From grover at sitepark.com Sun Aug 2 20:42:46 2015 From: grover at sitepark.com (Christoph =?ISO-8859-1?Q?Gr=F6ver?=) Date: Sun, 2 Aug 2015 22:42:46 +0200 Subject: Dovecot/Pigeonhole Issue (UTC modtime instead of localtime) In-Reply-To: <55BB8CDB.6010809@rename-it.nl> References: <20150731160115.2df2f412@aeshna> <55BB8CDB.6010809@rename-it.nl> Message-ID: <20150802224246.38aec854@anax.odonata.de> Hallo Stephan, > > Hmm. Probably, the timezone configuration (i.e. the contents of TZ > timezone environment variable) somehow doesn't reach the final stages of > e-mail delivery. I investigated further. I just did a telnet into the server and discoverered that dovecot knows the localtime. If I ask for the INTERNALDATE of a message the 2 hour offset of CEST is actually in the response: a fetch 166 INTERNALDATE * 166 FETCH (INTERNALDATE "02-Aug-2015 20:18:21 +0200") a OK Fetch completed (0.000 secs). See? The message above actually arrived at the server at 22:18 localtime. The Date-Header is set accordingly. The file modtime is being set to 20:18 of that day (for whatever reason). So the server responds with that time and adds the correct TZ offset (2 hours). So my assumption is that dovecot knows that we are at CEST (+0200). It's a maildir storage and the mails are received by netqmail-1.06. The last dovecot which we were using was version 2.2.14. With this version and system init and Mandriva 2010.2 we didn't have this kind of problem. I'd really like to understand this because some apple clients are having huge problems with the hours old INTERNALDATE. Thank you for your time. Greetings -- Christoph Gr?ver From phoffman at proper.com Mon Aug 3 01:07:16 2015 From: phoffman at proper.com (Paul Hoffman) Date: Sun, 02 Aug 2015 18:07:16 -0700 Subject: How to import an mbox to an existing user as a new mailbox Message-ID: Greetings. I have a bunch of mbox files that I want to import to an existing user, each into a new mailbox. That is, I want to import the file project1.mbox to user Foo, into a newly-created mailbox called "Project 1". Is there a dovecot tool to do this on the server? --Paul Hoffman From skdovecot at smail.inf.fh-brs.de Mon Aug 3 06:27:03 2015 From: skdovecot at smail.inf.fh-brs.de (Steffen Kaiser) Date: Mon, 3 Aug 2015 08:27:03 +0200 (CEST) Subject: How to import an mbox to an existing user as a new mailbox In-Reply-To: References: Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Sun, 2 Aug 2015, Paul Hoffman wrote: > Greetings. I have a bunch of mbox files that I want to import to an existing > user, each into a new mailbox. That is, I want to import the file > project1.mbox to user Foo, into a newly-created mailbox called "Project 1". > Is there a dovecot tool to do this on the server? do you have access to the file system of the mail storage? Can you call doveadm on the server? What mail storage do you use now, Maildir, mbox, ... ? - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVb8Jt3z1H7kL/d9rAQKCrAf9Em6faGXb0KFHcYp/kuf449P/Dxbinw4x 8aZKalHoarJzmPszQGWWqDxadaaKD0m61aCMWZ2ZM+FQrzPBA5nD/4ipRVtY9UZm hPXhnZ6ZDuMPosz9nNJlvqno6o17d4R8Gjf4EBjV1hva0cQr48FQv4d9K8bHCT/c 11WspPnT4EA0zYL3omygkA77d/8EUhxxp2Ju2eQ7ZVPH2UJUj2Ojfn4+NovqmqZs YYvIjzgDC+1B3+kvCgOm6o737XL1urozp3Rj1ccJ5zaB6dMkhUpbR172EHhZ4tM1 th1D0FsI1M7NawIsGsFmcl4m/ecH0oMv/voCU0IRLtWMjFXqWqCuew== =OutZ -----END PGP SIGNATURE----- From miloslav.hula at gmail.com Mon Aug 3 12:31:39 2015 From: miloslav.hula at gmail.com (=?UTF-8?Q?Miloslav_H=c5=afla?=) Date: Mon, 3 Aug 2015 14:31:39 +0200 Subject: Structure of dovecot.index.pvt.log Message-ID: <55BF5F2B.7080402@gmail.com> Hi, we are migrating from Cyrus to Dovecot and I would like to migrate seen flags for shared folder too. We have Dovecot 2.2.13 prepared as: location = maildir:/vmail/user/%%n/Maildir:INDEXPVT=/vmail/user/%n/Maildir/Shared/%%n Now I'm looking for 'dovecot.index.pvt.log' syntax to be able migrate Seen flags. All I know is, that index contains messages UID and Seen flag. May I ask you for a link to doc (if it exists) or into a source code? Should I care about 'dovecot.index.pvt.log' timestamps? Thank you, Milo From andy at thecsillags.com Mon Aug 3 13:48:46 2015 From: andy at thecsillags.com (Andy Csillag) Date: Mon, 03 Aug 2015 13:48:46 +0000 Subject: Renaming shared mailbox makes it disappear from sharee Message-ID: <07ae69872532d9190fbc498ea6372f17@www.thecsillags.com> I've run into an interesting thing.? If a mailbox has been shared to a second user (I'll call it the sharee), and there are subfolders on that mailbox, and the sharee renames one of those subfolders, it disappears from view to the sharee.? The only way I've been able discover to fix it is to have the sharee create a new subfolder -- then the renamed folder shows up again; in addition to the newly created folder. For example, first at example.org shares it's INBOX to second at example.org.? There's a folder INBOX/thing, which second@ renames to INBOX/renamed (ACLs permitting).? second@ can no longer can see INBOX/renamed (or INBOX/thing for that matter), via IMAP's LIST command.? But if second@ creates a new folder INBOX/other, INBOX/renamed is now visible. This is very reproducible for me.? I also tried it without the imappostlogin bit (in the dovecot -n).? Anyone else run into this? No NFS is involved, this is a single machine setup. Output of dovecot -n # 2.2.18: /dovecot/etc/dovecot/dovecot.conf # Pigeonhole version 0.4.8 # OS: Linux 2.6.32-504.el6.x86_64 x86_64 CentOS release 6.6 (Final) xfs auth_master_user_separator = * auth_mechanisms = plain login dict { ? acl = mysql:/dovecot/etc/dovecot/dovecot-dict-sql.conf.ext } disable_plaintext_auth = no lda_mailbox_autosubscribe = yes mail_home = mdbox:/dovecot/var/homedirs/%2Mu/%2.2Mu/%u mail_location = mdbox:/dovecot/var/mailboxes/%2Mu/%2.2Mu/%u mail_plugins = acl mail_shared_explicit_inbox = yes managesieve_notify_capability = mailto managesieve_sieve_capability = fileinto reject envelope encoded-character vacation subaddress comparator-i;ascii-numeric relational regex imap4flags copy include variables body enotify environment mailbox date index ihave duplicate vnd.dovecot.pipe vnd.dovecot.filter vnd.dovecot.execute namespace { ? list = children ? location = mdbox:/dovecot/var/mailboxes/%%2Mu/%%2.2Mu/%%u ? prefix = shared/%%u/ ? separator = / ? subscriptions = no ? type = shared } namespace inbox { ? inbox = yes ? location = ? mailbox Drafts { ??? special_use = Drafts ? } ? mailbox Junk { ??? special_use = Junk ? } ? mailbox Sent { ??? special_use = Sent ? } ? mailbox "Sent Messages" { ??? special_use = Sent ? } ? mailbox Trash { ??? special_use = Trash ? } ? prefix = } passdb { ? args = /dovecot/etc/dovecot/mastership-sql.conf ? driver = sql ? master = yes ? pass = yes } passdb { ? args = /dovecot/etc/dovecot/dovecot-sql.conf.ext ? driver = sql } plugin { ? acl = vfile ? acl_shared_dict = proxy::acl ? sieve_after = /dovecot/etc/dovecot/sieve/99-autofile.sieve ? sieve_after2 = /dovecot/etc/dovecot/sieve/98-notify-biz-address.sieve ? sieve_execute_bin_dir = /dovecot/etc/dovecot/sieve-execute ? sieve_execute_socket_dir = sieve-execute ? sieve_extensions = +vnd.dovecot.execute +vnd.dovecot.filter +vnd.dovecot.pipe ? sieve_filter_bin_dir = /dovecot/etc/dovecot/sieve-filter ? sieve_filter_socket_dir = sieve-filter ? sieve_pipe_bin_dir = /dovecot/etc/dovecot/sieve-pipe ? sieve_pipe_socket_dir = sieve-pipe ? sieve_plugins = sieve_extprograms } postmaster_address = postmaster at xxxxdomain.com sendmail_path = /dovecot/sbin/sendmail service dict { ? unix_listener dict { ??? user = dovecot ? } } service imap-postlogin { ? executable = script-login /dovecot/etc/dovecot/imappostlogin ? user = $default_internal_user } service imap { ? executable = imap imap-postlogin } ssl = no userdb { ? args = uid=dovecot gid=dovecot home=/dovecot/var/mailboxes/%%2Mu/%%2.2Mu/%%u ? driver = static } protocol lmtp { ? mail_plugins = acl sieve } protocol lda { ? mail_plugins = acl sieve } protocol imap { ? mail_plugins = acl imap_acl } ? From phoffman at proper.com Mon Aug 3 13:56:11 2015 From: phoffman at proper.com (Paul Hoffman) Date: Mon, 03 Aug 2015 06:56:11 -0700 Subject: How to import an mbox to an existing user as a new mailbox In-Reply-To: References: Message-ID: <97F9738A-901F-4E39-966B-6B092C2B5BA2@proper.com> On 2 Aug 2015, at 23:27, Steffen Kaiser wrote: > On Sun, 2 Aug 2015, Paul Hoffman wrote: > >> Greetings. I have a bunch of mbox files that I want to import to an >> existing user, each into a new mailbox. That is, I want to import the >> file project1.mbox to user Foo, into a newly-created mailbox called >> "Project 1". Is there a dovecot tool to do this on the server? > > do you have access to the file system of the mail storage? Can you > call doveadm on the server? What mail storage do you use now, Maildir, > mbox, ... ? Sorry for being unclear earlier. Yes, yes, mbox. So, I take it this is a doveadm command. :-) Will this be just "import", or a combination of "mailbox" and "import"? --Paul Hoffman From skdovecot at smail.inf.fh-brs.de Mon Aug 3 14:16:34 2015 From: skdovecot at smail.inf.fh-brs.de (Steffen Kaiser) Date: Mon, 3 Aug 2015 16:16:34 +0200 (CEST) Subject: How to import an mbox to an existing user as a new mailbox In-Reply-To: <97F9738A-901F-4E39-966B-6B092C2B5BA2@proper.com> References: <97F9738A-901F-4E39-966B-6B092C2B5BA2@proper.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 3 Aug 2015, Paul Hoffman wrote: > On 2 Aug 2015, at 23:27, Steffen Kaiser wrote: >> On Sun, 2 Aug 2015, Paul Hoffman wrote: >> >>> Greetings. I have a bunch of mbox files that I want to import to an >>> existing user, each into a new mailbox. That is, I want to import the file >>> project1.mbox to user Foo, into a newly-created mailbox called "Project >>> 1". Is there a dovecot tool to do this on the server? >> >> do you have access to the file system of the mail storage? Can you call >> doveadm on the server? What mail storage do you use now, Maildir, mbox, ... >> ? > > Sorry for being unclear earlier. Yes, yes, mbox. copy project1.mbox to '/path/to/user/mail/storage/Project 1.mbox' and make sure, the user has file system rights to access the file at all. Maybe, you need to add the new mailbox to the user's subscription file, in order to have it show up. - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVb93wnz1H7kL/d9rAQIsaQf8DMKWBQkIDcpj5T7+c4Fo4hK8CcSllEq3 1bjERFLVDXxMnMzInhwtUjlcoCnjgSI0D7bC4xe6xaO/viXCSFi94H56nGMrApih 0wkD0e3Dd3P44T+C/X1hZl6eAj6PPKETvWt8z00amUrtyhXhpGWggVeqQYob2YyM qprG3SuYqdsxfDGX5wOUiq0xtkLfQUQl5m60mwOiSnsv7VOCPsUl6y7XSgfb3wKv CBD23kL6cnsbLmrtqNXxGxD9RAMrTqpKEBRYw6Tj8K1xRJxbC7ysblDXK6S9ZqCi gjIs4m7pll8Ap8I45xXBD6gEZFfUy7qfhSWzkom073KivesGgh5MSw== =BbrX -----END PGP SIGNATURE----- From nutschi at talach.at Mon Aug 3 17:54:15 2015 From: nutschi at talach.at (Nutsch) Date: Mon, 03 Aug 2015 19:54:15 +0200 Subject: 451 4.3.0 Temporary internal failure Message-ID: <55BFAAC7.5080404@talach.at> Hi i can send mails without problems in any direction, except when the attachments are bigger than 1 MB. I alway get this message relay=mail.example.net[private/dovecot-lmtp], delay=35155, delays=35155/0.03/0.02/0.09, dsn=4.3.0, status=deferred (host mail.example.net[private/dovecot-lmtp] said: 451 4.3.0 Temporary internal failure (in reply to end of DATA command)) Aug 3 19:34:34 46185 dovecot: lmtp(6477): Disconnect from local: Temporary internal failure (in DATA) postconf message_size_limit message_size_limit = 0 postconf mail_size_limit mailbox_size_limit = 0 i can send big attachments to an extern address but not from intern to intern. Someone said that maybe the tmp directory is the problem so i checked the size. ls -ald /tmp/; df -h /tmp/ drwxrwxrwt 4 root root 80 Aug 3 19:44 /tmp/ Filesystem Size Used Avail Use% Mounted on - 1,0M 4,0K 1020K 1% /tmp root at example:/tmp# is there a connection between lmtp and the size of the tmp directory? And why is the size 1,0M, tmp is not an own partion. Should i increase the size? And if yes, how? df -h Filesystem Size Used Avail Use% Mounted on /dev/vzfs 80G 12G 69G 15% / ??? br. noctua From bind at enas.net Mon Aug 3 18:20:29 2015 From: bind at enas.net (Urban Loesch) Date: Mon, 03 Aug 2015 20:20:29 +0200 Subject: 451 4.3.0 Temporary internal failure In-Reply-To: <55BFAAC7.5080404@talach.at> References: <55BFAAC7.5080404@talach.at> Message-ID: <55BFB0ED.3010301@enas.net> Hi, according to dovecot.conf lmtp stores all mails temporarily in "/tmp/" that are bigger than 256KB. You can change the directory in dovecot.conf or you should increase the /tmp/ size. It could be that /tmp/ is a ramdisk. Check /etc/fstab. Which os you are using? What does "df -hT" say? What does /etc/fstab say? Regards Urban Am 03.08.2015 um 19:54 schrieb Nutsch: > Hi > > i can send mails without problems in any direction, except when the > attachments are bigger than 1 MB. I alway get this message > > relay=mail.example.net[private/dovecot-lmtp], delay=35155, > delays=35155/0.03/0.02/0.09, dsn=4.3.0, status=deferred (host > mail.example.net[private/dovecot-lmtp] said: 451 4.3.0 Temporary > internal failure (in reply to end of DATA command)) Aug 3 19:34:34 > 46185 dovecot: lmtp(6477): Disconnect from local: Temporary internal > failure (in DATA) > > > postconf message_size_limit > message_size_limit = 0 > postconf mail_size_limit > mailbox_size_limit = 0 > > i can send big attachments to an extern address but not from intern to > intern. Someone said that maybe the tmp directory is the problem so i > checked the size. > > ls -ald /tmp/; df -h /tmp/ > drwxrwxrwt 4 root root 80 Aug 3 19:44 /tmp/ > Filesystem Size Used Avail Use% Mounted on > - 1,0M 4,0K 1020K 1% /tmp > root at example:/tmp# > > is there a connection between lmtp and the size of the tmp directory? > And why is the size 1,0M, tmp is not an own partion. Should i increase > the size? And if yes, how? > > df -h > Filesystem Size Used Avail Use% Mounted on > /dev/vzfs 80G 12G 69G 15% / > > ??? > > br. noctua From matthias.lay at securepoint.de Tue Aug 4 09:27:48 2015 From: matthias.lay at securepoint.de (matthias lay) Date: Tue, 4 Aug 2015 11:27:48 +0200 Subject: ldap attribute modifiers. how to lowercase non ASCII In-Reply-To: <55B79C7D.8030504@securepoint.de> References: <55B79C7D.8030504@securepoint.de> Message-ID: <55C08594.9040706@securepoint.de> update of investigation... I encountered a different behavior in %Lu and %L{ldap:uid} when doing the search with lower cased user. # doveadm user bl?tula field value uid vmail gid vmail home /var/data/vmail/private/bl?tula mail maildir:~/Maildir:LAYOUT=fs:INBOX=~/Maildir/INBOX spuser private/bl?tula so in this case. the name gets lowered in the %Lu but not the %L{ldap:uid} case. - The user in ldap is all upper case. - system locale is en_US.UTF-8 log of the above doveadm query: mail.debug: Jul 29 12:45:28 dovecot: auth: Debug: master in: USER 1 bl?tula service=doveadm mail.debug: Jul 29 12:45:28 dovecot: auth: Debug: ldap(bl?tula): userdb cache miss mail.debug: Jul 29 12:45:28 dovecot: auth: Debug: ldap(bl?tula): user search: base=ou=users, dc=uma,dc=local scope=subtree filter=(&(objectClass=inetOrgPerson)(uid=bl?tula)) fields=uid,uid mail.debug: Jul 29 12:45:28 dovecot: auth: Debug: ldap(bl?tula): result: uid=BL?TULA; uid unused mail.debug: Jul 29 12:45:28 dovecot: auth: Debug: ldap(bl?tula): result: uid=BL?TULA mail.debug: Jul 29 12:45:28 dovecot: auth: Debug: userdb out: USER 1 bl?tula home=/var/data/vmail/private/bl?tulaspuser=private/bl?tula On 07/28/2015 05:15 PM, matthias lay wrote: > hi all, > > I encountered some problem with lowercasing chars like an "?" > > with user_attrs like this: > > user_attrs = uid=home=/mail/%Lu, =myvar=%L{ldap:uid} > > the ? doesnt get lower cased in both cases > > > # doveadm user BL?TULA > field value > user bl?tula > uid vmail > gid vmail > home /mail/bl?tula > mail maildir:~/Maildir:LAYOUT=fs:INBOX=~/Maildir/INBOX > myvar bl?tula > > > somebody knows if there a fix for that? > > > Greetz Matze > From sebastian.wiedenroth at skylime.net Tue Aug 4 10:09:17 2015 From: sebastian.wiedenroth at skylime.net (Sebastian Wiedenroth) Date: Tue, 4 Aug 2015 12:09:17 +0200 Subject: [Patch] Fix hang in safe_sendfile on SmartOS In-Reply-To: References: Message-ID: ping? If more information or a system to test the issue on is required please let me know, I?m sure I can provide those. If this list is not the correct place for patches I?m also happy to learn where else to send it. Thanks. Best regards, Sebastian > Am 16.07.2015 um 17:03 schrieb Sebastian Wiedenroth : > > # HG changeset patch > # User Sebastian Wiedenroth > # Date 1437050484 -7200 > # Thu Jul 16 14:41:24 2015 +0200 > # Node ID 7ef3a533b097e8e6590e754dc56ad308ab29233b > # Parent e3640ccaa76d77a9658126d1f8f306480dad8af7 > Fix hang in safe_sendfile on SmartOS > > The call to sendfile on SmartOS can fail with EOPNOTSUPP. This is a valid error > code and documented in the man page. This error code needs to be handled or > else dovecot will retry the sendfile call endlessly and hang. > > diff -r e3640ccaa76d -r 7ef3a533b097 src/lib/sendfile-util.c > --- a/src/lib/sendfile-util.c Sat Jan 10 04:32:42 2015 +0200 > +++ b/src/lib/sendfile-util.c Thu Jul 16 14:41:24 2015 +0200 > @@ -116,7 +116,7 @@ > if (errno == EINVAL) { > /* most likely trying to read past EOF */ > ret = 0; > - } else if (errno == EAFNOSUPPORT) { > + } else if (errno == EAFNOSUPPORT || errno == EOPNOTSUPP) { > /* not supported, return Linux-like EINVAL so caller > sees only consistent errnos. */ > errno = EINVAL; > From nutschi at talach.at Tue Aug 4 10:36:00 2015 From: nutschi at talach.at (Nutsch) Date: Tue, 04 Aug 2015 12:36:00 +0200 Subject: 451 4.3.0 Temporary internal failure In-Reply-To: <55BFB0ED.3010301@enas.net> References: <55BFAAC7.5080404@talach.at> <55BFB0ED.3010301@enas.net> Message-ID: <55C09590.2030204@talach.at> Hi, OS: Debian GNU/Linux 7 df -hT Filesystem Type Size Used Avail Use% Mounted on /dev/vzfs reiserfs 30G 12G 19G 38% / /etc/fstab proc /proc proc defaults 0 0 none /dev/pts devpts rw,gid=5,mode=620 0 0 none /run/shm tmpfs defaults 0 0 If someone knows an option to change the tmp directory in dovecot.conf, it would be very helpful. I can't find it. I can't increase the size of tmp, its not an partition and even my provider don't knows how it can be possible that the tmp directory is limited to 1 MB. Regards Noctua On 2015-08-03 20:20, Urban Loesch wrote: > Hi, > > according to dovecot.conf lmtp stores all mails temporarily in "/tmp/" > that are bigger than 256KB. You can change the directory in > dovecot.conf or you should increase the /tmp/ size. It could be that > /tmp/ is a ramdisk. Check /etc/fstab. > > Which os you are using? > What does "df -hT" say? > What does /etc/fstab say? > > Regards > Urban > > Am 03.08.2015 um 19:54 schrieb Nutsch: >> Hi >> >> i can send mails without problems in any direction, except when the >> attachments are bigger than 1 MB. I alway get this message >> >> relay=mail.example.net[private/dovecot-lmtp], delay=35155, >> delays=35155/0.03/0.02/0.09, dsn=4.3.0, status=deferred (host >> mail.example.net[private/dovecot-lmtp] said: 451 4.3.0 Temporary >> internal failure (in reply to end of DATA command)) Aug 3 19:34:34 >> 46185 dovecot: lmtp(6477): Disconnect from local: Temporary internal >> failure (in DATA) >> >> >> postconf message_size_limit >> message_size_limit = 0 >> postconf mail_size_limit >> mailbox_size_limit = 0 >> >> i can send big attachments to an extern address but not from intern to >> intern. Someone said that maybe the tmp directory is the problem so i >> checked the size. >> >> ls -ald /tmp/; df -h /tmp/ >> drwxrwxrwt 4 root root 80 Aug 3 19:44 /tmp/ >> Filesystem Size Used Avail Use% Mounted on >> - 1,0M 4,0K 1020K 1% /tmp >> root at example:/tmp# >> >> is there a connection between lmtp and the size of the tmp directory? >> And why is the size 1,0M, tmp is not an own partion. Should i increase >> the size? And if yes, how? >> >> df -h >> Filesystem Size Used Avail Use% Mounted on >> /dev/vzfs 80G 12G 69G 15% / >> >> ??? >> >> br. noctua From bind at enas.net Tue Aug 4 12:03:47 2015 From: bind at enas.net (Urban Loesch) Date: Tue, 04 Aug 2015 14:03:47 +0200 Subject: 451 4.3.0 Temporary internal failure In-Reply-To: <55C09590.2030204@talach.at> References: <55BFAAC7.5080404@talach.at> <55BFB0ED.3010301@enas.net> <55C09590.2030204@talach.at> Message-ID: <55C0AA23.2060203@enas.net> Hi, that should be: # Directory in which LDA/LMTP temporarily stores incoming mails >128 kB. #mail_temp_dir = /tmp Best Urban Am 04.08.2015 um 12:36 schrieb Nutsch: > Hi, > > OS: Debian GNU/Linux 7 > > df -hT > Filesystem Type Size Used Avail Use% Mounted on > /dev/vzfs reiserfs 30G 12G 19G 38% / > > /etc/fstab > proc /proc proc defaults 0 0 > none /dev/pts devpts rw,gid=5,mode=620 0 0 > none /run/shm tmpfs defaults 0 0 > > > If someone knows an option to change the tmp directory in dovecot.conf, it would be very helpful. I can't find it. > I can't increase the size of tmp, its not an partition and even my provider don't knows how it can be possible that the tmp directory is limited to 1 MB. > > Regards > Noctua > > On 2015-08-03 20:20, Urban Loesch wrote: >> Hi, >> >> according to dovecot.conf lmtp stores all mails temporarily in "/tmp/" that are bigger than 256KB. You can change the directory in dovecot.conf or >> you should increase the /tmp/ size. It could be that /tmp/ is a ramdisk. Check /etc/fstab. >> >> Which os you are using? >> What does "df -hT" say? >> What does /etc/fstab say? >> >> Regards >> Urban >> >> Am 03.08.2015 um 19:54 schrieb Nutsch: >>> Hi >>> >>> i can send mails without problems in any direction, except when the >>> attachments are bigger than 1 MB. I alway get this message >>> >>> relay=mail.example.net[private/dovecot-lmtp], delay=35155, >>> delays=35155/0.03/0.02/0.09, dsn=4.3.0, status=deferred (host >>> mail.example.net[private/dovecot-lmtp] said: 451 4.3.0 Temporary >>> internal failure (in reply to end of DATA command)) Aug 3 19:34:34 >>> 46185 dovecot: lmtp(6477): Disconnect from local: Temporary internal >>> failure (in DATA) >>> >>> >>> postconf message_size_limit >>> message_size_limit = 0 >>> postconf mail_size_limit >>> mailbox_size_limit = 0 >>> >>> i can send big attachments to an extern address but not from intern to >>> intern. Someone said that maybe the tmp directory is the problem so i >>> checked the size. >>> >>> ls -ald /tmp/; df -h /tmp/ >>> drwxrwxrwt 4 root root 80 Aug 3 19:44 /tmp/ >>> Filesystem Size Used Avail Use% Mounted on >>> - 1,0M 4,0K 1020K 1% /tmp >>> root at example:/tmp# >>> >>> is there a connection between lmtp and the size of the tmp directory? >>> And why is the size 1,0M, tmp is not an own partion. Should i increase >>> the size? And if yes, how? >>> >>> df -h >>> Filesystem Size Used Avail Use% Mounted on >>> /dev/vzfs 80G 12G 69G 15% / >>> >>> ??? >>> >>> br. noctua > From nutschi at talach.at Tue Aug 4 12:11:16 2015 From: nutschi at talach.at (Nutsch) Date: Tue, 04 Aug 2015 14:11:16 +0200 Subject: 451 4.3.0 Temporary internal failure In-Reply-To: <55C0AA23.2060203@enas.net> References: <55BFAAC7.5080404@talach.at> <55BFB0ED.3010301@enas.net> <55C09590.2030204@talach.at> <55C0AA23.2060203@enas.net> Message-ID: <55C0ABE4.9060701@talach.at> thx On 2015-08-04 14:03, Urban Loesch wrote: > Hi, > > that should be: > > # Directory in which LDA/LMTP temporarily stores incoming mails >128 kB. > #mail_temp_dir = /tmp > > Best > Urban > > Am 04.08.2015 um 12:36 schrieb Nutsch: >> Hi, >> >> OS: Debian GNU/Linux 7 >> >> df -hT >> Filesystem Type Size Used Avail Use% Mounted on >> /dev/vzfs reiserfs 30G 12G 19G 38% / >> >> /etc/fstab >> proc /proc proc defaults 0 0 >> none /dev/pts devpts rw,gid=5,mode=620 0 0 >> none /run/shm tmpfs defaults 0 0 >> >> >> If someone knows an option to change the tmp directory in dovecot.conf, it would be very helpful. I can't find it. >> I can't increase the size of tmp, its not an partition and even my provider don't knows how it can be possible that the tmp directory is limited to 1 MB. >> >> Regards >> Noctua >> >> On 2015-08-03 20:20, Urban Loesch wrote: >>> Hi, >>> >>> according to dovecot.conf lmtp stores all mails temporarily in "/tmp/" that are bigger than 256KB. You can change the directory in dovecot.conf or >>> you should increase the /tmp/ size. It could be that /tmp/ is a ramdisk. Check /etc/fstab. >>> >>> Which os you are using? >>> What does "df -hT" say? >>> What does /etc/fstab say? >>> >>> Regards >>> Urban >>> >>> Am 03.08.2015 um 19:54 schrieb Nutsch: >>>> Hi >>>> >>>> i can send mails without problems in any direction, except when the >>>> attachments are bigger than 1 MB. I alway get this message >>>> >>>> relay=mail.example.net[private/dovecot-lmtp], delay=35155, >>>> delays=35155/0.03/0.02/0.09, dsn=4.3.0, status=deferred (host >>>> mail.example.net[private/dovecot-lmtp] said: 451 4.3.0 Temporary >>>> internal failure (in reply to end of DATA command)) Aug 3 19:34:34 >>>> 46185 dovecot: lmtp(6477): Disconnect from local: Temporary internal >>>> failure (in DATA) >>>> >>>> >>>> postconf message_size_limit >>>> message_size_limit = 0 >>>> postconf mail_size_limit >>>> mailbox_size_limit = 0 >>>> >>>> i can send big attachments to an extern address but not from intern to >>>> intern. Someone said that maybe the tmp directory is the problem so i >>>> checked the size. >>>> >>>> ls -ald /tmp/; df -h /tmp/ >>>> drwxrwxrwt 4 root root 80 Aug 3 19:44 /tmp/ >>>> Filesystem Size Used Avail Use% Mounted on >>>> - 1,0M 4,0K 1020K 1% /tmp >>>> root at example:/tmp# >>>> >>>> is there a connection between lmtp and the size of the tmp directory? >>>> And why is the size 1,0M, tmp is not an own partion. Should i increase >>>> the size? And if yes, how? >>>> >>>> df -h >>>> Filesystem Size Used Avail Use% Mounted on >>>> /dev/vzfs 80G 12G 69G 15% / >>>> >>>> ??? >>>> >>>> br. noctua From grover at sitepark.com Tue Aug 4 15:03:17 2015 From: grover at sitepark.com (Christoph =?ISO-8859-1?Q?Gr=F6ver?=) Date: Tue, 4 Aug 2015 17:03:17 +0200 Subject: Dovecot/Pigeonhole Issue (UTC modtime instead of localtime) In-Reply-To: <55BB8CDB.6010809@rename-it.nl> References: <20150731160115.2df2f412@aeshna> <55BB8CDB.6010809@rename-it.nl> Message-ID: <20150804170317.1dbf7a36@aeshna> Hello ML, Hello Stephan, > Hmm. Probably, the timezone configuration (i.e. the contents of TZ > timezone environment variable) somehow doesn't reach the final stages > of e-mail delivery. Well. I tried several ways of telling the lda or whatever is setting up the INTERNALDATE to use the CEST +0200. Although it is already setup on the server via a symlink from /etc/localtime to /usr/share/zoneinfo/Europe/Berlin, I defined it at several stages, for example globally and for the account of the virtual user of the mails. But all of it to no avail. It just keeps getting a mtime in the past. So I actually had to patch the sources. I commented out the utime calls of src/lib-storage/index/maildir/maildir-save.c . This way the file gets the correct mtime. Still the clients saw the wrong INTERNALDATE. I also had to completely disable writing a index.cache by dovecot. Both are not ideal in my opinion. I'd still prefer a real solution. But for the moment I had to do it because my users became "desperate". If you have any ideas left, please tell me. I'd still like to here them. Thank you. Greetings -- Sitepark Gesellschaft f?r Informationsmanagement mbH Rothenburg 14-16, 48143 M?nster Telefon: +49 251 482655-0, Telefax: +49 251 482655-55 http://www.sitepark.com http://www.facebook.com/sitepark Gesch?ftsf?hrer: Thorsten Liebold Amtsgericht M?nster, HRB 5017 From lists at wiesinger.com Tue Aug 4 18:19:41 2015 From: lists at wiesinger.com (Gerhard Wiesinger) Date: Tue, 4 Aug 2015 20:19:41 +0200 Subject: Dovecot sieve pigeonhole permission Message-ID: <55C1023D.2040907@wiesinger.com> Hello, I'm running dovecot in a classical vmail.vmail setup with pigeonhole and LMTP. Permission worked well in the initial setup but currently (maybe after Fedora 22 update) I'm having the following permission issue: lmtp(root): Error: 7fPZFOL9wFXePQAABcdabc: sieve: binary open: failed to open: open(/etc/dovecot/sieve_after.svbin) failed: Permission denied (euid=9999(vmail) egid=9999(vmail) missing +r perm: /etc/dovecot/sieve_after.svbin, we're not in group 0(root), dir owned by 0:0 mode=0755) lmtp(root): Error: 7fPZFOL9wFXePQAABcdabc: sieve: binary save: failed to create temporary file: open(/etc/dovecot/sieve_after.svbin.myserver.mydomain.15838.) failed: Permission denied (euid=9999(vmail) egid=9999(vmail) missing +w perm: /etc/dovecot, dir owned by 0:0 mode=0755) lmtp(root): Error: 7fPZFOL9wFXePQAABcdcbc: sieve: The LDA Sieve plugin does not have permission to save global Sieve script binaries; global Sieve scripts like `/etc/dovecot/sieve_after.sieve' need to be pre-compiled using the sievec tool rpm -V dovecot dovecot-pigeonhole doesn't report any permission issues Versions: dovecot-2.2.18-2.fc22.x86_64 dovecot-pigeonhole-2.2.18-2.fc22.x86_64 Relevant config parts: mail_gid = vmail mail_uid = vmail plugin { sieve = ~/.dovecot.sieve sieve_after = /etc/dovecot/sieve_after.sieve sieve_dir = ~/sieve } ls -lad /etc/dovecot/ drwxr-xr-x. 3 root root 4096 Jul 30 18:13 /etc/dovecot/ ls -la /etc/dovecot/*sieve* -rwxr-x--- 1 vmail vmail 288 Aug 8 2014 /etc/dovecot/sieve_after.sieve -rw-r----- 1 root root 355 Jul 30 18:13 /etc/dovecot/sieve_after.svbin What's are the recommended permissions for the files/directories (also for root service startup of dovecot)? Did something change in permissions management? Thanx. Ciao, Gerhard -- http://www.wiesinger.com/ From skdovecot at smail.inf.fh-brs.de Wed Aug 5 11:06:08 2015 From: skdovecot at smail.inf.fh-brs.de (Steffen Kaiser) Date: Wed, 5 Aug 2015 13:06:08 +0200 (CEST) Subject: Dovecot sieve pigeonhole permission In-Reply-To: <55C1023D.2040907@wiesinger.com> References: <55C1023D.2040907@wiesinger.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tue, 4 Aug 2015, Gerhard Wiesinger wrote: > > I'm running dovecot in a classical vmail.vmail setup with pigeonhole and > LMTP. Permission worked well in the initial setup but currently (maybe after > Fedora 22 update) I'm having the following permission issue: > > lmtp(root): Error: 7fPZFOL9wFXePQAABcdabc: sieve: binary open: failed to > open: open(/etc/dovecot/sieve_after.svbin) failed: Permission denied > (euid=9999(vmail) egid=9999(vmail) missing +r perm: > /etc/dovecot/sieve_after.svbin, we're not in group 0(root), dir owned by 0:0 > mode=0755) Did you honored the error log and added the read permission for the vmail user, which most likely means: chmod a+r /etc/dovecot/sieve_after.svbin > lmtp(root): Error: 7fPZFOL9wFXePQAABcdabc: sieve: binary save: failed to > create temporary file: > open(/etc/dovecot/sieve_after.svbin.myserver.mydomain.15838.) failed: > Permission denied (euid=9999(vmail) egid=9999(vmail) missing +w perm: > /etc/dovecot, dir owned by 0:0 mode=0755) > lmtp(root): Error: 7fPZFOL9wFXePQAABcdcbc: sieve: The LDA Sieve plugin does > not have permission to save global Sieve script binaries; global Sieve > scripts like `/etc/dovecot/sieve_after.sieve' need to be pre-compiled using > the sievec tool Did you compiled the script manually as described in the error log? > > rpm -V dovecot dovecot-pigeonhole > doesn't report any permission issues > > Versions: > dovecot-2.2.18-2.fc22.x86_64 > dovecot-pigeonhole-2.2.18-2.fc22.x86_64 > > Relevant config parts: > mail_gid = vmail > mail_uid = vmail > > plugin { > sieve = ~/.dovecot.sieve > sieve_after = /etc/dovecot/sieve_after.sieve > sieve_dir = ~/sieve > } > > ls -lad /etc/dovecot/ > drwxr-xr-x. 3 root root 4096 Jul 30 18:13 /etc/dovecot/ > ls -la /etc/dovecot/*sieve* > -rwxr-x--- 1 vmail vmail 288 Aug 8 2014 /etc/dovecot/sieve_after.sieve > -rw-r----- 1 root root 355 Jul 30 18:13 /etc/dovecot/sieve_after.svbin > > What's are the recommended permissions for the files/directories (also for > root service startup of dovecot)? > Did something change in permissions management? - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVcHuIHz1H7kL/d9rAQK99wgAnp6aQJEioc/KbIWilj/JiNUJnWkMT090 DNvZBKXmGpD8IyNkAxvliVQyP3o8vbbBwaBMSoGXwlu7es9I1fKJI641pMgRNO/w r1iYkjFtP/sq4GvHoPVkTrs6QzKxVXQJZGfqsLvqAG58kieUM94QSyor5/7xa/1q XhGTH9ifJURqIDuwZkgcBKZPKJupd6+fyU8t9S27AVISjrPc5KVcuAh5yjYt2BrE 8cQRKysh+1xdLBswn4B/8jDcR9F04rjE2Py1AdmQpVjyC5AbfCbu9a9y5sCPuoEp g8NTF+kRrO6Y7rXU8aZwgpa9ScDoDMijOovpi3B/5U2r/40qpC4b7w== =bD4P -----END PGP SIGNATURE----- From lists at wiesinger.com Wed Aug 5 12:04:56 2015 From: lists at wiesinger.com (Gerhard Wiesinger) Date: Wed, 5 Aug 2015 14:04:56 +0200 Subject: Dovecot sieve pigeonhole permission In-Reply-To: References: <55C1023D.2040907@wiesinger.com> Message-ID: <55C1FBE8.90200@wiesinger.com> On 05.08.2015 13:06, Steffen Kaiser wrote: > On Tue, 4 Aug 2015, Gerhard Wiesinger wrote: > >> >> I'm running dovecot in a classical vmail.vmail setup with pigeonhole >> and LMTP. Permission worked well in the initial setup but currently >> (maybe after Fedora 22 update) I'm having the following permission >> issue: >> >> lmtp(root): Error: 7fPZFOL9wFXePQAABcdabc: sieve: binary open: failed >> to open: open(/etc/dovecot/sieve_after.svbin) failed: Permission >> denied (euid=9999(vmail) egid=9999(vmail) missing +r perm: >> /etc/dovecot/sieve_after.svbin, we're not in group 0(root), dir owned >> by 0:0 mode=0755) > > Did you honored the error log and added the read permission for the > vmail user, which most likely means: > > chmod a+r /etc/dovecot/sieve_after.svbin > Yes, I read it but I didn't like to give read permissions for all. Isn't there a better designed solution available? Nevertheless it works by setting the read permissions for all ... >> lmtp(root): Error: 7fPZFOL9wFXePQAABcdabc: sieve: binary save: failed >> to create temporary file: >> open(/etc/dovecot/sieve_after.svbin.myserver.mydomain.15838.) failed: >> Permission denied (euid=9999(vmail) egid=9999(vmail) missing +w perm: >> /etc/dovecot, dir owned by 0:0 mode=0755) >> lmtp(root): Error: 7fPZFOL9wFXePQAABcdcbc: sieve: The LDA Sieve >> plugin does not have permission to save global Sieve script binaries; >> global Sieve scripts like `/etc/dovecot/sieve_after.sieve' need to be >> pre-compiled using the sievec tool > > Did you compiled the script manually as described in the error log? > Yes, I did, but didn't work because of the missing read permissions for all. Ciao, Gerhard -- http://www.wiesinger.com/ From skdovecot at smail.inf.fh-brs.de Wed Aug 5 12:16:07 2015 From: skdovecot at smail.inf.fh-brs.de (Steffen Kaiser) Date: Wed, 5 Aug 2015 14:16:07 +0200 (CEST) Subject: Dovecot sieve pigeonhole permission In-Reply-To: <55C1FBE8.90200@wiesinger.com> References: <55C1023D.2040907@wiesinger.com> <55C1FBE8.90200@wiesinger.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wed, 5 Aug 2015, Gerhard Wiesinger wrote: > On 05.08.2015 13:06, Steffen Kaiser wrote: >> On Tue, 4 Aug 2015, Gerhard Wiesinger wrote: >> >>> >>> I'm running dovecot in a classical vmail.vmail setup with pigeonhole and >>> LMTP. Permission worked well in the initial setup but currently (maybe >>> after Fedora 22 update) I'm having the following permission issue: >>> >>> lmtp(root): Error: 7fPZFOL9wFXePQAABcdabc: sieve: binary open: failed to >>> open: open(/etc/dovecot/sieve_after.svbin) failed: Permission denied >>> (euid=9999(vmail) egid=9999(vmail) missing +r perm: >>> /etc/dovecot/sieve_after.svbin, we're not in group 0(root), dir owned by >>> 0:0 mode=0755) >> >> Did you honored the error log and added the read permission for the vmail >> user, which most likely means: >> >> chmod a+r /etc/dovecot/sieve_after.svbin > > Yes, I read it but I didn't like to give read permissions for all. Isn't well, what kind of users have shell or file level access to the server? > there a better designed solution available? > Nevertheless it works by setting the read permissions for all ... > -rwxr-x--- 1 vmail vmail 288 Aug 8 2014 /etc/dovecot/sieve_after.sieve > -rw-r----- 1 root root 355 Jul 30 18:13 /etc/dovecot/sieve_after.svbin 1st: sieve scripts are not executable. 2nd: it needs read perm for all your LMTP users, that probable means: chown vmail /etc/dovecot/sieve_after.svbin chmod o= /etc/dovecot/sieve_after.svbin If you have some system users as well, you might need to chgrp as well. - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVcH+h3z1H7kL/d9rAQJ7FwgAmY9KMnl6D/aUu2zabT2tptauOItjN/Rs hW9/Rkk15sRgsaqbVce+Qb/YMb80sH8VjdYQbtC9ea/PrwSoF/lDnp1W/ekJoc92 W8cQn1b6e+aoH2chUdw/sYGWuKEv8qpyzFmrxS8vGitcMzbICQcXh5WYxe0Ie19s GqpPKlPRKI/X84slV3BDV6NWILovrDY/UIa2wt4vW2rlUVDpwN1Z6Wfk3ExxoVo/ http0vL6XgXdrDeQ8SyeT+atxdOyQ5bhr/u485Ix0vsWR7AXJG07x6Ul10d8th/i nrTH6EfMCrLKdupD80lYhWUYvGZFmSm3dG+fbWKIgnd1YOr8Nth33Q== =7oLN -----END PGP SIGNATURE----- From matthias.lay at securepoint.de Wed Aug 5 14:30:58 2015 From: matthias.lay at securepoint.de (matthias lay) Date: Wed, 5 Aug 2015 16:30:58 +0200 Subject: question on autch cache parameters Message-ID: <55C21E22.10808@securepoint.de> Hi list, I have a question on auth caching in 2.2.18. I am using acl_groups for a master user, appended in a static userdb file # snip ############################### master at uma:{SHA}XXXX=::::::userdb_acl_groups=umareadmaster allow_nets=127.0.0.1 # snap ############################### and use this group in a global ACL file. I discovered this only works on first NOT-cached login environment in imap-postlogin script on first login: AUTH_TOKEN=e96b5a32ceb2cafc4460c210ad2e92e3d7ab388c MASTER_USER=master at uma SPUSER=private/pdf LOCAL_IP=127.0.0.1 USER=pdf AUTH_USER=master at uma PWD=/var/run/dovecot USERDB_KEYS=ACL_GROUPS HOME SPUSER MASTER_USER AUTH_TOKEN AUTH_USER SHLVL=1 HOME=/var/data/vmail/private/pdf ACL_GROUPS=umareadmaster IP=127.0.0.1 _=/usr/bin/env on the second cached login it looks like this AUTH_TOKEN=12703b11932f233520f6d4b33559c33aeb1cfc7f MASTER_USER=master at uma SPUSER=private/pdf LOCAL_IP=127.0.0.1 USER=pdf AUTH_USER=master at uma PWD=/var/run/dovecot USERDB_KEYS=HOME SPUSER MASTER_USER AUTH_TOKEN AUTH_USER SHLVL=1 HOME=/var/data/vmail/private/pdf IP=127.0.0.1 _=/usr/bin/env so the ACL_GROUPS is gone. is this intended to be like that. so groups not included in cache and I have to find another approach? anybody else encountered similar problems with some auth Variables and caching? Greetz Matze From dale.gallagher at gmail.com Wed Aug 5 14:59:38 2015 From: dale.gallagher at gmail.com (Dale Gallagher) Date: Wed, 5 Aug 2015 16:59:38 +0200 Subject: 2.2.18: Mailbox INBOX sync: mailbox_delete failed: INBOX can't be deleted. Message-ID: Hello I have 2.2.18 built with: ./configure \ --prefix="$prefix" \ --bindir="$prefix/bin" \ --sbindir="$prefix/bin" \ --sysconfdir="/etc" \ --with-storages="maildir,imapc,pop3c" \ --without-vpopmail \ --without-ldap I'm attempting to sync (backup) from another Dovecot server running version 2.0.21: Both servers don't have no explicit namespaces configured and both are using the Maildir storage plugin. doveadm -Dv -o imapc_user=remote_user -o imapc_host=source.host -o imapc_port=993 -o imapc_ssl=imaps -o imapc_password=remote_pass -o imapc_features="rfc822.size fetch-headers" backup -R -u local_user imapc: All seems to go well, showing local and remote mailbox syncing without errors, then finally: dsync(mcg): Debug: brain M: Deleting mailbox 'INBOX' (GUID 59023c2a911dc255842b0000d09efc50): UIDVALIDITY changed (1355416734 -> 1438784913) dsync(mcg): Error: Mailbox INBOX sync: mailbox_delete failed: INBOX can't be deleted. No mail is synced to the local server (2.2.18) after this error halts execution. What should I be looking for, in terms of possible causes of this error? I've tried syncing on a completely fresh mailbox, which hasn't been accessed by a mail client and the same error occurs. I've seen similar reports of this error before, without clear solutions / suggestions. Help would be greatly appreciated. Thanks From matthias.lay at securepoint.de Wed Aug 5 15:33:24 2015 From: matthias.lay at securepoint.de (matthias lay) Date: Wed, 5 Aug 2015 17:33:24 +0200 Subject: question on autch cache parameters In-Reply-To: <55C21E22.10808@securepoint.de> References: <55C21E22.10808@securepoint.de> Message-ID: <55C22CC4.5030006@securepoint.de> just tested against dovecot 2.2.15 everythings works fine. so might be a bug introduced between 2.2.16 and 2.2.18 On 08/05/2015 04:30 PM, matthias lay wrote: > Hi list, > > I have a question on auth caching in 2.2.18. > > I am using acl_groups for a master user, appended in a static userdb file > > # snip ############################### > master at uma:{SHA}XXXX=::::::userdb_acl_groups=umareadmaster > allow_nets=127.0.0.1 > # snap ############################### > > and use this group in a global ACL file. > I discovered this only works on first NOT-cached login > > > > environment in imap-postlogin script on first login: > > > AUTH_TOKEN=e96b5a32ceb2cafc4460c210ad2e92e3d7ab388c > MASTER_USER=master at uma > SPUSER=private/pdf > LOCAL_IP=127.0.0.1 > USER=pdf > AUTH_USER=master at uma > PWD=/var/run/dovecot > USERDB_KEYS=ACL_GROUPS HOME SPUSER MASTER_USER AUTH_TOKEN AUTH_USER > SHLVL=1 > HOME=/var/data/vmail/private/pdf > ACL_GROUPS=umareadmaster > IP=127.0.0.1 > _=/usr/bin/env > > > on the second cached login it looks like this > > > AUTH_TOKEN=12703b11932f233520f6d4b33559c33aeb1cfc7f > MASTER_USER=master at uma > SPUSER=private/pdf > LOCAL_IP=127.0.0.1 > USER=pdf > AUTH_USER=master at uma > PWD=/var/run/dovecot > USERDB_KEYS=HOME SPUSER MASTER_USER AUTH_TOKEN AUTH_USER > SHLVL=1 > HOME=/var/data/vmail/private/pdf > IP=127.0.0.1 > _=/usr/bin/env > > so the ACL_GROUPS is gone. > > is this intended to be like that. > so groups not included in cache and I have to find another approach? > > anybody else encountered similar problems with some auth Variables and > caching? > > > Greetz Matze > -------------- next part -------------- A non-text attachment was scrubbed... Name: 0x7BCC653A.asc Type: application/pgp-keys Size: 3116 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From andy at andybev.com Wed Aug 5 23:07:04 2015 From: andy at andybev.com (Andrew Beverley) Date: Thu, 06 Aug 2015 00:07:04 +0100 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <1437780774.7496.7.camel@andybev.com> References: <1437780774.7496.7.camel@andybev.com> Message-ID: <1438816024.7950.16.camel@andybev.com> On Sat, 2015-07-25 at 00:32 +0100, Andrew Beverley wrote: > Hi, > > I've just upgrade from Debian Wheezy to Debian Jessie and am getting: > > Fatal: master: service(lmtp): child 6761 killed with signal 11 (core dumped) > > It seems to be something to do with sieve. When I disable that from lmtp then > everything works fine. > > OS: Debian Jessie > Dovecot version: 2.2.13 > CPU: x86 > > This is the gdb output: > > Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1". > Core was generated by `dovecot/lmtp'. > Program terminated with signal SIGSEGV, Segmentation fault. > #0 0x00007f8e4c94f626 in sieve_validator_register_command () from > /usr/lib/dovecot/libdovecot-sieve.so.0 I've just tried upgrading to 2.2.18 (Stretch) but get the same error. I've opened a Debian bug report: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=794718 I have also tried downgrading to 2.1.7 (Wheezy) and everything works correctly. Thanks, Andy From stephan at rename-it.nl Thu Aug 6 06:46:02 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Thu, 6 Aug 2015 08:46:02 +0200 Subject: Dovecot/Pigeonhole Issue (UTC modtime instead of localtime) In-Reply-To: <20150804170317.1dbf7a36@aeshna> References: <20150731160115.2df2f412@aeshna> <55BB8CDB.6010809@rename-it.nl> <20150804170317.1dbf7a36@aeshna> Message-ID: <55C302AA.8010003@rename-it.nl> Op 8/4/2015 om 5:03 PM schreef Christoph Gr?ver: > Hello ML, Hello Stephan, > >> Hmm. Probably, the timezone configuration (i.e. the contents of TZ >> timezone environment variable) somehow doesn't reach the final stages >> of e-mail delivery. > Well. I tried several ways of telling the lda or whatever is setting up > the INTERNALDATE to use the CEST +0200. > Although it is already setup on the server via a symlink > from /etc/localtime to /usr/share/zoneinfo/Europe/Berlin, I defined it > at several stages, for example globally and for the account of the > virtual user of the mails. But all of it to no avail. > It just keeps getting a mtime in the past. > > So I actually had to patch the sources. I commented out the utime calls > of src/lib-storage/index/maildir/maildir-save.c . > This way the file gets the correct mtime. > > Still the clients saw the wrong INTERNALDATE. > I also had to completely disable writing a index.cache by dovecot. > > Both are not ideal in my opinion. I'd still prefer a real solution. > But for the moment I had to do it because my users became "desperate". > > If you have any ideas left, please tell me. I'd still like to here them. I have none so far. It can take a while before Timo has time to investigate this further. Regards, Stephan. From stephan at rename-it.nl Thu Aug 6 07:12:13 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Thu, 6 Aug 2015 09:12:13 +0200 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <1437780774.7496.7.camel@andybev.com> References: <1437780774.7496.7.camel@andybev.com> Message-ID: <55C308CD.5070308@rename-it.nl> Op 7/25/2015 om 1:32 AM schreef Andrew Beverley: > Hi, > > I've just upgrade from Debian Wheezy to Debian Jessie and am getting: > > Fatal: master: service(lmtp): child 6761 killed with signal 11 (core dumped) > > It seems to be something to do with sieve. When I disable that from lmtp then > everything works fine. > > OS: Debian Jessie > Dovecot version: 2.2.13 > CPU: x86 > > This is the gdb output: > > Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1". > Core was generated by `dovecot/lmtp'. > Program terminated with signal SIGSEGV, Segmentation fault. > #0 0x00007f8e4c94f626 in sieve_validator_register_command () from /usr/lib/dovecot/libdovecot-sieve.so.0 Could you provide a full backtrace using the gdb command `bt full'? Regards, Stephan. From andy at andybev.com Thu Aug 6 07:35:56 2015 From: andy at andybev.com (Andrew Beverley) Date: Thu, 06 Aug 2015 08:35:56 +0100 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <55C308CD.5070308@rename-it.nl> References: <1437780774.7496.7.camel@andybev.com> <55C308CD.5070308@rename-it.nl> Message-ID: <1438846556.14063.7.camel@andybev.com> On Thu, 2015-08-06 at 09:12 +0200, Stephan Bosch wrote: > Could you provide a full backtrace using the gdb command `bt full'? Thanks for the reply. Is this everything you need? #0 0x00007f8553969626 in sieve_validator_register_command () from /usr/lib/dovecot/libdovecot-sieve.so.0 No symbol table info available. #1 0x00007f8552eff4e6 in ext_extdata_validator_load (ext=0x7f8556f60280, valdtr=0xffffffff) at ext-extdata.c:46 No locals. #2 0x00007f85539791a4 in sieve_extension_unregister () from /usr/lib/dovecot/libdovecot-sieve.so.0 No symbol table info available. #3 0x00007f855397950e in sieve_plugins_unload () from /usr/lib/dovecot/libdovecot-sieve.so.0 No symbol table info available. #4 0x00007f855397e8dc in sieve_deinit () from /usr/lib/dovecot/libdovecot-sieve.so.0 No symbol table info available. #5 0x00007f8553bc7948 in ?? () from /usr/lib/dovecot/modules/lib90_sieve_plugin.so No symbol table info available. #6 0x00007f855497ad69 in mail_deliver () from /usr/lib/dovecot/libdovecot-lda.so.0 No symbol table info available. #7 0x00007f8554daa171 in ?? () No symbol table info available. #8 0x00007f85543f4d0f in io_loop_call_io () from /usr/lib/dovecot/libdovecot.so.0 No symbol table info available. #9 0x00007f85543f5d09 in io_loop_handler_run_internal () from /usr/lib/dovecot/libdovecot.so.0 No symbol table info available. #10 0x00007f85543f4d79 in io_loop_handler_run () from /usr/lib/dovecot/libdovecot.so.0 No symbol table info available. #11 0x00007f85543f4df8 in io_loop_run () from /usr/lib/dovecot/libdovecot.so.0 No symbol table info available. #12 0x00007f855439fdc3 in master_service_run () from /usr/lib/dovecot/libdovecot.so.0 No symbol table info available. #13 0x00007f8554da89b5 in main () From dovecot at tlinx.org Thu Aug 6 07:36:36 2015 From: dovecot at tlinx.org (Linda A. Walsh) Date: Thu, 06 Aug 2015 00:36:36 -0700 Subject: Dovecot under Linux with mail-extension and case insesitiv In-Reply-To: <55B93AB8.5030603@valckenier-kips.de> References: <55B93AB8.5030603@valckenier-kips.de> Message-ID: <55C30E84.7010008@tlinx.org> Sascha wrote: > Hello, > > i use doevecot 2.2.18 current. My Problem is with email-extension and > case sensitiv folders. > I've had similar problems. As I run Win-7 against a Samba-linux server. Win-7 does case-preserving+case-ignore -- samba does case Mangling -- it seems to uppercase machine names & lowercase user names... causes alot of chaos.... I'm only 1-user, so I changed my /etc/passwd file to have 2-3 names for each user-id I wanted to show up: root:x:0:0:root:/root:/bin/bash Bliss\root:x:0:0:root:/root:/bin/bash BLISS\root:x:0:0:root:/root:/bin/bash At times/over the years, I've had Samba give me each variation at least once. But it's not just samba... different brands of 'ssh' will pass different things -- some just 'user', some Domain\User, some Domain/User(cygwin). Since most of my personal users are all lower case, they don't usually need special attention, but in home directories under 'Bliss', have a symlink BLISS->Bliss. On some shares also needed another symlink bliss->Bliss. (ARRGGGG)...(BTW - Bliss was meant as "irony"... so symlinks might help but will look ugly)...it can get worse -- in user/group names, sometimes spaces are removed.. 'Power Users' => powerusers...... ARG!!!.... But when I've talked to dovecot... I always have just needed the user name (no Domain)... Maybe your email software has some setting to tell it to "play nice"?.... Sorry for the "bad idea for large-scale implementation, but it was easiest to do for a few users..(home network)... BTW... some file systems are moving toward ignoring case... xfs has had a case-ignore option but only works in ascii range. Heard apple' implemented case-preserve+ignore on theirs, and seem to remember reading somthing about Solaris(Sun/Oracle)' ZFS getting a case-ignore+preserve option. Good luck! > Example: > user+extenstion will be delivered to the user and subfolder extenstion > so this is okay. > but user+extenstion will not be delivert to the user and exiting > subfolder Extenstion so that is not okay. > The mail will be also delivered in the subfolder extension. The exiting > Folder Extension will not used. > > Give's a way to tell dovecot, use the extension as a Folder, if a > exiting folder with Upper/Lower Cases will be detect and use the right > folder to delivery? > > -- Regards Sascha > > --- > Diese E-Mail wurde von Avast Antivirus-Software auf Viren gepr?ft. > https://www.avast.com/antivirus > > From stephan at rename-it.nl Thu Aug 6 07:49:35 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Thu, 6 Aug 2015 09:49:35 +0200 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <1438846556.14063.7.camel@andybev.com> References: <1437780774.7496.7.camel@andybev.com> <55C308CD.5070308@rename-it.nl> <1438846556.14063.7.camel@andybev.com> Message-ID: <55C3118F.3050805@rename-it.nl> Op 8/6/2015 om 9:35 AM schreef Andrew Beverley: > On Thu, 2015-08-06 at 09:12 +0200, Stephan Bosch wrote: >> Could you provide a full backtrace using the gdb command `bt full'? > Thanks for the reply. Is this everything you need? You have no debug symbols installed. Could you install dovecot-dbg package and try again? Regards, Stephan From andy at andybev.com Thu Aug 6 08:13:31 2015 From: andy at andybev.com (Andrew Beverley) Date: Thu, 06 Aug 2015 09:13:31 +0100 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <55C3118F.3050805@rename-it.nl> References: <1437780774.7496.7.camel@andybev.com> <55C308CD.5070308@rename-it.nl> <1438846556.14063.7.camel@andybev.com> <55C3118F.3050805@rename-it.nl> Message-ID: <1438848811.18074.3.camel@andybev.com> On Thu, 2015-08-06 at 09:49 +0200, Stephan Bosch wrote: > You have no debug symbols installed. Could you install dovecot-dbg > package and try again? Sorry, how's this: #0 sieve_validator_find_command_registration (valdtr=0xffffffff, command=0x7fc573c4bcd8 "extdata") at sieve-validator.c:309 No locals. #1 sieve_validator_register_command (valdtr=0xffffffff, ext=0x7fc5761b1280, cmd_def=0x7fc573e4c400 ) at sieve-validator.c:331 cmd_reg = #2 0x00007fc573c4b4e6 in ext_extdata_validator_load (ext=0x7fc5761b1280, valdtr=0xffffffff) at ext-extdata.c:46 No locals. #3 0x00007fc5746c51a4 in _sieve_extension_unload (ext=) at sieve-extensions.c:316 No locals. #4 sieve_extension_unregister (ext=) at sieve-extensions.c:426 ext_reg = ext_id = #5 0x00007fc5746c550e in sieve_plugins_unload (svinst=0x7fc5761b09b0) at sieve-plugins.c:166 module = 0x7fc5761b5500 unload_func = plugin = 0x7fc5761b1268 __FUNCTION__ = "sieve_plugins_unload" #6 0x00007fc5746ca8dc in sieve_deinit (svinst=svinst at entry=0x7ffdd3a36050) at sieve.c:136 No locals. #7 0x00007fc574913948 in lda_sieve_deliver_mail (mdctx=, storage_r=0x7ffdd3a36208) at lda-sieve-plugin.c:948 srctx = {svinst = 0x7fc5761b09b0, mdctx = 0x7ffdd3a36230, home_dir = 0x7fc5761acf78 "/var/mail/vhosts/xx/yy", scripts = 0x7fc576171088, script_count = 0, user_script = 0x0, main_script = 0x0, msgdata = 0x0, scriptenv = 0x0, user_ehandler = 0x0, master_ehandler = 0x7fc5761b56c0, userlog = 0x0} debug = svenv = {hostname = 0x7fc57618cb98 "xx.com", domainname = 0x0, base_dir = 0x7fc5761ac8c0 "/var/run/dovecot", username = 0x7fc5761abda8 "yy at xx.com", home_dir = 0x7fc5761acf78 "/var/mail/vhosts/xx/yy", flags = SIEVE_FLAG_HOME_RELATIVE, location = SIEVE_ENV_LOCATION_MDA, delivery_phase = SIEVE_DELIVERY_PHASE_DURING} i = ret = #8 0x00007fc5756c6d69 in mail_deliver (ctx=ctx at entry=0x7ffdd3a36230, storage_r=storage_r at entry=0x7ffdd3a36208) at mail-deliver.c:400 ret = #9 0x00007fc575af6171 in client_deliver (session=0x7fc5761aa060, src_mail=0x7fc5761a5840, rcpt=0x7fc57617e7d0, client=0x7fc57617df50) at commands.c:689 lda_set = 0x7fc57618cb28 ns = set_parser = line = storage = 0x0 sets = mail_error = 1981272992 ret = dctx = {pool = 0x7fc5761aa040, set = 0x7fc57618cb28, session = 0x7fc5761aa060, dup_ctx = 0x7fc5761b08d0, session_id = 0x7fc57617e760 "8WveM2IVw1V5fgAAChoztw", src_mail = 0x7fc5761a5840, src_envelope_sender = 0x7fc57617e778 "xx at hotmail.com", dest_user = 0x7fc5761abcd0, dest_addr = 0x7fc57617edd0 "yy at xx.com", final_dest_addr = 0x7fc57617edd0 "yy at xx.com", dest_mailbox_name = 0x7fc575af8ad0 "INBOX", dest_mail = 0x0, var_expand_table = 0x0, tempfail_error = 0x0, tried_default_save = false, saved_mail = false, save_dest_mail = false, mailbox_full = false, dsn = false} input = mail_set = error = username = #10 client_deliver_next (session=0x7fc5761aa060, src_mail=0x7fc5761a5840, client=0x7fc57617df50) at commands.c:732 count = #11 client_input_data_write_local (input=, client=0x7fc57617df50) at commands.c:827 src_mail = 0x7fc5761a5840 first_uid = 4294967295 session = 0x7fc5761aa060 old_uid = 0 #12 client_input_data_write (client=0x7fc57617df50) at commands.c:939 input = 0x7fc57618dc50 ret = true #13 client_input_data_handle (client=0x7fc57617df50) at commands.c:1033 data = size = 934 ret = #14 0x00007fc575140d0f in io_loop_call_io (io=0x7fc57617d3a0) at ioloop.c:441 ioloop = 0x7fc576178730 t_id = 2 __FUNCTION__ = "io_loop_call_io" #15 0x00007fc575141d09 in io_loop_handler_run_internal (ioloop=ioloop at entry=0x7fc576178730) at ioloop-epoll.c:220 ctx = 0x7fc57617b550 io = tv = {tv_sec = 299, tv_usec = 983558} events_count = msecs = ret = 1 i = 0 j = call = __FUNCTION__ = "io_loop_handler_run_internal" #16 0x00007fc575140d79 in io_loop_handler_run (ioloop=ioloop at entry=0x7fc576178730) at ioloop.c:488 No locals. #17 0x00007fc575140df8 in io_loop_run (ioloop=0x7fc576178730) at ioloop.c:465 __FUNCTION__ = "io_loop_run" #18 0x00007fc5750ebdc3 in master_service_run (service=0x7fc5761785c0, callback=) at master-service.c:566 No locals. #19 0x00007fc575af49b5 in main (argc=1, argv=0x7fc576178390) at main.c:122 set_roots = {0x7fc5758ca4c0 , 0x7fc575cfa6c0 , 0x0} service_flags = storage_service_flags = c = From matthias.lay at securepoint.de Thu Aug 6 10:07:29 2015 From: matthias.lay at securepoint.de (matthias lay) Date: Thu, 6 Aug 2015 12:07:29 +0200 Subject: question on autch cache parameters In-Reply-To: <55C22CC4.5030006@securepoint.de> References: <55C21E22.10808@securepoint.de> <55C22CC4.5030006@securepoint.de> Message-ID: <55C331E1.8060702@securepoint.de> hi timo, I checked out the commit causing this. its this one: http://hg.dovecot.org/dovecot-2.2/diff/5e445c659f89/src/auth/auth-request.c#l1.32 if I move this block back as it was. everything is fine diff -r a46620d6e0ff -r 5e445c659f89 src/auth/auth-request.c --- a/src/auth/auth-request.c Tue May 05 13:35:52 2015 +0300 +++ b/src/auth/auth-request.c Tue May 05 14:16:31 2015 +0300 @@ -618,30 +627,28 @@ auth_request_want_skip_passdb(request, next_passdb)) next_passdb = next_passdb->next; + if (*result == PASSDB_RESULT_OK) { + /* this passdb lookup succeeded, preserve its extra fields */ + auth_fields_snapshot(request->extra_fields); + request->snapshot_have_userdb_prefetch_set = + request->userdb_prefetch_set; + if (request->userdb_reply != NULL) + auth_fields_snapshot(request->userdb_reply); + } else { + /* this passdb lookup failed, remove any extra fields it set */ + auth_fields_rollback(request->extra_fields); + if (request->userdb_reply != NULL) { + auth_fields_rollback(request->userdb_reply); + request->userdb_prefetch_set = + request->snapshot_have_userdb_prefetch_set; + } + } + if (passdb_continue && next_passdb != NULL) { /* try next passdb. */ request->passdb = next_passdb; request->passdb_password = NULL; - if (*result == PASSDB_RESULT_OK) { - /* this passdb lookup succeeded, preserve its extra - fields */ - auth_fields_snapshot(request->extra_fields); - request->snapshot_have_userdb_prefetch_set = - request->userdb_prefetch_set; - if (request->userdb_reply != NULL) - auth_fields_snapshot(request->userdb_reply); - } else { - /* this passdb lookup failed, remove any extra fields - it set */ - auth_fields_rollback(request->extra_fields); - if (request->userdb_reply != NULL) { - auth_fields_rollback(request->userdb_reply); - request->userdb_prefetch_set = - request->snapshot_have_userdb_prefetch_set; - } - } - if (*result == PASSDB_RESULT_USER_UNKNOWN) { /* remember that we did at least one successful passdb lookup */ On 08/05/2015 05:33 PM, matthias lay wrote: > just tested against dovecot 2.2.15 > > everythings works fine. so might be a bug introduced between 2.2.16 and > 2.2.18 > > > > > > On 08/05/2015 04:30 PM, matthias lay wrote: >> Hi list, >> >> I have a question on auth caching in 2.2.18. >> >> I am using acl_groups for a master user, appended in a static userdb file >> >> # snip ############################### >> master at uma:{SHA}XXXX=::::::userdb_acl_groups=umareadmaster >> allow_nets=127.0.0.1 >> # snap ############################### >> >> and use this group in a global ACL file. >> I discovered this only works on first NOT-cached login >> >> >> >> environment in imap-postlogin script on first login: >> >> >> AUTH_TOKEN=e96b5a32ceb2cafc4460c210ad2e92e3d7ab388c >> MASTER_USER=master at uma >> SPUSER=private/pdf >> LOCAL_IP=127.0.0.1 >> USER=pdf >> AUTH_USER=master at uma >> PWD=/var/run/dovecot >> USERDB_KEYS=ACL_GROUPS HOME SPUSER MASTER_USER AUTH_TOKEN AUTH_USER >> SHLVL=1 >> HOME=/var/data/vmail/private/pdf >> ACL_GROUPS=umareadmaster >> IP=127.0.0.1 >> _=/usr/bin/env >> >> >> on the second cached login it looks like this >> >> >> AUTH_TOKEN=12703b11932f233520f6d4b33559c33aeb1cfc7f >> MASTER_USER=master at uma >> SPUSER=private/pdf >> LOCAL_IP=127.0.0.1 >> USER=pdf >> AUTH_USER=master at uma >> PWD=/var/run/dovecot >> USERDB_KEYS=HOME SPUSER MASTER_USER AUTH_TOKEN AUTH_USER >> SHLVL=1 >> HOME=/var/data/vmail/private/pdf >> IP=127.0.0.1 >> _=/usr/bin/env >> >> so the ACL_GROUPS is gone. >> >> is this intended to be like that. >> so groups not included in cache and I have to find another approach? >> >> anybody else encountered similar problems with some auth Variables and >> caching? >> >> >> Greetz Matze >> > From dale.gallagher at gmail.com Thu Aug 6 10:10:25 2015 From: dale.gallagher at gmail.com (Dale Gallagher) Date: Thu, 6 Aug 2015 12:10:25 +0200 Subject: 2.2.18: Mailbox INBOX sync: mailbox_delete failed: INBOX can't be deleted. In-Reply-To: References: Message-ID: Hello Could my previously reported error above have something to do with the IMAP path prefix? It's not configured on either of the previously mentioned servers. I see a similar error, which may be related, when trying to sync using offlineimap.py: Folder INBOX [acc: test_account]: ERROR: ERROR in syncfolder for test_account folder INBOX Thanks Dale On 5 August 2015 at 16:59, Dale Gallagher wrote: > Hello > > I have 2.2.18 built with: > > ./configure \ > --prefix="$prefix" \ > --bindir="$prefix/bin" \ > --sbindir="$prefix/bin" \ > --sysconfdir="/etc" \ > --with-storages="maildir,imapc,pop3c" \ > --without-vpopmail \ > --without-ldap > > I'm attempting to sync (backup) from another Dovecot server running > version 2.0.21: > > Both servers don't have no explicit namespaces configured and both are > using the Maildir storage plugin. > > doveadm -Dv -o imapc_user=remote_user -o imapc_host=source.host -o > imapc_port=993 -o imapc_ssl=imaps -o imapc_password=remote_pass -o > imapc_features="rfc822.size fetch-headers" backup -R -u local_user imapc: > > All seems to go well, showing local and remote mailbox syncing without > errors, then finally: > > dsync(mcg): Debug: brain M: Deleting mailbox 'INBOX' (GUID > 59023c2a911dc255842b0000d09efc50): UIDVALIDITY changed (1355416734 -> > 1438784913) > dsync(mcg): Error: Mailbox INBOX sync: mailbox_delete failed: INBOX can't > be deleted. > > No mail is synced to the local server (2.2.18) after this error halts > execution. > > What should I be looking for, in terms of possible causes of this error? > > I've tried syncing on a completely fresh mailbox, which hasn't been > accessed by a mail client and the same error occurs. > > I've seen similar reports of this error before, without clear solutions / > suggestions. > > Help would be greatly appreciated. > > Thanks > > > > From AchilovRN at granch.ru Thu Aug 6 11:15:05 2015 From: AchilovRN at granch.ru (=?UTF-8?B?0JDRh9C40LvQvtCyINCg0LDRiNC40LQg0J3Rg9GA0LzRg9GF0LDQvNC10LQ=?= =?UTF-8?B?0L7QstC40Yc=?=) Date: Thu, 6 Aug 2015 17:15:05 +0600 Subject: SiS does not work Message-ID: <55C341B9.9080203@granch.ru> How do you do. I have tried to use SiS to deduplicate attachments in mail storage. Unsuccesful. When goes mail with reasonable size, and SiS try to start, I have receive error, when this mail must be delivered for group of receivers (i.e. through aliases). Below is error log, valuable parts of mail addresses masked by X and Y chars Aug 6 16:11:12 mailsrv dovecot: lmtp(xxxxx): AL1SNrkyw1VPBAAABJcpmA: msgid=<2015080614471325760993 at chinacoalchem.com> (from zsy at chinacoalchem.com): saved mail to INBOX This receiver has received mail. But next one does not. And to receive I must restart _sendmail_ (not dovecot!). I must restart sendmail as many times as there were recipients in envelope. Aug 6 16:11:12 mailsrv dovecot: lmtp(yyyy): Error: read(/usr/share/smbusers/xxxxx/mail/attachments /e4/3d/e43db6e8af6fb1df4a08f9c1f3f31e93e61781c4c1d2395e04dd96a583aa-01bd5236b932c3554f04000004972998[base64 :19 b/l]) failed: Stream is smaller than expected (0 < 4080128) Aug 6 16:11:12 mailsrv dovecot: lmtp(yyyy): Error: read(attachments-connector(/usr/share/smbusers/xxxxx/mail/storage/m.35)) failed: read(/usr/share/smbusers/xxxxx/mail/attachments/e4/3d/e43db6e8af6f b1df4a08f9c1f3f31e93e61781c4c1d2395e04dd96a583aa-01bd5236b932c3554f04000004972998[base64:19 b/l]) failed: Stream is smaller than expected (0 < 4080128) Aug 6 16:11:12 mailsrv dovecot: lmtp(yyyy): Error: copy: i_stream_read(attachments-connector(/usr/share/smbusers/xxxxx/mail/storage/m.35)) failed: read(/usr/share/smbusers/xxxxx/mail/attachments/e4/ 3d/e43db6e8af6fb1df4a08f9c1f3f31e93e61781c4c1d2395e04dd96a583aa-01bd5236b932c3554f04000004972998[base64:19 b/l]) failed: Stream is smaller than expected (0 < 4080128) Aug 6 16:11:12 mailsrv dovecot: lmtp(yyyy): Error: read(attachments-connector(/usr/share/smbusers/xxxxx/mail/storage/m.35)) failed: read(/usr/share/smbusers/xxxxx/mail/attachments/e4/3d/e43db6e8af6f b1df4a08f9c1f3f31e93e61781c4c1d2395e04dd96a583aa-01bd5236b932c3554f04000004972998[base64:19 b/l]) failed: Stream is smaller than expected (0 < 4080128) (uid=0, box=INBOX) Aug 6 16:11:12 mailsrv dovecot: lmtp(yyyy): AL1SNrkyw1VPBAAABJcpmA: msgid=<2015080614471325760993 at chinacoalchem.com> (from zsy at chinacoalchem.com): save failed to INBOX: Internal error occurred. Refer to server log for more information. [2015-08-06 16:11:12] Why LMTP, when try to deliver mail to local user yyyy tries to look up into a home directory of local user xxxxx? All these and similar home directories has a 'mail' subdir, which keeps mail and rights on this subdir is 0700. To avoid this completely misunderstood occassion, I have temporary set mail_attachment_min_size to 2048kb - but now I cannot receive advantage from SiS. -- ? ?????????. ?????? ????? ?????????????? (AchilovRN at granch.ru) ??????? ?????????? ?? ?????? ?????????? ??? ??? "?????", ???: +7 (383) 233-35-12, ???. 107 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2606 bytes Desc: ?????????????????????????????????? ?????????????? S/MIME URL: From stephan at rename-it.nl Thu Aug 6 14:10:58 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Thu, 06 Aug 2015 16:10:58 +0200 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <1438848811.18074.3.camel@andybev.com> References: <1437780774.7496.7.camel@andybev.com> <55C308CD.5070308@rename-it.nl> <1438846556.14063.7.camel@andybev.com> <55C3118F.3050805@rename-it.nl> <1438848811.18074.3.camel@andybev.com> Message-ID: <55C36AF2.5090202@rename-it.nl> What version of the extdata plugin are you using? This looks like a version built for Dovecot v2.1/Pigeonhole 0.3.0. Keep in mind that the extdata plugin is not part of the Pigeonhole distribution, so that will not be upgraded along with Pigeonhole. Since the extdata plugin is not part of the normal Debian packages (not afaik anyway), you likely compiled and installed the extdata plugin manually in the past while using Dovecot 2.1. Do that again, but use this repository: http://hg.rename-it.nl/pigeonhole-0.4-sieve-extdata/ Regards, Stephan. Andrew Beverley schreef op 6-8-2015 om 10:13: > On Thu, 2015-08-06 at 09:49 +0200, Stephan Bosch wrote: >> You have no debug symbols installed. Could you install dovecot-dbg >> package and try again? > Sorry, how's this: > > #0 sieve_validator_find_command_registration (valdtr=0xffffffff, command=0x7fc573c4bcd8 "extdata") at sieve-validator.c:309 > No locals. > #1 sieve_validator_register_command (valdtr=0xffffffff, ext=0x7fc5761b1280, cmd_def=0x7fc573e4c400 ) > at sieve-validator.c:331 > cmd_reg = > #2 0x00007fc573c4b4e6 in ext_extdata_validator_load (ext=0x7fc5761b1280, valdtr=0xffffffff) at ext-extdata.c:46 > No locals. > #3 0x00007fc5746c51a4 in _sieve_extension_unload (ext=) at sieve-extensions.c:316 > No locals. > #4 sieve_extension_unregister (ext=) at sieve-extensions.c:426 > ext_reg = > ext_id = > #5 0x00007fc5746c550e in sieve_plugins_unload (svinst=0x7fc5761b09b0) at sieve-plugins.c:166 > module = 0x7fc5761b5500 > unload_func = > plugin = 0x7fc5761b1268 > __FUNCTION__ = "sieve_plugins_unload" > #6 0x00007fc5746ca8dc in sieve_deinit (svinst=svinst at entry=0x7ffdd3a36050) at sieve.c:136 > No locals. > #7 0x00007fc574913948 in lda_sieve_deliver_mail (mdctx=, storage_r=0x7ffdd3a36208) at lda-sieve-plugin.c:948 > srctx = {svinst = 0x7fc5761b09b0, mdctx = 0x7ffdd3a36230, > home_dir = 0x7fc5761acf78 "/var/mail/vhosts/xx/yy", scripts = 0x7fc576171088, script_count = 0, > user_script = 0x0, main_script = 0x0, msgdata = 0x0, scriptenv = 0x0, user_ehandler = 0x0, > master_ehandler = 0x7fc5761b56c0, userlog = 0x0} > debug = > svenv = {hostname = 0x7fc57618cb98 "xx.com", domainname = 0x0, base_dir = 0x7fc5761ac8c0 "/var/run/dovecot", > username = 0x7fc5761abda8 "yy at xx.com", > home_dir = 0x7fc5761acf78 "/var/mail/vhosts/xx/yy", flags = SIEVE_FLAG_HOME_RELATIVE, > location = SIEVE_ENV_LOCATION_MDA, delivery_phase = SIEVE_DELIVERY_PHASE_DURING} > i = > ret = > #8 0x00007fc5756c6d69 in mail_deliver (ctx=ctx at entry=0x7ffdd3a36230, storage_r=storage_r at entry=0x7ffdd3a36208) at mail-deliver.c:400 > ret = > #9 0x00007fc575af6171 in client_deliver (session=0x7fc5761aa060, src_mail=0x7fc5761a5840, rcpt=0x7fc57617e7d0, > client=0x7fc57617df50) at commands.c:689 > lda_set = 0x7fc57618cb28 > ns = > set_parser = > line = > storage = 0x0 > sets = > mail_error = 1981272992 > ret = > dctx = {pool = 0x7fc5761aa040, set = 0x7fc57618cb28, session = 0x7fc5761aa060, dup_ctx = 0x7fc5761b08d0, > session_id = 0x7fc57617e760 "8WveM2IVw1V5fgAAChoztw", src_mail = 0x7fc5761a5840, > src_envelope_sender = 0x7fc57617e778 "xx at hotmail.com", dest_user = 0x7fc5761abcd0, > dest_addr = 0x7fc57617edd0 "yy at xx.com", final_dest_addr = 0x7fc57617edd0 "yy at xx.com", > dest_mailbox_name = 0x7fc575af8ad0 "INBOX", dest_mail = 0x0, var_expand_table = 0x0, tempfail_error = 0x0, > tried_default_save = false, saved_mail = false, save_dest_mail = false, mailbox_full = false, dsn = false} > input = > mail_set = > error = > username = > #10 client_deliver_next (session=0x7fc5761aa060, src_mail=0x7fc5761a5840, client=0x7fc57617df50) at commands.c:732 > count = > #11 client_input_data_write_local (input=, client=0x7fc57617df50) at commands.c:827 > src_mail = 0x7fc5761a5840 > first_uid = 4294967295 > session = 0x7fc5761aa060 > old_uid = 0 > #12 client_input_data_write (client=0x7fc57617df50) at commands.c:939 > input = 0x7fc57618dc50 > ret = true > #13 client_input_data_handle (client=0x7fc57617df50) at commands.c:1033 > data = > size = 934 > ret = > #14 0x00007fc575140d0f in io_loop_call_io (io=0x7fc57617d3a0) at ioloop.c:441 > ioloop = 0x7fc576178730 > t_id = 2 > __FUNCTION__ = "io_loop_call_io" > #15 0x00007fc575141d09 in io_loop_handler_run_internal (ioloop=ioloop at entry=0x7fc576178730) at ioloop-epoll.c:220 > ctx = 0x7fc57617b550 > io = > tv = {tv_sec = 299, tv_usec = 983558} > events_count = > msecs = > ret = 1 > i = 0 > j = > call = > __FUNCTION__ = "io_loop_handler_run_internal" > #16 0x00007fc575140d79 in io_loop_handler_run (ioloop=ioloop at entry=0x7fc576178730) at ioloop.c:488 > No locals. > #17 0x00007fc575140df8 in io_loop_run (ioloop=0x7fc576178730) at ioloop.c:465 > __FUNCTION__ = "io_loop_run" > #18 0x00007fc5750ebdc3 in master_service_run (service=0x7fc5761785c0, callback=) at master-service.c:566 > No locals. > #19 0x00007fc575af49b5 in main (argc=1, argv=0x7fc576178390) at main.c:122 > set_roots = {0x7fc5758ca4c0 , 0x7fc575cfa6c0 , 0x0} > service_flags = > storage_service_flags = > c = From superinterstellar at gmail.com Thu Aug 6 15:06:21 2015 From: superinterstellar at gmail.com (Kevin Laurie) Date: Thu, 6 Aug 2015 23:06:21 +0800 Subject: backing up IMAP server on a hard drive Message-ID: Hi, I am trying to back up my IMAP server to a hard drive. Later I intend to extract all mails for attachments. What do you reckon is the best too to perform this ? Imapsync or Thunderbird (or something else, please recommend) One problem I am having with imapsync is the setting for host2 (being a localhost computer). How does one set a parameter for host2 being a desktop computer. Thanks Kevin From rick at havokmon.com Thu Aug 6 15:21:12 2015 From: rick at havokmon.com (Rick Romero) Date: Thu, 06 Aug 2015 10:21:12 -0500 Subject: backing up IMAP server on a hard drive In-Reply-To: Message-ID: <20150806102112.Horde.vSPTHO6e4PuheGsNw3FVyg1@www.vfemail.net> Quoting Kevin Laurie : > Hi, > I am trying to back up my IMAP server to a hard drive. Later I intend to > extract all mails for attachments. What do you reckon is the best too to > perform this ? > > Imapsync or Thunderbird (or something else, please recommend) > > One problem I am having with imapsync is the setting for host2 (being a > localhost computer). > How does one set a parameter for host2 being a desktop computer. > > ThanksKevin > ? You could run Mercury/32 http://www.pmail.com/ as a local IMAP server that could be your 'desktop destination'. Though I'd assume the typical backup solution for a smaller environment would probably use rsync (unless your server is a VM, then you could image the whole VM via whatever utilities the host provides). Rick From superinterstellar at gmail.com Thu Aug 6 15:51:41 2015 From: superinterstellar at gmail.com (Kevin Laurie) Date: Thu, 6 Aug 2015 23:51:41 +0800 Subject: backing up IMAP server on a hard drive In-Reply-To: <20150806102112.Horde.vSPTHO6e4PuheGsNw3FVyg1@www.vfemail.net> References: <20150806102112.Horde.vSPTHO6e4PuheGsNw3FVyg1@www.vfemail.net> Message-ID: Dear Rick, Thanks for your feedback. I think rsync might be a better option. Its(imap server)with gmail so I dont think it would work . Furthermore I am running a linux system(Ubuntu 14.04 to be precise). Does Mercury 32 support it? >From the site it seems not. Please advise. Regards Kevin On Thursday, August 6, 2015, Rick Romero wrote: > Quoting Kevin Laurie : > > Hi, >> I am trying to back up my IMAP server to a hard drive. Later I intend to >> extract all mails for attachments. What do you reckon is the best too to >> perform this ? >> >> Imapsync or Thunderbird (or something else, please recommend) >> >> One problem I am having with imapsync is the setting for host2 (being a >> localhost computer). >> How does one set a parameter for host2 being a desktop computer. >> >> ThanksKevin >> >> > > You could run Mercury/32 http://www.pmail.com/ as a local IMAP server that > could be your 'desktop destination'. > > Though I'd assume the typical backup solution for a smaller environment > would probably use rsync (unless your server is a VM, then you could image > the whole VM via whatever utilities the host provides). > > Rick > From rick at havokmon.com Thu Aug 6 16:06:13 2015 From: rick at havokmon.com (Rick Romero) Date: Thu, 06 Aug 2015 11:06:13 -0500 Subject: backing up IMAP server on a hard drive In-Reply-To: References: <20150806102112.Horde.vSPTHO6e4PuheGsNw3FVyg1@www.vfemail.net> Message-ID: <20150806110613.Horde.Hi0mD3dS1oJ42DKObmZ9SQ8@www.vfemail.net> Hi Kevin, Ahh. When you said desktop, I assumed you meant Windows.? Dovecot is an IMAP Server, so I assume that was the server you wanted to backup. If you're running Linux, then you could run Dovecot on it and use IMAPSync to sync your Gmail locally. Rick Quoting Kevin Laurie : > Dear Rick, > Thanks for your feedback.? > I think rsync might be a better option. Its(imap server)with gmail so > I dont think it would work . > ? > Furthermore I am running a linux system(Ubuntu 14.04 to be precise). > Does Mercury 32 support it? > From the site it seems not. > ? > Please advise.? > Regards > Kevin? > ? > ? > ? > On Thursday, August 6, 2015, Rick Romero wrote: > >> ?Quoting Kevin Laurie : >> >>> Hi, >>> I am trying to back up my IMAP server to a hard drive. Later I intend to >>> extract all mails for attachments. What do you reckon is the best too to >>> perform this ? >>> >>> Imapsync or Thunderbird (or something else, please recommend) >>> >>> One problem I am having with imapsync is the setting for host2 (being a >>> localhost computer). >>> How does one set a parameter for host2 being a desktop computer. >>> >>> ThanksKevin >>> ? >> >> You could run Mercury/32 http://www.pmail.com/ as a local IMAP server >> that >> could be your 'desktop destination'. >> >> Though I'd assume the typical backup solution for a smaller environment >> would probably use rsync (unless your server is a VM, then you could >> image >> the whole VM via whatever utilities the host provides). >> >> Rick Links: ------ [1] From superinterstellar at gmail.com Thu Aug 6 16:11:05 2015 From: superinterstellar at gmail.com (Kevin Laurie) Date: Fri, 7 Aug 2015 00:11:05 +0800 Subject: backing up IMAP server on a hard drive In-Reply-To: <20150806110613.Horde.Hi0mD3dS1oJ42DKObmZ9SQ8@www.vfemail.net> References: <20150806102112.Horde.vSPTHO6e4PuheGsNw3FVyg1@www.vfemail.net> <20150806110613.Horde.Hi0mD3dS1oJ42DKObmZ9SQ8@www.vfemail.net> Message-ID: Dear Rick, Thanks. I guess I'll have to install dovecot. I am not sure what the parameter for the -host2 should be(apart from localhost? ) Its my first time doing dovecot on a Linux desktop. I was thinking of using Thunderbird to download all mails but I guess that will be too long of a process. On Friday, August 7, 2015, Rick Romero wrote: > Hi Kevin, > > Ahh. When you said desktop, I assumed you meant Windows. Dovecot is an > IMAP Server, so I assume that was the server you wanted to backup. > > If you're running Linux, then you could run Dovecot on it and use IMAPSync > to sync your Gmail locally. > > Rick > > Quoting Kevin Laurie >: > > Dear Rick, > Thanks for your feedback. > I think rsync might be a better option. Its(imap server)with gmail so I > dont think it would work . > > Furthermore I am running a linux system(Ubuntu 14.04 to be precise). Does > Mercury 32 support it? > From the site it seems not. > > Please advise. > Regards > Kevin > > > > On Thursday, August 6, 2015, Rick Romero > wrote: > >> Quoting Kevin Laurie : >> >> Hi, >>> I am trying to back up my IMAP server to a hard drive. Later I intend to >>> extract all mails for attachments. What do you reckon is the best too to >>> perform this ? >>> >>> Imapsync or Thunderbird (or something else, please recommend) >>> >>> One problem I am having with imapsync is the setting for host2 (being a >>> localhost computer). >>> How does one set a parameter for host2 being a desktop computer. >>> >>> ThanksKevin >>> >>> >> >> You could run Mercury/32 http://www.pmail.com/ as a local IMAP server >> that >> could be your 'desktop destination'. >> >> Though I'd assume the typical backup solution for a smaller environment >> would probably use rsync (unless your server is a VM, then you could image >> the whole VM via whatever utilities the host provides). >> >> Rick > > > > From superinterstellar at gmail.com Thu Aug 6 16:26:07 2015 From: superinterstellar at gmail.com (Kevin Laurie) Date: Thu, 6 Aug 2015 21:56:07 +0530 Subject: backing up IMAP server on a hard drive In-Reply-To: References: <20150806102112.Horde.vSPTHO6e4PuheGsNw3FVyg1@www.vfemail.net> <20150806110613.Horde.Hi0mD3dS1oJ42DKObmZ9SQ8@www.vfemail.net> Message-ID: Dear Rick, I just got dovecot up and running on my localhost PC. Thanks On Thu, Aug 6, 2015 at 9:41 PM, Kevin Laurie wrote: > Dear Rick, > Thanks. > I guess I'll have to install dovecot. > I am not sure what the parameter for the -host2 should be(apart from > localhost? ) > Its my first time doing dovecot on a Linux desktop. > > > I was thinking of using Thunderbird to download all mails but I guess that > will be too long of a process. > > On Friday, August 7, 2015, Rick Romero wrote: >> >> Hi Kevin, >> >> Ahh. When you said desktop, I assumed you meant Windows. Dovecot is an >> IMAP Server, so I assume that was the server you wanted to backup. >> >> If you're running Linux, then you could run Dovecot on it and use IMAPSync >> to sync your Gmail locally. >> >> Rick >> >> Quoting Kevin Laurie : >> >> Dear Rick, >> >> Thanks for your feedback. >> I think rsync might be a better option. Its(imap server)with gmail so I >> dont think it would work . >> >> Furthermore I am running a linux system(Ubuntu 14.04 to be precise). Does >> Mercury 32 support it? >> From the site it seems not. >> >> Please advise. >> Regards >> Kevin >> >> >> >> On Thursday, August 6, 2015, Rick Romero wrote: >>> >>> Quoting Kevin Laurie : >>> >>>> Hi, >>>> I am trying to back up my IMAP server to a hard drive. Later I intend to >>>> extract all mails for attachments. What do you reckon is the best too to >>>> perform this ? >>>> >>>> Imapsync or Thunderbird (or something else, please recommend) >>>> >>>> One problem I am having with imapsync is the setting for host2 (being a >>>> localhost computer). >>>> How does one set a parameter for host2 being a desktop computer. >>>> >>>> ThanksKevin >>>> >>> >>> >>> You could run Mercury/32 http://www.pmail.com/ as a local IMAP server >>> that >>> could be your 'desktop destination'. >>> >>> Though I'd assume the typical backup solution for a smaller environment >>> would probably use rsync (unless your server is a VM, then you could >>> image >>> the whole VM via whatever utilities the host provides). >>> >>> Rick >> >> >> > From andy at andybev.com Thu Aug 6 17:30:49 2015 From: andy at andybev.com (Andrew Beverley) Date: Thu, 06 Aug 2015 18:30:49 +0100 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <55C36AF2.5090202@rename-it.nl> References: <1437780774.7496.7.camel@andybev.com> <55C308CD.5070308@rename-it.nl> <1438846556.14063.7.camel@andybev.com> <55C3118F.3050805@rename-it.nl> <1438848811.18074.3.camel@andybev.com> <55C36AF2.5090202@rename-it.nl> Message-ID: <1438882249.5340.1.camel@andybev.com> On Thu, 2015-08-06 at 16:10 +0200, Stephan Bosch wrote: > What version of the extdata plugin are you using? This looks like a > version built for Dovecot v2.1/Pigeonhole 0.3.0. Keep in mind that the > extdata plugin is not part of the Pigeonhole distribution, so that will > not be upgraded along with Pigeonhole. > > Since the extdata plugin is not part of the normal Debian packages (not > afaik anyway), you likely compiled and installed the extdata plugin > manually in the past while using Dovecot 2.1. Do that again, but use > this repository: http://hg.rename-it.nl/pigeonhole-0.4-sieve-extdata/ You're absolutely right; sorry, I had completely forgotten. I'll give that a go and let you know how I get on. Thanks, Andy From superinterstellar at gmail.com Fri Aug 7 08:54:00 2015 From: superinterstellar at gmail.com (Kevin Laurie) Date: Fri, 7 Aug 2015 14:24:00 +0530 Subject: opening a maildir ? Message-ID: Hello, I have a maildir on my Ubuntu PC. I used imapsync + dovecot to copy all my emails on it. What would be a convenient way to access all the emails ( I want to go through and download all attachments). So a tool with some GUI will be nice. Appreciate if someone could help here. Thanks Kevin From mail at marc-stuermer.de Fri Aug 7 09:31:10 2015 From: mail at marc-stuermer.de (Marc =?utf-8?b?U3TDvHJtZXI=?=) Date: Fri, 07 Aug 2015 11:31:10 +0200 Subject: opening a maildir ? In-Reply-To: Message-ID: <20150807113110.Horde.N3Tp9SQmjPtQdHSQDUPEW56@webmail.no-carrier.info> Zitat von Kevin Laurie : > What would be a convenient way to access all the emails ( I want to go > through and download all attachments). So a tool with some GUI will be > nice. Thunderbird with spool access. https://addons.mozilla.org/de/thunderbird/addon/attachmentextractor/ From matthias.lay at securepoint.de Fri Aug 7 10:19:22 2015 From: matthias.lay at securepoint.de (matthias lay) Date: Fri, 7 Aug 2015 12:19:22 +0200 Subject: pigeonhole/lda accessing -m folder Message-ID: <55C4862A.7070101@securepoint.de> Hi all, anybody knows, if there is a way to acces the folder from a call like this deovecot-lda -m destfolder from within a sieve script? thx in advance matze From Jost.Krieger+dovecot at rub.de Fri Aug 7 10:26:26 2015 From: Jost.Krieger+dovecot at rub.de (Jost Krieger) Date: Fri, 7 Aug 2015 12:26:26 +0200 Subject: pigeonhole/lda accessing -m folder In-Reply-To: <55C4862A.7070101@securepoint.de> References: <55C4862A.7070101@securepoint.de> Message-ID: <20150807102626.GG5380@ruhr-uni-bochum.de> On Fri Aug 7 12:19:22 2015, matthias lay wrote: > anybody knows, if there is a way to acces the folder from a call like this > > deovecot-lda -m destfolder > > from within a sieve script? Depending on what you mean by "access", this could be as easy as require "fileinto"; fileinto "destfolder"; Yours Jost Krieger -- | Jost.Krieger+sig at ruhr-uni-bochum.de Please help stamp out spam! | | Postmaster, JAPH, resident answer machine at RUB Comp. Center | | Sincere words are not sweet, sweet words are not sincere. | | Lao Tse, Tao Te King 81 | -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 4784 bytes Desc: not available URL: From matthias.lay at securepoint.de Fri Aug 7 10:40:13 2015 From: matthias.lay at securepoint.de (matthias lay) Date: Fri, 7 Aug 2015 12:40:13 +0200 Subject: pigeonhole/lda accessing -m folder In-Reply-To: <20150807102626.GG5380@ruhr-uni-bochum.de> References: <55C4862A.7070101@securepoint.de> <20150807102626.GG5380@ruhr-uni-bochum.de> Message-ID: <55C48B0D.5040200@securepoint.de> hi jost thx for the reply, by access I mean to read the variable require ["fileinto", "variables", "?destfolder?" ]; if anyof ( destfolder :matches "*") { fileinto "${1}/subfolder"; }else{ fileinto "INBOX/subfolder"; } On 08/07/2015 12:26 PM, Jost Krieger wrote: > On Fri Aug 7 12:19:22 2015, matthias lay wrote: > >> anybody knows, if there is a way to acces the folder from a call like this >> >> deovecot-lda -m destfolder >> >> from within a sieve script? > > Depending on what you mean by "access", this could be as easy as > > require "fileinto"; > > fileinto "destfolder"; > > Yours > Jost Krieger > From matthias.lay at securepoint.de Fri Aug 7 14:03:31 2015 From: matthias.lay at securepoint.de (matthias lay) Date: Fri, 7 Aug 2015 16:03:31 +0200 Subject: pigeonhole/lda accessing -m folder In-Reply-To: <55C48B0D.5040200@securepoint.de> References: <55C4862A.7070101@securepoint.de> <20150807102626.GG5380@ruhr-uni-bochum.de> <55C48B0D.5040200@securepoint.de> Message-ID: <55C4BAB3.4050507@securepoint.de> fixed my problem. if somebody is interested. I added an variable to the environment extension, which is quite easy. And the value of the folder is already there. so this is all: src/lib-sieve/plugins/environment/ext-environment-common.c static const char *envit_spfolder_get_value (struct sieve_instance *svinst, const struct sieve_script_env *senv) { return senv->default_mailbox; } const struct sieve_environment_item spfolder_env_item = { .name = "spfolder", .get_value = envit_spfolder_get_value, }; after that the variable can easily be used in a script like if anyof ( environment :matches "spfolder" "*") { set "myfolder" "${1}"; } On 08/07/2015 12:40 PM, matthias lay wrote: > hi jost thx for the reply, > > by access I mean to read the variable > > > require ["fileinto", "variables", "?destfolder?" ]; > > if anyof ( destfolder :matches "*") { > > fileinto "${1}/subfolder"; > > }else{ > > fileinto "INBOX/subfolder"; > > } > > > > On 08/07/2015 12:26 PM, Jost Krieger wrote: >> On Fri Aug 7 12:19:22 2015, matthias lay wrote: >> >>> anybody knows, if there is a way to acces the folder from a call like this >>> >>> deovecot-lda -m destfolder >>> >>> from within a sieve script? >> >> Depending on what you mean by "access", this could be as easy as >> >> require "fileinto"; >> >> fileinto "destfolder"; >> >> Yours >> Jost Krieger >> From ml at x-net.be Fri Aug 7 16:04:50 2015 From: ml at x-net.be (Gerry) Date: Fri, 7 Aug 2015 18:04:50 +0200 Subject: Using a separate passdb per service Message-ID: <55C4D722.3030204@x-net.be> Situation: one front-facing server running Dovecot as IMAP/POP3/ ManageSieve proxy, a mixture of IMAP servers (Dovecot, Exchange, ...) in the back-end. Dovecot's passdb does lookups against MySQL which contains a simple user/host mapping, the actual authentication happens on the back-end IMAP servers. The configuration is more or less as described here: http://wiki2.dovecot.org/PasswordDatabase/ExtraFields/Proxy Now I would like to add a Postfix instance on the front-facing server which listens on the submission port and authenticates users via SASL using the local Dovecot's UNIX socket. The idea being that a user only needs to remember one single hostname, one username and one password for all mail-related services. The problem is that Dovecot is operating in proxy mode, which means that the password_query returns NULL as the password and explicitly returns a field "nopasswd" containing "Y". Thus, users can not authenticate against the UNIX socket. What I think I want to do is convince Dovecot to use one passdb for the imap/pop3/managesieve services and different one for the "auth" service. The configuration snippet below doesn't work, but it should illustrate what I want to achieve: > protocols = imap pop3 sieve > > service auth { > passdb sql { > driver = sql > args = /etc/dovecot/mysql-auth-sasl.conf.ext > } > > unix_listener /var/spool/postfix/private/auth { > user = postfix > group = postfix > mode = 0666 > } > } > > # IMAP/POP3/ManageSieve auth against MySQL > passdb sql { > driver = sql > args = /etc/dovecot/mysql-auth-default.conf.ext > } Example mysql-auth-sasl.conf.ext > driver = mysql > connect = host=127.0.0.1 dbname=mail user=mail password=somethingrandom > password_query = SELECT password AS password FROM users WHERE login = '%u' Example mysql-auth-default.conf.ext: > driver = mysql > connect = host=127.0.0.1 dbname=mail user=mail password=somethingrandom > password_query = SELECT NULL AS password, 'Y' as nopassword, host, 'Y' AS proxy FROM users WHERE login = '%u' Any pointers? Gerry From edgar at pettijohn-web.com Sat Aug 8 03:57:51 2015 From: edgar at pettijohn-web.com (Edgar Pettijohn) Date: Fri, 7 Aug 2015 22:57:51 -0500 Subject: Using a separate passdb per service In-Reply-To: <55C4D722.3030204@x-net.be> References: <55C4D722.3030204@x-net.be> Message-ID: <55C57E3F.6050402@pettijohn-web.com> I'm not sure if this would work, but possibly having two separate instances of dovecot with separate configs running may work for you. http://wiki2.dovecot.org/RunningDovecot On 08/07/15 11:04, Gerry wrote: > Situation: one front-facing server running Dovecot as IMAP/POP3/ > ManageSieve proxy, a mixture of IMAP servers (Dovecot, Exchange, ...) > in the back-end. Dovecot's passdb does lookups against MySQL which > contains a simple user/host mapping, the actual authentication happens > on the back-end IMAP servers. The configuration is more or less as > described here: http://wiki2.dovecot.org/PasswordDatabase/ExtraFields/Proxy > > Now I would like to add a Postfix instance on the front-facing server > which listens on the submission port and authenticates users via SASL > using the local Dovecot's UNIX socket. The idea being that a user only > needs to remember one single hostname, one username and one password > for all mail-related services. > > The problem is that Dovecot is operating in proxy mode, which means > that the password_query returns NULL as the password and explicitly > returns a field "nopasswd" containing "Y". Thus, users can not > authenticate against the UNIX socket. > > What I think I want to do is convince Dovecot to use one passdb for the > imap/pop3/managesieve services and different one for the "auth" service. > > The configuration snippet below doesn't work, but it should illustrate > what I want to achieve: > >> protocols = imap pop3 sieve >> >> service auth { >> passdb sql { >> driver = sql >> args = /etc/dovecot/mysql-auth-sasl.conf.ext >> } >> >> unix_listener /var/spool/postfix/private/auth { >> user = postfix >> group = postfix >> mode = 0666 >> } >> } >> >> # IMAP/POP3/ManageSieve auth against MySQL >> passdb sql { >> driver = sql >> args = /etc/dovecot/mysql-auth-default.conf.ext >> } > Example mysql-auth-sasl.conf.ext > >> driver = mysql >> connect = host=127.0.0.1 dbname=mail user=mail password=somethingrandom >> password_query = SELECT password AS password FROM users WHERE login = '%u' > Example mysql-auth-default.conf.ext: > >> driver = mysql >> connect = host=127.0.0.1 dbname=mail user=mail password=somethingrandom >> password_query = SELECT NULL AS password, 'Y' as nopassword, host, 'Y' AS proxy FROM users WHERE login = '%u' > Any pointers? > > Gerry From superinterstellar at gmail.com Sat Aug 8 04:56:55 2015 From: superinterstellar at gmail.com (Kevin Laurie) Date: Sat, 8 Aug 2015 10:26:55 +0530 Subject: backing up email / saving maildir on external hard drives Message-ID: Hello, Yesterday I tried to back up a 40GB maildir . I tried to move the maildir from home to external HDD but failed. Decided then to compress it(which took several hours). Now changing the disk format from FAT to exFAT to allow the transfer for the large compressed file. How does one back up emails on a external drive? Some advice would be greatly appreciated. From ml+dovecot at valo.at Sat Aug 8 05:06:59 2015 From: ml+dovecot at valo.at (Christian Kivalo) Date: Sat, 08 Aug 2015 07:06:59 +0200 Subject: backing up email / saving maildir on external hard drives In-Reply-To: References: Message-ID: <4CE2F84F-7995-41B6-B7A8-852FBCB69072@valo.at> Am 08. August 2015 06:56:55 MESZ, schrieb Kevin Laurie : >Hello, >Yesterday I tried to back up a 40GB maildir . >I tried to move the maildir from home to external HDD but failed. >Decided >then to compress it(which took several hours). Now changing the disk >format >from FAT to exFAT to allow the transfer for the large compressed file. > >How does one back up emails on a external drive? >Some advice would be greatly appreciated. i backup my maildir with rsync to my home nas. i assume you are using linux or similar, so when backing up to an external hdd i'd also use rsync to copy over the maildir as is to the external hdd but use a different filesystem n that hdd like ext4 or xfs or anything else than fat/vfat/exfat. when something goes wrong you can just copy the files back. what was the error message you got when you tried to copy to the ext. hdd? regards - c From superinterstellar at gmail.com Sat Aug 8 16:15:39 2015 From: superinterstellar at gmail.com (Kevin Laurie) Date: Sat, 8 Aug 2015 21:45:39 +0530 Subject: backing up email / saving maildir on external hard drives In-Reply-To: <4CE2F84F-7995-41B6-B7A8-852FBCB69072@valo.at> References: <4CE2F84F-7995-41B6-B7A8-852FBCB69072@valo.at> Message-ID: Dear Christian, Thanks for your feedback. The HDD will not accept larger than 4GB (as its in FAT format). Its a new external HDD. Thinking of the best format(that would work with Mac , Win and Linux) .seems like a challenge. What's your view on NTFS? And why not exFAT? Thanks Kevin On Saturday, August 8, 2015, Christian Kivalo wrote: > > > Am 08. August 2015 06:56:55 MESZ, schrieb Kevin Laurie < > superinterstellar at gmail.com >: > >Hello, > >Yesterday I tried to back up a 40GB maildir . > >I tried to move the maildir from home to external HDD but failed. > >Decided > >then to compress it(which took several hours). Now changing the disk > >format > >from FAT to exFAT to allow the transfer for the large compressed file. > > > >How does one back up emails on a external drive? > >Some advice would be greatly appreciated. > > i backup my maildir with rsync to my home nas. > > i assume you are using linux or similar, so when backing up to an external > hdd i'd also use rsync to copy over the maildir as is to the external hdd > but use a different filesystem n that hdd like ext4 or xfs or anything else > than fat/vfat/exfat. > > when something goes wrong you can just copy the files back. > > what was the error message you got when you tried to copy to the ext. hdd? > > regards > - c > From dale.gallagher at gmail.com Sat Aug 8 17:55:18 2015 From: dale.gallagher at gmail.com (Dale Gallagher) Date: Sat, 8 Aug 2015 19:55:18 +0200 Subject: 2.2.18: Mailbox INBOX sync: mailbox_delete failed: INBOX can't be deleted. In-Reply-To: References: Message-ID: Hello An update: I managed to sync mailboxes between the 2 servers using imapsync, which worked perfectly. I suspect there's a bug lurking in the imapc code, because why would an external utility work flawlessly with 2 dovecot servers, but dovecot's own syncing client won't? Regards Dale On 6 August 2015 at 12:10, Dale Gallagher wrote: > Hello > > Could my previously reported error above have something to do with the > IMAP path prefix? It's not configured on either of the previously mentioned > servers. > > I see a similar error, which may be related, when trying to sync using > offlineimap.py: > > Folder INBOX [acc: test_account]: > ERROR: ERROR in syncfolder for test_account folder INBOX > > Thanks > Dale > > > On 5 August 2015 at 16:59, Dale Gallagher > wrote: > >> Hello >> >> I have 2.2.18 built with: >> >> ./configure \ >> --prefix="$prefix" \ >> --bindir="$prefix/bin" \ >> --sbindir="$prefix/bin" \ >> --sysconfdir="/etc" \ >> --with-storages="maildir,imapc,pop3c" \ >> --without-vpopmail \ >> --without-ldap >> >> I'm attempting to sync (backup) from another Dovecot server running >> version 2.0.21: >> >> Both servers don't have no explicit namespaces configured and both are >> using the Maildir storage plugin. >> >> doveadm -Dv -o imapc_user=remote_user -o imapc_host=source.host -o >> imapc_port=993 -o imapc_ssl=imaps -o imapc_password=remote_pass -o >> imapc_features="rfc822.size fetch-headers" backup -R -u local_user imapc: >> >> All seems to go well, showing local and remote mailbox syncing without >> errors, then finally: >> >> dsync(mcg): Debug: brain M: Deleting mailbox 'INBOX' (GUID >> 59023c2a911dc255842b0000d09efc50): UIDVALIDITY changed (1355416734 -> >> 1438784913) >> dsync(mcg): Error: Mailbox INBOX sync: mailbox_delete failed: INBOX can't >> be deleted. >> >> No mail is synced to the local server (2.2.18) after this error halts >> execution. >> >> What should I be looking for, in terms of possible causes of this error? >> >> I've tried syncing on a completely fresh mailbox, which hasn't been >> accessed by a mail client and the same error occurs. >> >> I've seen similar reports of this error before, without clear solutions / >> suggestions. >> >> Help would be greatly appreciated. >> >> Thanks >> >> >> >> > From superinterstellar at gmail.com Sat Aug 8 18:17:20 2015 From: superinterstellar at gmail.com (Kevin Laurie) Date: Sat, 8 Aug 2015 23:47:20 +0530 Subject: Trying to back Message-ID: Hi, I am trying to backup multiple mail accounts by using imapsync and Dovecot on my local machine. I am using maildir as a mail storage format and currently have the following setting on my local machine :- mail_location = maildir:~/Maildir I am afraid that if I imapsync multiple accounts they will all get stored in the same location. Is there a way to seperate the accounts based on the email address? Appreciate if someone could advise. Thanks! Kevin From me at junc.eu Sat Aug 8 18:38:45 2015 From: me at junc.eu (Benny Pedersen) Date: Sat, 08 Aug 2015 20:38:45 +0200 Subject: Trying to back In-Reply-To: References: Message-ID: <9c4b24caa4b901b449eb3218f97e0a92@junc.eu> Kevin Laurie skrev den 2015-08-08 20:17: > Appreciate if someone could advise. http://imapsync.lamiral.info/ or try google imapsync howto but do read man pages first From eduardo at freedominterface.org Sat Aug 8 23:26:16 2015 From: eduardo at freedominterface.org (Eduardo Ramos) Date: Sat, 8 Aug 2015 20:26:16 -0300 Subject: Trying to back In-Reply-To: <9c4b24caa4b901b449eb3218f97e0a92@junc.eu> References: <9c4b24caa4b901b449eb3218f97e0a92@junc.eu> Message-ID: <55C69017.3050609@freedominterface.org> Kevin, Why don't you use dsync? http://wiki2.dovecot.org/Tools/Doveadm/Sync On 08/08/2015 03:38 PM, Benny Pedersen wrote: > Kevin Laurie skrev den 2015-08-08 20:17: > >> Appreciate if someone could advise. > > http://imapsync.lamiral.info/ > > or try google imapsync howto > but do read man pages first From superinterstellar at gmail.com Sun Aug 9 04:29:52 2015 From: superinterstellar at gmail.com (Kevin Laurie) Date: Sun, 9 Aug 2015 09:59:52 +0530 Subject: Trying to back In-Reply-To: <55C69017.3050609@freedominterface.org> References: <9c4b24caa4b901b449eb3218f97e0a92@junc.eu> <55C69017.3050609@freedominterface.org> Message-ID: Hi Eduardo, Thank you! Best Regards Kevin On Sunday, August 9, 2015, Eduardo Ramos wrote: > Kevin, > > Why don't you use dsync? > > http://wiki2.dovecot.org/Tools/Doveadm/Sync > > On 08/08/2015 03:38 PM, Benny Pedersen wrote: > >> Kevin Laurie skrev den 2015-08-08 20:17: >> >> Appreciate if someone could advise. >>> >> >> http://imapsync.lamiral.info/ >> >> or try google imapsync howto >> but do read man pages first >> > From fzielcke at z-51.de Sun Aug 9 06:46:02 2015 From: fzielcke at z-51.de (Felix Zielcke) Date: Sun, 09 Aug 2015 08:46:02 +0200 Subject: backing up email / saving maildir on external hard drives In-Reply-To: References: <4CE2F84F-7995-41B6-B7A8-852FBCB69072@valo.at> Message-ID: <1439102762.4408.5.camel@z-51.de> Am Samstag, den 08.08.2015, 21:45 +0530 schrieb Kevin Laurie: > Dear Christian, > Thanks for your feedback. > The HDD will not accept larger than 4GB (as its in FAT format). Its a > new > external HDD. Thinking of the best format(that would work with Mac , > Win > and Linux) .seems like a challenge. > What's your view on NTFS? And why not exFAT? > > Thanks > Kevin > Hi Kevin, you can have multiple partitions even on an external HDD. Or do you really need one big partition which works on all 3 OS? For Win and OS X I'd use exFAT. OS X AFAIK doestn't have write support on NTFS enabled by default. And then for Linux a native FS like ext4,btrfs,XFS or whatever you personally prefer. Though the best would be indeed 3 partitions for each OS. NTFS for Windows And then HFS+ for OS X For example I use on my backup HDD NTFS for Windows and btrfs for my Linux Systems From ml at x-net.be Sun Aug 9 09:09:20 2015 From: ml at x-net.be (Gerry) Date: Sun, 9 Aug 2015 11:09:20 +0200 Subject: Using a separate passdb per service In-Reply-To: <55C57E3F.6050402@pettijohn-web.com> References: <55C4D722.3030204@x-net.be> <55C57E3F.6050402@pettijohn-web.com> Message-ID: <55C718C0.6060502@x-net.be> On 08/08/2015 05:57 AM, Edgar Pettijohn wrote: > I'm not sure if this would work, but possibly having two separate > instances of dovecot with separate configs running may work for you. > > http://wiki2.dovecot.org/RunningDovecot Hi Edgar, Thank you for your suggestion. Yes, that would probably work, but it would be rather fiddly to run two Dovecot instances. I was hoping to be able to do it with just one instance. Gerry From superinterstellar at gmail.com Sun Aug 9 14:10:32 2015 From: superinterstellar at gmail.com (Kevin Laurie) Date: Sun, 9 Aug 2015 22:10:32 +0800 Subject: backing up email / saving maildir on external hard drives In-Reply-To: <1439102762.4408.5.camel@z-51.de> References: <4CE2F84F-7995-41B6-B7A8-852FBCB69072@valo.at> <1439102762.4408.5.camel@z-51.de> Message-ID: Hi Felix, I would prefer having one HDD that works with all three OSes. That would be very convenient. Guess exFAT is my best bet. Its supported on Linux,Win and OS X. What do you reckon? Best Regards Kevin On Sun, Aug 9, 2015 at 2:46 PM, Felix Zielcke wrote: > Am Samstag, den 08.08.2015, 21:45 +0530 schrieb Kevin Laurie: >> Dear Christian, >> Thanks for your feedback. >> The HDD will not accept larger than 4GB (as its in FAT format). Its a >> new >> external HDD. Thinking of the best format(that would work with Mac , >> Win >> and Linux) .seems like a challenge. >> What's your view on NTFS? And why not exFAT? >> >> Thanks >> Kevin >> > > Hi Kevin, > > you can have multiple partitions even on an external HDD. > Or do you really need one big partition which works on all 3 OS? > > For Win and OS X I'd use exFAT. > OS X AFAIK doestn't have write support on NTFS enabled by default. > And then for Linux a native FS like ext4,btrfs,XFS or whatever you > personally prefer. > > Though the best would be indeed 3 partitions for each OS. > NTFS for Windows > And then HFS+ for OS X > > For example I use on my backup HDD NTFS for Windows and btrfs for my > Linux Systems From phoffman at proper.com Sun Aug 9 15:40:30 2015 From: phoffman at proper.com (Paul Hoffman) Date: Sun, 09 Aug 2015 08:40:30 -0700 Subject: How to import an mbox to an existing user as a new mailbox In-Reply-To: References: <97F9738A-901F-4E39-966B-6B092C2B5BA2@proper.com> Message-ID: On 3 Aug 2015, at 7:16, Steffen Kaiser wrote: > On Mon, 3 Aug 2015, Paul Hoffman wrote: >> On 2 Aug 2015, at 23:27, Steffen Kaiser wrote: >>> On Sun, 2 Aug 2015, Paul Hoffman wrote: >>> >>>> Greetings. I have a bunch of mbox files that I want to import to an >>>> existing user, each into a new mailbox. That is, I want to import >>>> the file project1.mbox to user Foo, into a newly-created mailbox >>>> called "Project 1". Is there a dovecot tool to do this on the >>>> server? >>> >>> do you have access to the file system of the mail storage? Can you >>> call doveadm on the server? What mail storage do you use now, >>> Maildir, mbox, ... ? >> >> Sorry for being unclear earlier. Yes, yes, mbox. > > copy project1.mbox to '/path/to/user/mail/storage/Project 1.mbox' and > make sure, the user has file system rights to access the file at all. > > Maybe, you need to add the new mailbox to the user's subscription > file, in order to have it show up. After a week, I'm at this again. Now having read up on doveadm, I am trying the following: sudo doveadm import -u Foo mbox:/home/phoffman/project1.mbox "Project 1" all However, that gets the result: Initializing mail storage from mail_location parameter failed: mbox: mbox root directory can't be a file Which bit am I missing here? --Paul Hoffman From slitt at troubleshooters.com Sun Aug 9 15:44:49 2015 From: slitt at troubleshooters.com (Steve Litt) Date: Sun, 9 Aug 2015 11:44:49 -0400 Subject: backing up email / saving maildir on external hard drives In-Reply-To: References: Message-ID: <20150809114449.7b5a273f@mydesq2.domain.cxm> On Sat, 8 Aug 2015 10:26:55 +0530 Kevin Laurie wrote: > Hello, > Yesterday I tried to back up a 40GB maildir . > I tried to move the maildir from home to external HDD but failed. If you tried to *move* it it's an archive, not a backup. If you tried to *copy* it, with the intent of keeping the original on the original hard disk and using it further, and keeping today's copy on some other media, *that's* a backup. I'm not trying to be pedantic, but there are many distinctions between the two. Archives must be re-transferred frequently: Backups merely need to be redone at intervals. > Decided then to compress it(which took several hours). Now changing > the disk format from FAT to exFAT to allow the transfer for the large > compressed file. Reading https://en.wikipedia.org/wiki/File_Allocation_Table#exFAT , I personally wouldn't use exFAT. Regular FAT32 has a max filesize of 2GB-1, which is 50 times the size of your whole uncompressed maildir. > > How does one back up emails on a external drive? > Some advice would be greatly appreciated. Check this out: ================================================ slitt at mydesq2:~$ df -h ~/mail/Maildir Filesystem Size Used Avail Use% Mounted on /dev/sdb8 116G 11G 100G 10% /home/slitt/mail/Maildir slitt at mydesq2:~$ ================================================ I don't have 40 GB, but * have 11, which is less than an order of magnitude away. I just back up this puppy to my backup server with my normal rsync based backup procedures, which you can read about here: * http://www.troubleshooters.com/lpm/200609/200609.htm * http://www.troubleshooters.com/linux/blu-ray-backup.htm * http://www.troubleshooters.com/lpm/201408/201408.htm The stuff about Blu-Ray is important only if you back up to blu-ray. I like to keep some backups on write-once media, because kept in the shade at reasonable temperatures and humidities, it tends to last longer. And spinning disks that spend the majority of their time not spinning tend to have problems. If this is a *backup*, I'd leave it uncompressed so you can take incremental backups regularly. If it's an *archive*, meaning that the data is immediately removed from your computer after copy, compression might be in order, but you should make two copies and test them both thoroughly before deleting the original, and you should test them every couple months and if either goes bad, copy the other one to something good. Archives are a PITA. For 40GB in these days of $150 2TB drives, I'd keep the data intact, back it up, and when you outgrow your hard drive, just get a bigger one. In other parts of this thread you ask how to separate backups from different accounts from different computers. As far as accounts, I think that Maildir directory structures would take care of that. As far as different machines, just put the hostname at the front of each destination directory. SteveT Steve Litt July 2015 featured book: Rapid Learning for the 21st Century http://www.troubleshooters.com/rl21 From me at junc.eu Sun Aug 9 15:46:27 2015 From: me at junc.eu (Benny Pedersen) Date: Sun, 09 Aug 2015 17:46:27 +0200 Subject: How to import an mbox to an existing user as a new mailbox In-Reply-To: References: <97F9738A-901F-4E39-966B-6B092C2B5BA2@proper.com> Message-ID: <92803d84546f5f40419a660a6c0c9913@junc.eu> Paul Hoffman skrev den 2015-08-09 17:40: > Initializing mail storage from mail_location parameter failed: mbox: > mbox root directory can't be a file > > Which bit am I missing here? sudo doveadm import -u Foo mbox:/home/phoffman/project1/ "Project 1" all untested From superinterstellar at gmail.com Sun Aug 9 16:33:14 2015 From: superinterstellar at gmail.com (Kevin Laurie) Date: Mon, 10 Aug 2015 00:33:14 +0800 Subject: backing up email / saving maildir on external hard drives In-Reply-To: <20150809114449.7b5a273f@mydesq2.domain.cxm> References: <20150809114449.7b5a273f@mydesq2.domain.cxm> Message-ID: Dear Steve, Very valuable info. Appreciate it and will be careful when using terms. Actually I think I should just use rsync without compressing. The reason why I started compressing was because the GUI gave some errors when I was trying to copy then files. I'll just rsync the data from my laptop HDD to my external drive(without compressing) Thanks Kevin On Sunday, August 9, 2015, Steve Litt wrote: > On Sat, 8 Aug 2015 10:26:55 +0530 > Kevin Laurie > wrote: > > > Hello, > > Yesterday I tried to back up a 40GB maildir . > > I tried to move the maildir from home to external HDD but failed. > > If you tried to *move* it it's an archive, not a backup. If you tried > to *copy* it, with the intent of keeping the original on the original > hard disk and using it further, and keeping today's copy on some other > media, *that's* a backup. I'm not trying to be pedantic, but there are > many distinctions between the two. Archives must be re-transferred > frequently: Backups merely need to be redone at intervals. > > > Decided then to compress it(which took several hours). Now changing > > the disk format from FAT to exFAT to allow the transfer for the large > > compressed file. > > Reading https://en.wikipedia.org/wiki/File_Allocation_Table#exFAT , I > personally wouldn't use exFAT. Regular FAT32 has a max filesize of > 2GB-1, which is 50 times the size of your whole uncompressed maildir. > > > > > How does one back up emails on a external drive? > > Some advice would be greatly appreciated. > > Check this out: > > ================================================ > slitt at mydesq2:~$ df -h ~/mail/Maildir > Filesystem Size Used Avail Use% Mounted on > /dev/sdb8 116G 11G 100G 10% /home/slitt/mail/Maildir > slitt at mydesq2:~$ > ================================================ > > I don't have 40 GB, but * have 11, which is less than an order of > magnitude away. I just back up this puppy to my backup server with my > normal rsync based backup procedures, which you can read about here: > > * http://www.troubleshooters.com/lpm/200609/200609.htm > > * http://www.troubleshooters.com/linux/blu-ray-backup.htm > > * http://www.troubleshooters.com/lpm/201408/201408.htm > > The stuff about Blu-Ray is important only if you back up to blu-ray. I > like to keep some backups on write-once media, because kept in the > shade at reasonable temperatures and humidities, it tends to last > longer. And spinning disks that spend the majority of their time not > spinning tend to have problems. > > If this is a *backup*, I'd leave it uncompressed so you can take > incremental backups regularly. If it's an *archive*, meaning that the > data is immediately removed from your computer after copy, compression > might be in order, but you should make two copies and test them both > thoroughly before deleting the original, and you should test them every > couple months and if either goes bad, copy the other one to something > good. Archives are a PITA. For 40GB in these days of $150 2TB drives, > I'd keep the data intact, back it up, and when you outgrow your hard > drive, just get a bigger one. > > In other parts of this thread you ask how to separate backups from > different accounts from different computers. As far as accounts, I > think that Maildir directory structures would take care of that. As far > as different machines, just put the hostname at the front of each > destination directory. > > > > SteveT > > Steve Litt > July 2015 featured book: Rapid Learning for the 21st Century > http://www.troubleshooters.com/rl21 > From phoffman at proper.com Sun Aug 9 18:20:57 2015 From: phoffman at proper.com (Paul Hoffman) Date: Sun, 09 Aug 2015 11:20:57 -0700 Subject: How to import an mbox to an existing user as a new mailbox In-Reply-To: <92803d84546f5f40419a660a6c0c9913@junc.eu> References: <97F9738A-901F-4E39-966B-6B092C2B5BA2@proper.com> <92803d84546f5f40419a660a6c0c9913@junc.eu> Message-ID: On 9 Aug 2015, at 8:46, Benny Pedersen wrote: > Paul Hoffman skrev den 2015-08-09 17:40: > >> Initializing mail storage from mail_location parameter failed: mbox: >> mbox root directory can't be a file >> >> Which bit am I missing here? > > sudo doveadm import -u Foo mbox:/home/phoffman/project1/ "Project 1" > all > > untested When I try this: Fatal: Import namespace initialization failed: Initializing mail storage from mail_location parameter failed: mkdir(/home/phoffman/project1) failed: Permission denied (euid=1013(phoffprop) egid=1013(phoffprop) missing +w perm: /home/phoffman, dir owned by 1001:1001 mode=0755) Does this error indicate that it is trying to write to the mbox file instead of reading it? If so, why? --Paul Hoffman From me at junc.eu Sun Aug 9 18:36:41 2015 From: me at junc.eu (Benny Pedersen) Date: Sun, 09 Aug 2015 20:36:41 +0200 Subject: How to import an mbox to an existing user as a new mailbox In-Reply-To: References: <97F9738A-901F-4E39-966B-6B092C2B5BA2@proper.com> <92803d84546f5f40419a660a6c0c9913@junc.eu> Message-ID: Paul Hoffman skrev den 2015-08-09 20:20: > Fatal: Import namespace initialization failed: Initializing mail > storage from mail_location parameter failed: > mkdir(/home/phoffman/project1) failed: Permission denied > (euid=1013(phoffprop) egid=1013(phoffprop) missing +w perm: > /home/phoffman, dir owned by 1001:1001 mode=0755) > > Does this error indicate that it is trying to write to the mbox file > instead of reading it? If so, why? nope, you have to tell dovecot auth backend more how to write to homedir of 'id proffprop' uid and gid does not match as i read it, proffprop have uid 1001 and dovecot tryed to write to 1013 search that error and solve it From edgar at pettijohn-web.com Sun Aug 9 18:58:47 2015 From: edgar at pettijohn-web.com (Edgar Pettijohn) Date: Sun, 9 Aug 2015 13:58:47 -0500 Subject: How to import an mbox to an existing user as a new mailbox In-Reply-To: References: <97F9738A-901F-4E39-966B-6B092C2B5BA2@proper.com> <92803d84546f5f40419a660a6c0c9913@junc.eu> Message-ID: <55C7A2E7.6030305@pettijohn-web.com> On 08/09/15 13:36, Benny Pedersen wrote: > Paul Hoffman skrev den 2015-08-09 20:20: > >> Fatal: Import namespace initialization failed: Initializing mail >> storage from mail_location parameter failed: >> mkdir(/home/phoffman/project1) failed: Permission denied >> (euid=1013(phoffprop) egid=1013(phoffprop) missing +w perm: >> /home/phoffman, dir owned by 1001:1001 mode=0755) >> >> Does this error indicate that it is trying to write to the mbox file >> instead of reading it? If so, why? > > nope, you have to tell dovecot auth backend more how to write to > homedir of 'id proffprop' > > uid and gid does not match > > as i read it, proffprop have uid 1001 and dovecot tryed to write to 1013 > > search that error and solve it There are prob. a lot of ways around this, but the easiest would most likely be something like: # chmod o+w /home/phoffman From me at junc.eu Sun Aug 9 19:13:37 2015 From: me at junc.eu (Benny Pedersen) Date: Sun, 09 Aug 2015 21:13:37 +0200 Subject: How to import an mbox to an existing user as a new mailbox In-Reply-To: <55C7A2E7.6030305@pettijohn-web.com> References: <97F9738A-901F-4E39-966B-6B092C2B5BA2@proper.com> <92803d84546f5f40419a660a6c0c9913@junc.eu> <55C7A2E7.6030305@pettijohn-web.com> Message-ID: <4203326cde80c2e90c3a74ee7ce891bb@junc.eu> Edgar Pettijohn skrev den 2015-08-09 20:58: >> search that error and solve it > There are prob. a lot of ways around this, but the easiest would most > likely be something like: > # chmod o+w /home/phoffman why use dovecot ? that will make it totaly insecure, dont do it ! i have here no access on groups and others not even on READS post dovecot -n and lets help solve it From edgar at pettijohn-web.com Sun Aug 9 19:43:31 2015 From: edgar at pettijohn-web.com (Edgar Pettijohn) Date: Sun, 9 Aug 2015 14:43:31 -0500 Subject: How to import an mbox to an existing user as a new mailbox In-Reply-To: <4203326cde80c2e90c3a74ee7ce891bb@junc.eu> References: <97F9738A-901F-4E39-966B-6B092C2B5BA2@proper.com> <92803d84546f5f40419a660a6c0c9913@junc.eu> <55C7A2E7.6030305@pettijohn-web.com> <4203326cde80c2e90c3a74ee7ce891bb@junc.eu> Message-ID: <55C7AD63.5080201@pettijohn-web.com> On 08/09/15 14:13, Benny Pedersen wrote: > Edgar Pettijohn skrev den 2015-08-09 20:58: > >>> search that error and solve it >> There are prob. a lot of ways around this, but the easiest would most >> likely be something like: >> # chmod o+w /home/phoffman > > why use dovecot ? > > that will make it totaly insecure, dont do it ! > afterwards #chmod o-w /home/phoffman > i have here no access on groups and others not even on READS > > post dovecot -n > > and lets help solve it From skdovecot at smail.inf.fh-brs.de Mon Aug 10 07:58:40 2015 From: skdovecot at smail.inf.fh-brs.de (Steffen Kaiser) Date: Mon, 10 Aug 2015 09:58:40 +0200 (CEST) Subject: Using a separate passdb per service In-Reply-To: <55C4D722.3030204@x-net.be> References: <55C4D722.3030204@x-net.be> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, 7 Aug 2015, Gerry wrote: > The problem is that Dovecot is operating in proxy mode, which means > that the password_query returns NULL as the password and explicitly > returns a field "nopasswd" containing "Y". Thus, users can not > authenticate against the UNIX socket. > What I think I want to do is convince Dovecot to use one passdb for the > imap/pop3/managesieve services and different one for the "auth" service. As far as I know, all services use the "auth" in the back. But you have the "%s" / service variable. You should be able to craft a SQL query, that returns NULL & nopasswd=Y, if postfix is not querying Dovecot. I don't know, which service name postfix passes to Dovecot, though, - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVchZsHz1H7kL/d9rAQIaLwf/WXnI9PMGuN042g7VzfDlZxbsVTvck55X DqPdy1P+YDtMCFpEbTxQG8m9EMfI82Zcd3rzqGbcaMbFqatG7TsucBg06S5j7XSX fs/jNX6DwAdmNIRzjrEU5a8M+Zpo6ifWohBO1IMax3sAs4Z9v+O/hEjX1wiKed24 nFA1QNFG/s2bjDUbf7WBYnU0MnFPeUmMJzy5sR+zFC9lWbaj+Y9b6ayMbdlhVvcz 8qu827/i+2McHlDiS3a2JmwuYTyGpPwfryIojjgTnYvxB1Ex4qsI/mfk8s2am6hY SIvi0Btdlb/ZUmxMy8WKj/hko4Mb+nxO6FBpMU8V8opTJHHUuLf0UA== =1Wip -----END PGP SIGNATURE----- From ml at x-net.be Mon Aug 10 09:23:41 2015 From: ml at x-net.be (Gerry) Date: Mon, 10 Aug 2015 11:23:41 +0200 Subject: Using a separate passdb per service In-Reply-To: References: <55C4D722.3030204@x-net.be> Message-ID: <55C86D9D.5000603@x-net.be> On 08/10/2015 09:58 AM, Steffen Kaiser wrote: > As far as I know, all services use the "auth" in the back. > > But you have the "%s" / service variable. You should be able to craft a > SQL query, that returns NULL & nopasswd=Y, if postfix is not querying > Dovecot. > > I don't know, which service name postfix passes to Dovecot, though, Hmm, that's an interesting idea. I'll explore it further. Thanks! Gerry From matthias.lay at securepoint.de Mon Aug 10 11:16:12 2015 From: matthias.lay at securepoint.de (matthias lay) Date: Mon, 10 Aug 2015 13:16:12 +0200 Subject: pigeonhole/lda accessing -m folder In-Reply-To: <55C4BAB3.4050507@securepoint.de> References: <55C4862A.7070101@securepoint.de> <20150807102626.GG5380@ruhr-uni-bochum.de> <55C48B0D.5040200@securepoint.de> <55C4BAB3.4050507@securepoint.de> Message-ID: <55C887FC.7020002@securepoint.de> complete patch. some parts were missing before diff --git a/src/lib-sieve/plugins/environment/ext-environment-common.c b/src/lib-sieve/plugins/environment/ext-environment-common.c --- a/src/lib-sieve/plugins/environment/ext-environment-common.c +++ b/src/lib-sieve/plugins/environment/ext-environment-common.c @@ -24,7 +24,8 @@ static const struct sieve_environment_item *core_env_items[] = { &location_env_item, &phase_env_item, &name_env_item, - &version_env_item + &version_env_item, + &defaultfolder_env_item }; static unsigned int core_env_items_count = N_ELEMENTS(core_env_items); @@ -233,7 +234,19 @@ const struct sieve_environment_item version_env_item = { .value = PIGEONHOLE_VERSION, }; +/* "defaultfolder": + * the default folder where mail is stored when no rule matches + */ +static const char *envit_defaultfolder_get_value +(struct sieve_instance *svinst, + const struct sieve_script_env *senv) +{ + return senv->default_mailbox; +} - +const struct sieve_environment_item defaultfolder_env_item = { + .name = "defaultfolder", + .get_value = envit_defaultfolder_get_value, +}; diff --git a/src/lib-sieve/plugins/environment/ext-environment-common.h b/src/lib-sieve/plugins/environment/ext-environment-common.h --- a/src/lib-sieve/plugins/environment/ext-environment-common.h +++ b/src/lib-sieve/plugins/environment/ext-environment-common.h @@ -38,6 +38,7 @@ extern const struct sieve_environment_item location_env_item; extern const struct sieve_environment_item phase_env_item; extern const struct sieve_environment_item name_env_item; extern const struct sieve_environment_item version_env_item; +extern const struct sieve_environment_item defaultfolder_env_item; /* * Initialization On 08/07/2015 04:03 PM, matthias lay wrote: > fixed my problem. > > if somebody is interested. > I added an variable to the environment extension, which is quite easy. > And the value of the folder is already there. so this is all: > > > src/lib-sieve/plugins/environment/ext-environment-common.c > > static const char *envit_spfolder_get_value > (struct sieve_instance *svinst, > const struct sieve_script_env *senv) > { > return senv->default_mailbox; > } > > const struct sieve_environment_item spfolder_env_item = { > .name = "spfolder", > .get_value = envit_spfolder_get_value, > }; > > > after that the variable can easily be used in a script like > > if anyof ( environment :matches "spfolder" "*") { > set "myfolder" "${1}"; > } > > > > > > > > > > On 08/07/2015 12:40 PM, matthias lay wrote: >> hi jost thx for the reply, >> >> by access I mean to read the variable >> >> >> require ["fileinto", "variables", "?destfolder?" ]; >> >> if anyof ( destfolder :matches "*") { >> >> fileinto "${1}/subfolder"; >> >> }else{ >> >> fileinto "INBOX/subfolder"; >> >> } >> >> >> >> On 08/07/2015 12:26 PM, Jost Krieger wrote: >>> On Fri Aug 7 12:19:22 2015, matthias lay wrote: >>> >>>> anybody knows, if there is a way to acces the folder from a call like this >>>> >>>> deovecot-lda -m destfolder >>>> >>>> from within a sieve script? >>> >>> Depending on what you mean by "access", this could be as easy as >>> >>> require "fileinto"; >>> >>> fileinto "destfolder"; >>> >>> Yours >>> Jost Krieger >>> From delrio at mie.utoronto.ca Mon Aug 10 13:35:57 2015 From: delrio at mie.utoronto.ca (Oscar del Rio) Date: Mon, 10 Aug 2015 09:35:57 -0400 Subject: SiS does not work In-Reply-To: <55C341B9.9080203@granch.ru> References: <55C341B9.9080203@granch.ru> Message-ID: <55C8A8BD.9090204@mie.utoronto.ca> On 08/ 6/15 07:15 AM, ?????? ????? ?????????????? wrote: > > I have tried to use SiS to deduplicate attachments in mail storage... > > Aug 6 16:11:12 mailsrv dovecot: lmtp(yyyy): Error: > read(/usr/share/smbusers/xxxxx/mail/attachments You forgot to post your "doveconf -n" configuration. It looks like your attachment storage includes the username in the path. That won't work. The path should be global, e.g. /usr/share/mail/attachments, and readable/writable by the dovecot process. SiS deduplicates the attachments regardless of username since the same attachment might be sent/received by multiple users. SiS will save a single copy in the global path and transparently link it to the users' mailboxes. From skdovecot at smail.inf.fh-brs.de Mon Aug 10 14:29:01 2015 From: skdovecot at smail.inf.fh-brs.de (Steffen Kaiser) Date: Mon, 10 Aug 2015 16:29:01 +0200 (CEST) Subject: PATCH for Stephan: Re: pigeonhole/lda accessing -m folder In-Reply-To: <55C887FC.7020002@securepoint.de> References: <55C4862A.7070101@securepoint.de> <20150807102626.GG5380@ruhr-uni-bochum.de> <55C48B0D.5040200@securepoint.de> <55C4BAB3.4050507@securepoint.de> <55C887FC.7020002@securepoint.de> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 10 Aug 2015, matthias lay wrote: Hi Matthias, there is another possibility: require ["fileinto", "variables", "subaddress", "envelope"]; if envelope :detail :matches "to" "*" { set "myfolder" "${1}"; } else { set "myfolder" "INBOX"; } fileinto "my_${myfolder}"; stop; /usr/local/dovecot/libexec/dovecot/deliver -d user -m box -a user+box And enable the recipient_delimiter option. ===== Dear Stephan, If you find this code OK, please merge it into Pigeonhole. I find this information useful, too. The +detail and the default mailbox are not necessarily the same. Maybe, there is a namespace to expose this information easily? For the archive: + This works for both LDA and LMTP. + The Dovecot LDA requires option -m . + LMTP requires an active recipient_delimiter option. + If there is no -m option or no +detail, INBOX is the defaultfolder. > complete patch. some parts were missing before > > > diff --git a/src/lib-sieve/plugins/environment/ext-environment-common.c > b/src/lib-sieve/plugins/environment/ext-environment-common.c > --- a/src/lib-sieve/plugins/environment/ext-environment-common.c > +++ b/src/lib-sieve/plugins/environment/ext-environment-common.c > @@ -24,7 +24,8 @@ static const struct sieve_environment_item > *core_env_items[] = { > &location_env_item, > &phase_env_item, > &name_env_item, > - &version_env_item > + &version_env_item, > + &defaultfolder_env_item > }; > > static unsigned int core_env_items_count = N_ELEMENTS(core_env_items); > @@ -233,7 +234,19 @@ const struct sieve_environment_item > version_env_item = { > .value = PIGEONHOLE_VERSION, > }; > > +/* "defaultfolder": > + * the default folder where mail is stored when no rule matches > + */ > +static const char *envit_defaultfolder_get_value > +(struct sieve_instance *svinst, > + const struct sieve_script_env *senv) > +{ > + return senv->default_mailbox; > +} > > - > +const struct sieve_environment_item defaultfolder_env_item = { > + .name = "defaultfolder", > + .get_value = envit_defaultfolder_get_value, > +}; > > > diff --git a/src/lib-sieve/plugins/environment/ext-environment-common.h > b/src/lib-sieve/plugins/environment/ext-environment-common.h > --- a/src/lib-sieve/plugins/environment/ext-environment-common.h > +++ b/src/lib-sieve/plugins/environment/ext-environment-common.h > @@ -38,6 +38,7 @@ extern const struct sieve_environment_item > location_env_item; > extern const struct sieve_environment_item phase_env_item; > extern const struct sieve_environment_item name_env_item; > extern const struct sieve_environment_item version_env_item; > +extern const struct sieve_environment_item defaultfolder_env_item; > > /* > * Initialization > > > > On 08/07/2015 04:03 PM, matthias lay wrote: >> fixed my problem. >> >> if somebody is interested. >> I added an variable to the environment extension, which is quite easy. >> And the value of the folder is already there. so this is all: >> >> >> src/lib-sieve/plugins/environment/ext-environment-common.c >> >> static const char *envit_spfolder_get_value >> (struct sieve_instance *svinst, >> const struct sieve_script_env *senv) >> { >> return senv->default_mailbox; >> } >> >> const struct sieve_environment_item spfolder_env_item = { >> .name = "spfolder", >> .get_value = envit_spfolder_get_value, >> }; >> >> >> after that the variable can easily be used in a script like >> >> if anyof ( environment :matches "spfolder" "*") { >> set "myfolder" "${1}"; >> } >> >> >> On 08/07/2015 12:40 PM, matthias lay wrote: >>> hi jost thx for the reply, >>> >>> by access I mean to read the variable >>> >>> >>> require ["fileinto", "variables", "?destfolder?" ]; >>> >>> if anyof ( destfolder :matches "*") { >>> >>> fileinto "${1}/subfolder"; >>> >>> }else{ >>> >>> fileinto "INBOX/subfolder"; >>> >>> } >>> >>> >>> >>> On 08/07/2015 12:26 PM, Jost Krieger wrote: >>>> On Fri Aug 7 12:19:22 2015, matthias lay wrote: >>>> >>>>> anybody knows, if there is a way to acces the folder from a call like this >>>>> >>>>> deovecot-lda -m destfolder >>>>> >>>>> from within a sieve script? >>>> >>>> Depending on what you mean by "access", this could be as easy as >>>> >>>> require "fileinto"; >>>> >>>> fileinto "destfolder"; >>>> >>>> Yours >>>> Jost Krieger >>>> > - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVci1LXz1H7kL/d9rAQJJogf+PJl87IL3KEmc3uPDi7q8yJBRTChmqlGN 39SrJ6DGfe8+IZSlipJZzAAnE/nn+D/YHGcS20X40Nas4QassoZi16ZfX6OGId1O 8IvYjGc0k0Mu8Nnk0vyX0DwTA23oId0y934HCQxhxvgS6bM9iexA+Fs2KPA2ta+9 xOEtpjQCTvYFX6VvqhjQaz6lv/f+mSEoU/EFbvC9jMNV+v4a2SQ3Dazfg6OAoRUp WfXjYAgmvsH/xbL/QGqzFkJX/1vVTBX8wJRDoZHC/XcWNr2775VFQHaO4PQrPAr/ 48dHbiSwinGLBD4RkLVq7l9zZWfuxf6EMTmdNWHRO49cE3ZahxBtoQ== =Jq9R -----END PGP SIGNATURE----- From stephan at rename-it.nl Mon Aug 10 15:05:58 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Mon, 10 Aug 2015 17:05:58 +0200 Subject: PATCH for Stephan: Re: pigeonhole/lda accessing -m folder In-Reply-To: References: <55C4862A.7070101@securepoint.de> <20150807102626.GG5380@ruhr-uni-bochum.de> <55C48B0D.5040200@securepoint.de> <55C4BAB3.4050507@securepoint.de> <55C887FC.7020002@securepoint.de> Message-ID: <55C8BDD6.1060007@rename-it.nl> Op 8/10/2015 om 4:29 PM schreef Steffen Kaiser: > On Mon, 10 Aug 2015, matthias lay wrote: > > Dear Stephan, > > If you find this code OK, please merge it into Pigeonhole. Non-standard entries like this require a vnd.dovecot prefix. > I find this information useful, too. The +detail and the default > mailbox are not necessarily the same. Maybe, there is a namespace to > expose this information easily? The Pigeonhole internals support custom variable namespaces. The extdata plugin uses this too, so adding something like that is architecturally not problematic. I would make this dependent on an extension like "vnd.dovecot.environment", so that non-standard features are only added when this is in the require line. Currently, the "auth" envelope field (CMUSieve) is the only exception and I like to keep it that way. I'll give this a look. Also, are there any other environment items that may be useful? Regards, Stephan. From matthias.lay at securepoint.de Mon Aug 10 15:51:57 2015 From: matthias.lay at securepoint.de (matthias lay) Date: Mon, 10 Aug 2015 17:51:57 +0200 Subject: PATCH for Stephan: Re: pigeonhole/lda accessing -m folder In-Reply-To: <55C8BDD6.1060007@rename-it.nl> References: <55C4862A.7070101@securepoint.de> <20150807102626.GG5380@ruhr-uni-bochum.de> <55C48B0D.5040200@securepoint.de> <55C4BAB3.4050507@securepoint.de> <55C887FC.7020002@securepoint.de> <55C8BDD6.1060007@rename-it.nl> Message-ID: <55C8C89D.1040707@securepoint.de> hi stephan, On 08/10/2015 05:05 PM, Stephan Bosch wrote: > > I'll give this a look. Also, are there any other environment items that > may be useful? > when I checked the environment plugin I was wondering cuz it was not what I was thinking it would be, .....before reading the RFC ;) would it be a bad idea to have a generic environment extension working with getenv() to be able to get all kind of self defined, prefixed variables in program environment? Greetz Matze From foss-ml at c-14.de Mon Aug 10 16:19:57 2015 From: foss-ml at c-14.de (Simon Thelen) Date: Mon, 10 Aug 2015 18:19:57 +0200 Subject: Dovecot sends "OK Logged in" with empty tag In-Reply-To: <20150731203749.GB2027@anonymous> References: <20150731203749.GB2027@anonymous> Message-ID: <20150810161957.GA2262@anonymous> On 15-07-31 at 22:37, Simon Thelen wrote: > Hello, > > I'm currently writing my own IMAP client and testing it against Dovecot, > and when doing PLAIN AUTH Dovecot sends the "OK Logged in" with an empty > tag at the start. I tried reproducing the conversation with netcat, but > Dovecot sends a correctly tagged OK at the end of the AUTH so I assume > it's timing related. I ended up finding the issue and thought I'd report the fix here. When sending the CAPABILITY command directly after receiving the untagged OK during PREGREET, I sent a trailing '\0' after the '\r\n'. This then presumably messed up the tokenizer in dovecot causing it to think that the tag for the AUTHENTICATE command was "". I can't find a spot in the RFC that states what a server should do if it receives a bad tag, so I don't know if dovecot handles it correctly but in any case this was a bug on my part. Sorry for the noise. -- Simon Thelen From dovecot.org at veggiechinese.net Mon Aug 10 18:02:58 2015 From: dovecot.org at veggiechinese.net (Will Yardley) Date: Mon, 10 Aug 2015 11:02:58 -0700 Subject: userdb / user map with director Message-ID: <20150810180258.GV74304@aura.veggiechinese.net> This is with dovecot-ee 2.2.18.2 on RHEL6.... To handle backend and frontend on same machine, I'm using the following hack, as previously outlined here: [in director instance config] passdb { # See thread ending in: # http://dovecot.org/pipermail/dovecot/2012-June/083817.html # for why this has to be sql instead of 'static' driver = sql args = /etc/dovecot/dovecot-sql.conf } [in /etc/dovecot/dovecot-sql.conf] driver = sqlite connect = /etc/dovecot/empty.db password_query = select 'y' as proxy, \ NULL as password, \ 'y' as nopassword, \ case '%a' \ when '110' then '10110' \ when '995' then '10110' \ when '143' then '10143' \ when '993' then '10143' end \ as port; This works, but I was getting a failure on usernames, since director instance didn't have a userdb. So I tried the following: [in config] passdb { driver = sql args = /etc/dovecot/dovecot-sql.conf } userdb { driver = prefetch } and appended the following to the SQL query: password_query = select '%u' AS user, \ Then I see the following error in the log when I try to run the map "Trying to iterate users, but userdbs don't support it" and doveadm -i director director map returns: Error: User listing returned failure Error: user listing failed [...] I'm sure I'm missing something obvious, so is there an easy fix to make this mapping work? w From fzielcke at z-51.de Mon Aug 10 20:09:19 2015 From: fzielcke at z-51.de (Felix Zielcke) Date: Mon, 10 Aug 2015 22:09:19 +0200 Subject: backing up email / saving maildir on external hard drives In-Reply-To: References: <4CE2F84F-7995-41B6-B7A8-852FBCB69072@valo.at> <1439102762.4408.5.camel@z-51.de> Message-ID: <1439237359.3315.3.camel@z-51.de> Am Sonntag, den 09.08.2015, 22:10 +0800 schrieb Kevin Laurie: > Hi Felix, > > I would prefer having one HDD that works with all three OSes. That > would be very convenient. > Guess exFAT is my best bet. Its supported on Linux,Win and OS X. > What do you reckon? > Best Regards > Kevin > With one HDD you can still have at least 4 different partitions on it without problems. So the question is still if you need one big partition for all 3 OS or not. I don't know how well exFAT is supported in Linux. I never used it at all. From j.echter at echter-kuechen-elektro.de Tue Aug 11 07:35:04 2015 From: j.echter at echter-kuechen-elektro.de (J. Echter) Date: Tue, 11 Aug 2015 09:35:04 +0200 Subject: Convert from Maildir to Mdbox Message-ID: <55C9A5A8.9010501@echter-kuechen-elektro.de> Hi, i'd like to convert from maildir to mdbox. what would be the best way to do this? i have read the dsync part of the dovecot wiki but i don't get what i have to use for my setup. my dovecot.conf (the snippets i think would matter): mail_gid = 5000 mail_home = /DATEN/vmail/%d/%n mail_location = maildir:/DATEN/vmail/%d/%n/mail/ mail_plugins = " zlib" mail_uid = 5000 Thanks for your input. Juergen From goetz.reinicke at filmakademie.de Tue Aug 11 13:45:33 2015 From: goetz.reinicke at filmakademie.de (=?UTF-8?B?R8O2dHogUmVpbmlja2UgLSBJVCBLb29yZGluYXRvcg==?=) Date: Tue, 11 Aug 2015 15:45:33 +0200 Subject: backing up email / saving maildir on external hard drives In-Reply-To: References: <20150809114449.7b5a273f@mydesq2.domain.cxm> Message-ID: <55C9FC7D.4080201@filmakademie.de> Hi, talking of rsync and compression is may be also a bit misleading. On the destination there will be no compressed files if you transfer with rsync! The transfere on the network by rsync might be compressed! /G?tz Am 09.08.15 um 18:33 schrieb Kevin Laurie: > Dear Steve, > Very valuable info. Appreciate it and will be careful when using terms. > Actually I think I should just use rsync without compressing. The reason > why I started compressing was because the GUI gave some errors when I was > trying to copy then files. > > I'll just rsync the data from my laptop HDD to my external drive(without > compressing) > > Thanks > Kevin > > On Sunday, August 9, 2015, Steve Litt wrote: > >> On Sat, 8 Aug 2015 10:26:55 +0530 >> Kevin Laurie > wrote: >> >>> Hello, >>> Yesterday I tried to back up a 40GB maildir . >>> I tried to move the maildir from home to external HDD but failed. >> >> If you tried to *move* it it's an archive, not a backup. If you tried >> to *copy* it, with the intent of keeping the original on the original >> hard disk and using it further, and keeping today's copy on some other >> media, *that's* a backup. I'm not trying to be pedantic, but there are >> many distinctions between the two. Archives must be re-transferred >> frequently: Backups merely need to be redone at intervals. >> >>> Decided then to compress it(which took several hours). Now changing >>> the disk format from FAT to exFAT to allow the transfer for the large >>> compressed file. >> >> Reading https://en.wikipedia.org/wiki/File_Allocation_Table#exFAT , I >> personally wouldn't use exFAT. Regular FAT32 has a max filesize of >> 2GB-1, which is 50 times the size of your whole uncompressed maildir. >> >>> >>> How does one back up emails on a external drive? >>> Some advice would be greatly appreciated. >> >> Check this out: >> >> ================================================ >> slitt at mydesq2:~$ df -h ~/mail/Maildir >> Filesystem Size Used Avail Use% Mounted on >> /dev/sdb8 116G 11G 100G 10% /home/slitt/mail/Maildir >> slitt at mydesq2:~$ >> ================================================ >> >> I don't have 40 GB, but * have 11, which is less than an order of >> magnitude away. I just back up this puppy to my backup server with my >> normal rsync based backup procedures, which you can read about here: >> >> * http://www.troubleshooters.com/lpm/200609/200609.htm >> >> * http://www.troubleshooters.com/linux/blu-ray-backup.htm >> >> * http://www.troubleshooters.com/lpm/201408/201408.htm >> >> The stuff about Blu-Ray is important only if you back up to blu-ray. I >> like to keep some backups on write-once media, because kept in the >> shade at reasonable temperatures and humidities, it tends to last >> longer. And spinning disks that spend the majority of their time not >> spinning tend to have problems. >> >> If this is a *backup*, I'd leave it uncompressed so you can take >> incremental backups regularly. If it's an *archive*, meaning that the >> data is immediately removed from your computer after copy, compression >> might be in order, but you should make two copies and test them both >> thoroughly before deleting the original, and you should test them every >> couple months and if either goes bad, copy the other one to something >> good. Archives are a PITA. For 40GB in these days of $150 2TB drives, >> I'd keep the data intact, back it up, and when you outgrow your hard >> drive, just get a bigger one. >> >> In other parts of this thread you ask how to separate backups from >> different accounts from different computers. As far as accounts, I >> think that Maildir directory structures would take care of that. As far >> as different machines, just put the hostname at the front of each >> destination directory. >> >> >> >> SteveT >> >> Steve Litt >> July 2015 featured book: Rapid Learning for the 21st Century >> http://www.troubleshooters.com/rl21 >> -- G?tz Reinicke IT-Koordinator Tel. +49 7141 969 82420 E-Mail goetz.reinicke at filmakademie.de Filmakademie Baden-W?rttemberg GmbH Akademiehof 10 71638 Ludwigsburg www.filmakademie.de Eintragung Amtsgericht Stuttgart HRB 205016 Vorsitzender des Aufsichtsrats: J?rgen Walter MdL Staatssekret?r im Ministerium f?r Wissenschaft, Forschung und Kunst Baden-W?rttemberg Gesch?ftsf?hrer: Prof. Thomas Schadt -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 5481 bytes Desc: S/MIME Cryptographic Signature URL: From goetz.reinicke at filmakademie.de Tue Aug 11 13:59:01 2015 From: goetz.reinicke at filmakademie.de (=?UTF-8?B?R8O2dHogUmVpbmlja2UgLSBJVCBLb29yZGluYXRvcg==?=) Date: Tue, 11 Aug 2015 15:59:01 +0200 Subject: backing up email / saving maildir on external hard drives In-Reply-To: References: <4CE2F84F-7995-41B6-B7A8-852FBCB69072@valo.at> Message-ID: <55C9FFA5.1090709@filmakademie.de> Hi Kevin, if you use maildir, I bet there is not a single mail bigger than 100 MB. So ALL mails are 40 GB, but may be hundreds and thousands small ones ...? Or did you compress all files in one 40GB tar/gz/zip/... ? So fat32 is good for small files. BUT if you like to transfer as well e.g. DVD iso images larger than > 4 GB that might be a problem. From what I think is good for you to go with, you could do a "rsync ... magic needed options" that maildir to the FAT32 Disk. So you can read it on all systems without extra tools. Furthermore if you want to do an other backup or "update" your files run an other rsync ... so you dont have to compress a big archive all the time. As the others pointed out, what do you want to do, what is your requirement? Especially in a Linux/Mac/Win World there is no one sizes fits all. An other thing is: If you dont need to move your external disk around a lot or take it with you on the go, think of a NAS; e.g. two disks, mirrored raid so you can access it without big problems from all OSs .. my2cents. /G?tz Am 08.08.15 um 18:15 schrieb Kevin Laurie: > Dear Christian, > Thanks for your feedback. > The HDD will not accept larger than 4GB (as its in FAT format). Its a new > external HDD. Thinking of the best format(that would work with Mac , Win > and Linux) .seems like a challenge. > What's your view on NTFS? And why not exFAT? > > Thanks > Kevin > > On Saturday, August 8, 2015, Christian Kivalo wrote: > >> >> >> Am 08. August 2015 06:56:55 MESZ, schrieb Kevin Laurie < >> superinterstellar at gmail.com >: >>> Hello, >>> Yesterday I tried to back up a 40GB maildir . >>> I tried to move the maildir from home to external HDD but failed. >>> Decided >>> then to compress it(which took several hours). Now changing the disk >>> format >> >from FAT to exFAT to allow the transfer for the large compressed file. >>> >>> How does one back up emails on a external drive? >>> Some advice would be greatly appreciated. >> >> i backup my maildir with rsync to my home nas. >> >> i assume you are using linux or similar, so when backing up to an external >> hdd i'd also use rsync to copy over the maildir as is to the external hdd >> but use a different filesystem n that hdd like ext4 or xfs or anything else >> than fat/vfat/exfat. >> >> when something goes wrong you can just copy the files back. >> >> what was the error message you got when you tried to copy to the ext. hdd? >> >> regards >> - c >> -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 5481 bytes Desc: S/MIME Cryptographic Signature URL: From Christian.Schmidt at chemie.uni-hamburg.de Tue Aug 11 15:56:21 2015 From: Christian.Schmidt at chemie.uni-hamburg.de (Christian Schmidt) Date: Tue, 11 Aug 2015 17:56:21 +0200 Subject: backing up email / saving maildir on external hard drives In-Reply-To: <55C9FC7D.4080201@filmakademie.de> References: <20150809114449.7b5a273f@mydesq2.domain.cxm> <55C9FC7D.4080201@filmakademie.de> Message-ID: <55CA1B25.7080100@chemie.uni-hamburg.de> On 11.08.2015 15:45, G?tz Reinicke - IT Koordinator wrote: > talking of rsync and compression is may be also a bit misleading. > > On the destination there will be no compressed files if you transfer > with rsync! ...unless your dovecot is working with compressed maildir files. ;-) Kind Regards, Christian Schmidt -- No signature available. -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 5306 bytes Desc: S/MIME Cryptographic Signature URL: From goetz.reinicke at filmakademie.de Tue Aug 11 16:32:02 2015 From: goetz.reinicke at filmakademie.de (=?utf-8?Q?G=C3=B6tz_Reinicke?=) Date: Tue, 11 Aug 2015 18:32:02 +0200 Subject: backing up email / saving maildir on external hard drives In-Reply-To: <55CA1B25.7080100@chemie.uni-hamburg.de> References: <20150809114449.7b5a273f@mydesq2.domain.cxm> <55C9FC7D.4080201@filmakademie.de> <55CA1B25.7080100@chemie.uni-hamburg.de> Message-ID: > Am 11.08.2015 um 17:56 schrieb Christian Schmidt : > > On 11.08.2015 15:45, G?tz Reinicke - IT Koordinator wrote: >> talking of rsync and compression is may be also a bit misleading. >> >> On the destination there will be no compressed files if you transfer >> with rsync! > > ...unless your dovecot is working with compressed maildir files. ;-) > I was talking about the rsync file transfer, which will not compress the output on the backup server?. not about anything dovecote has configured ? may be uhh he uses a filesystem compression too ? Or some rsync piping trough gzip? kidding :) /G?tz From skdovecot at smail.inf.fh-brs.de Wed Aug 12 06:19:30 2015 From: skdovecot at smail.inf.fh-brs.de (Steffen Kaiser) Date: Wed, 12 Aug 2015 08:19:30 +0200 (CEST) Subject: PATCH for Stephan: Re: pigeonhole/lda accessing -m folder In-Reply-To: <55C8BDD6.1060007@rename-it.nl> References: <55C4862A.7070101@securepoint.de> <20150807102626.GG5380@ruhr-uni-bochum.de> <55C48B0D.5040200@securepoint.de> <55C4BAB3.4050507@securepoint.de> <55C887FC.7020002@securepoint.de> <55C8BDD6.1060007@rename-it.nl> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 10 Aug 2015, Stephan Bosch wrote: > I would make this dependent on an extension like > "vnd.dovecot.environment", so that non-standard features are only added Yes, sure :-) Maybe to place it into a vnd.dovecot. namespace of the variable extension is even simplier? > I'll give this a look. Also, are there any other environment items that > may be useful? Hm, + the current user for global scripts. + can you differ between "INBOX" specified explicitly or implicitly, I mean as result of the -m option or in LMTP via subaddress? This is what I need in some scripts, "if there is no explicit destination folder given, fileinto XYZ". The default_folder information comes very near to it. + Because I use LMTP rather then LDA as Matthias I wonder if you can make user-related information available, that can be configured via conf-file? + There had been a request about matching the subaddress against existing mail folders case-insensitively. I guess it is great performance penalty to read a complete list of mail folders of the user into the environment, so one can match against it? - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVcrlcnz1H7kL/d9rAQLNqAgAwL191ZlJicE2DNGbgKPV1VrvC4/Vg7iD HShfyiTqLFSmPUM1W07CBYtXKFdcnYPno/jbnhM8Qz3R2lgDoOIPvUFaxL8CKUnt t2/z1ZOg6qrMynLDr986Zvc6Yjbu0aX5/Ga02uvRWiKl8Zxp/NkIzA8F99b2Vn3i G0aDCYy6PG3pdNsPJ0zthIDVdkkmF4Kre3N2ufECF9tOcFizbs4tRrUXD1Gu0xlu W3pM+oRtACKkgHJufgXO0jhx0WBt370E7JSVO4M0KFnFfpoOCZk16OiYMJDDj/wL YlI52+TTlPy1+CBQTuWA8jWY5g5n8gBrCXZ3mFEiMjM03wRhn2hmSQ== =7bN5 -----END PGP SIGNATURE----- From cms at balius.com Wed Aug 12 15:19:27 2015 From: cms at balius.com (Chad M Stewart) Date: Wed, 12 Aug 2015 10:19:27 -0500 Subject: multiple messages per second to a single mailbox Message-ID: Dovecot 2.2.18 on CentOS 6 I have a pair of servers setup with MySQL, Postfix, and Dovecot. Replication is setup and working between the two dovecot instances. The problem I'm running into is that a single mailbox receives a lot of messages, at times the rate is multiple messages per second. Delivery from Postfix to Dovecot is via tcp based LMTP. When I do 'ps -aef|grep lmtp|wc -l' I get 62 and does not appear to go higher than that. At the moment I have 4500 and 8300 messages queued on two Postfix instances waiting to deliver via LMTP to the same dovecot instance. Deliveries only happen via LMTP and only one of the two nodes actually gets the deliveries. What I'm seeing is very high load on the system (40) and queues building on the Postfix side. Replication is keeping up. Looking at the logs now I see anywhere from 4-7 messages per second delivered to this single mailbox. I would like to increase that rate a lot. These machines are VMs hosted on Xenserver 6.x. I have them setup with 8 vCPUs (2 sockets with 4 cores per socket), the dom0 machines have dual HBA connectors back to a SAN and have 128 CPUs and 256GB of RAM and are not taxed. I added a 2nd virtual disk that is used for storing mailbox data. It is ext4 and has noatime set during mount. /var is also mounted with noatime. The performance graphs in XenCenter show nearly all 8 vCPUs at about 50%, and the writes on the mailbox data disk are about 20%. iostat is showing mostly <5 for await times for the disks, though I do see a 10 now and again. I'm guessing that maybe I'm hitting a mailbox locking issue and not sure how to reduce the contention and thereby increase the delivery rate to this mailbox. -Chad From andrzej.filip at gmail.com Wed Aug 12 16:04:03 2015 From: andrzej.filip at gmail.com (Andrzej A. Filip) Date: Wed, 12 Aug 2015 18:04:03 +0200 Subject: multiple messages per second to a single mailbox In-Reply-To: (Chad M. Stewart's message of "Wed, 12 Aug 2015 10:19:27 -0500") References: Message-ID: Chad M Stewart wrote: > Dovecot 2.2.18 on CentOS 6 > > I have a pair of servers setup with MySQL, Postfix, and Dovecot. Replication is setup and working between the two dovecot instances. > > The problem I'm running into is that a single mailbox receives a lot > of messages, at times the rate is multiple messages per > second. Delivery from Postfix to Dovecot is via tcp based LMTP. When > I do 'ps -aef|grep lmtp|wc -l' I get 62 and does not appear to go > higher than that. At the moment I have 4500 and 8300 messages queued > on two Postfix instances waiting to deliver via LMTP to the same > dovecot instance. Deliveries only happen via LMTP and only one of the > two nodes actually gets the deliveries. > > What I'm seeing is very high load on the system (40) and queues > building on the Postfix side. Replication is keeping up. Looking at > the logs now I see anywhere from 4-7 messages per second delivered to > this single mailbox. I would like to increase that rate a lot. > > These machines are VMs hosted on Xenserver 6.x. I have them setup > with 8 vCPUs (2 sockets with 4 cores per socket), the dom0 machines > have dual HBA connectors back to a SAN and have 128 CPUs and 256GB of > RAM and are not taxed. I added a 2nd virtual disk that is used for > storing mailbox data. It is ext4 and has noatime set during > mount. /var is also mounted with noatime. > > The performance graphs in XenCenter show nearly all 8 vCPUs at about > 50%, and the writes on the mailbox data disk are about 20%. iostat is > showing mostly <5 for await times for the disks, though I do see a 10 > now and again. > > > I'm guessing that maybe I'm hitting a mailbox locking issue and not sure how > to reduce the contention and thereby increase the delivery rate to this > mailbox. Could you provide the following info: a) mailbox type (maildir/mbox/dbox/...) [mail_location in dovecot's config] b) file system type (ext2/ext3/ext4/fat32/...) [provided by "df -T" command on my system] -- A. Filip From cms at balius.com Wed Aug 12 16:25:56 2015 From: cms at balius.com (Chad M Stewart) Date: Wed, 12 Aug 2015 11:25:56 -0500 Subject: multiple messages per second to a single mailbox In-Reply-To: References: Message-ID: <84C57ADD-1F76-4276-9DDD-B5B347600A80@balius.com> On Aug 12, 2015, at 11:04 AM, Andrzej A. Filip wrote: > > <..snip..> > Could you provide the following info: > a) mailbox type (maildir/mbox/dbox/...) maildir > [mail_location in dovecot's config] /srv/mail/// > b) file system type (ext2/ext3/ext4/fat32/...) > [provided by "df -T" command on my system] As I said ext4. Since I posted I've changed a couple of things: ulimit -n 8192, and disabled fsync as in mail_fsync = never. I'm not sure if I'll put it back in the LMTP section or not. Given all the hardware abstraction layers. -Chad From cms at balius.com Wed Aug 12 16:33:13 2015 From: cms at balius.com (Chad M Stewart) Date: Wed, 12 Aug 2015 11:33:13 -0500 Subject: multiple messages per second to a single mailbox In-Reply-To: <84C57ADD-1F76-4276-9DDD-B5B347600A80@balius.com> References: <84C57ADD-1F76-4276-9DDD-B5B347600A80@balius.com> Message-ID: On Aug 12, 2015, at 11:25 AM, Chad M Stewart wrote: > <..snip..> > Since I posted I've changed a couple of things: ulimit -n 8192, and disabled fsync as in mail_fsync = never. I'm not sure if I'll put it back in the LMTP section or not. Given all the hardware abstraction layers. I forgot to mention that on a clean start of dovecot (after making those changes), I counted 25 msgs delivered to this single mailbox (all INBOX) in a second. Then it slowed down. Maybe the answer is less LMTP connections and increase the number of messages per connection. I have Postfix configured to do outbound connection caching and I've seen entries in the logs indicating 30 messages on a given connection. I'll have to look into how to limit dovecot to a certain number of LMTP processes. -Chad From troeder at univention.de Wed Aug 12 16:58:42 2015 From: troeder at univention.de (=?windows-1252?Q?Daniel_Tr=F6der?=) Date: Wed, 12 Aug 2015 18:58:42 +0200 Subject: multiple messages per second to a single mailbox In-Reply-To: References: Message-ID: <55CB7B42.104@univention.de> On 08/12/2015 17:19, Chad M Stewart wrote: > What I'm seeing is very high load on the system (40) and queues building on the Postfix side. High load means, that there are a lot of processes waiting to run. The most likely cause for this is not CPU consumption, but I/O wait. Please run vmstat and iostat and post their output. Greetings Daniel -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From cms at balius.com Wed Aug 12 17:35:26 2015 From: cms at balius.com (Chad M Stewart) Date: Wed, 12 Aug 2015 12:35:26 -0500 Subject: multiple messages per second to a single mailbox In-Reply-To: <55CB7B42.104@univention.de> References: <55CB7B42.104@univention.de> Message-ID: <8D617AEC-0038-4B4F-B393-C95DCD737F96@balius.com> On Aug 12, 2015, at 11:58 AM, Daniel Tr?der wrote: > On 08/12/2015 17:19, Chad M Stewart wrote: >> What I'm seeing is very high load on the system (40) and queues building on the Postfix side. > High load means, that there are a lot of processes waiting to run. The > most likely cause for this is not CPU consumption, but I/O wait. > > Please run vmstat and iostat and post their output. I was watching iostat and avg service times, and maybe once every 30-45 seconds I'd see times of 10ms, but otherwise it was below that. I achieved the biggest impact by limiting the number of outbound connections from Postfix to Dovecot. I limited Postfix to 5 connections, which means a total of 10 inbound LMTP to Dovecot. Then I saw near 500 msgs per LMTP connection. I suspect the problem was a locking issue on the mailbox in question. Too many simultaneous delivery attempts via too many LMTP sessions. The backlog has cleared so I'm done troubleshooting for now. If this happens again I'll resume looking into it more. These are new servers so I'm tuning for the load, etc.. -Chad From andrew at mcnaughty.com Wed Aug 12 20:08:52 2015 From: andrew at mcnaughty.com (Andrew McN) Date: Thu, 13 Aug 2015 06:08:52 +1000 Subject: imap segfaults when using antispam plugin, and moving mail to an 'unsure' folder. Message-ID: <55CBA7D4.3040101@mcnaughty.com> I put up a question on serverfault first, which has the details, including a stack trace: http://serverfault.com/questions/713687/dovecot-antispam-error-storing-mail . Details also copied below. Is this a known bug? If I built dovecot from source rather than using the Ubuntu 14.04 packages, would I expect that to help? Are there any configuration details that might be relevant? (dovecot -n output shown Any help much appreciated. Andrew McNaughton Details on the serverfault page repeated below, plus a little more. ----------------------------------------------------- Ubuntu 14.04.3 x86_64 Running inside lxc-docker 1.7.0 dovecot-imapd 1:2.2.9-1ubuntu2 dovecot-antispam 2.0+20130822-2build1 Filesystem is via docker, so aufs is involved, and under that on the host systems is ext4 ----------------------------------------------------- I know mail is not supposed to be moved into a mailbox of type 'unsure'. The server should refuse this action. What actually happens though is that the imap process throws an error and dies: Aug 12 17:53:07 255c7fc2d2f0 imap: antispam: mailbox_is_unsure(Review-Classification): 1 Aug 12 17:53:07 255c7fc2d2f0 dovecot: imap(user at example.com): Panic: file mail-storage.c: line 2141 (mailbox_copy): assertion failed: (!ctx->unfinished) Aug 12 17:53:07 255c7fc2d2f0 dovecot: imap(user at example.com): Error: Raw backtrace: /usr/lib/dovecot/libdovecot.so.0(+0x5e271) [0x7f718e7dd271] -> /usr/lib/dovecot/libdovecot.so.0(+0x5e34e) [0x7f718e7dd34e] -> /usr/lib/dovecot/libdovecot.so.0(i_fatal+0) [0x7f718e798a9e] -> /usr/lib/dovecot/libdovecot-storage.so.0(+0x7196f) [0x7f718eab896f] -> /usr/lib/dovecot/libdovecot-storage.so.0(mailbox_move+0x15) [0x7f718eab8985] -> dovecot/imap(+0xded5) [0x7f718ef84ed5] -> dovecot/imap(command_exec+0x3c) [0x7f718ef8f6dc] -> dovecot/imap(+0x176cf) [0x7f718ef8e6cf] -> dovecot/imap(+0x17785) [0x7f718ef8e785] -> dovecot/imap(client_handle_input+0x125) [0x7f718ef8eaa5] -> dovecot/imap(client_input+0x75) [0x7f718ef8ee35] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_call_io+0x27) [0x7f718e7ed247] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_handler_run+0xd7) [0x7f718e7edfd7] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_run+0x38) [0x7f718e7ecde8] -> /usr/lib/dovecot/libdovecot.so.0(master_service_run+0x13) [0x7f718e79dc93] -> dovecot/imap(main+0x2c4) [0x7f718ef832f4] -> /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf5) [0x7f718e3dbec5] -> dovecot/imap(+0xc460) [0x7f718ef83460] Aug 12 17:53:07 255c7fc2d2f0 dovecot: imap(user at example.com): Fatal: master: service(imap): child 58 killed with signal 6 (core dumped) ----------------------------------------------------------------------- root at 8ec9579667fc:/# gdb /usr/lib/dovecot/imap /tmp/61 GNU gdb (Ubuntu 7.7.1-0ubuntu5~14.04.2) 7.7.1 Copyright (C) 2014 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "x86_64-linux-gnu". Type "show configuration" for configuration details. For bug reporting instructions, please see: . Find the GDB manual and other documentation resources online at: . For help, type "help". Type "apropos word" to search for commands related to "word"... Reading symbols from /usr/lib/dovecot/imap...Reading symbols from /usr/lib/debug//usr/lib/dovecot/imap...done. done. [New LWP 61] Core was generated by `dovecot/imap'. Program terminated with signal SIGABRT, Aborted. #0 0x00007f5dd5394cc9 in __GI_raise (sig=sig at entry=6) at ../nptl/sysdeps/unix/sysv/linux/raise.c:56 56 ../nptl/sysdeps/unix/sysv/linux/raise.c: No such file or directory. (gdb) bt full #0 0x00007f5dd5394cc9 in __GI_raise (sig=sig at entry=6) at ../nptl/sysdeps/unix/sysv/linux/raise.c:56 resultvar = 0 pid = 61 selftid = 61 #1 0x00007f5dd53980d8 in __GI_abort () at abort.c:89 save_stage = 2 act = {__sigaction_handler = {sa_handler = 0x800, sa_sigaction = 0x800}, sa_mask = {__val = {140041026109520, 178, 140041026077952, 512, 140040990095251, 140041026077952, 140041026077952, 512, 140040990087565, 140734823456136, 140040990231826, 140041026077952, 140734823456040, 0, 140040990231929, 140041026077952}}, sa_flags = -713567062, sa_restorer = 0x7fff6128dd01} sigs = {__val = {32, 0 }} #2 0x00007f5dd5781298 in default_fatal_finish (type=, status=status at entry=0) at failures.c:192 backtrace = 0x7f5dd79d0538 "/usr/lib/dovecot/libdovecot.so.0(+0x5e271) [0x7f5dd5781271] -> /usr/lib/dovecot/libdovecot.so.0(+0x5e34e) [0x7f5dd578134e] -> /usr/lib/dovecot/libdovecot.so.0(i_fatal+0) [0x7f5dd573ca9e] -> /usr/lib/d"... #3 0x00007f5dd578134e in i_internal_fatal_handler (ctx=0x7fff6128dd70, format=, args=) at failures.c:653 status = 0 #4 0x00007f5dd573ca9e in i_panic (format=format at entry=0x7f5dd5ab1e68 "file %s: line %d (%s): assertion failed: (%s)") at failures.c:264 ctx = {type = LOG_TYPE_PANIC, exit_status = 0, timestamp = 0x0} args = {{gp_offset = 40, fp_offset = 48, overflow_arg_area = 0x7fff6128de60, reg_save_area = 0x7fff6128dda0}} #5 0x00007f5dd5a5c96f in mailbox_copy (_ctx=, mail=mail at entry=0x7f5dd79ffbc0) at mail-storage.c:2141 ctx = 0x7f5dd7a043e0 t = 0x7f5dd79fdf00 keywords = 0x0 pvt_flags = 0 real_mail = ret = -1 __FUNCTION__ = "mailbox_copy" #6 0x00007f5dd5a5c985 in mailbox_move (_ctx=, mail=0x7f5dd79ffbc0) at mail-storage.c:2153 ctx = 0x7f5dd7a043e0 #7 0x00007f5dd5f28ed5 in fetch_and_copy (copy_count_r=, src_uidset_r=, search_args=, src_trans_r=0x7fff6128df18, t=, move=true, client=0x7f5dd79ee440) at cmd-copy.c:67 search_ctx = 0x7f5dd79ff920 src_trans = 0x7f5dd79fec10 srcset_ctx = {str = 0x7f5dd79d0350, first_uid = 0, last_uid = 4294967295} ret = 1 save_ctx = 0x0 mail = 0x7f5dd79ffbc0 copy_count = 1 src_uidset = 0x7f5dd79d0350 #8 cmd_copy_full (cmd=0x7f5dd79eee70, move=) at cmd-copy.c:123 client = 0x7f5dd79ee440 dest_storage = destbox = 0x7f5dd79f8d10 t = 0x7f5dd79fdf00 src_trans = 0x7f5dd573d029 search_args = 0x7f5dd79f73a0 messageset = 0x7f5dd79dbc10 "76" mailbox = 0x7f5dd79dbc18 "Review-Classification" src_uidset = sync_flags = (unknown: 0) imap_flags = (unknown: 0) changes = {pool = 0x7f5dd79d0350, uid_validity = 0, saved_uids = {arr = {buffer = 0x1, element_size = 140040986301132}, v = 0x1, v_modifiable = 0x1}, ignored_modseq_changes = 3617438728, changed = 93, no_read_perm = 127} msg = ret = __FUNCTION__ = "cmd_copy_full" #9 0x00007f5dd5f336dc in command_exec (cmd=0x7f5dd79eee70) at imap-commands.c:158 hook = 0x7f5dd79d90d0 ret = #10 0x00007f5dd5f326cf in client_command_input (cmd=0x7f5dd79eee70) at imap-client.c:780 client = 0x7f5dd79ee440 command = __FUNCTION__ = "client_command_input" #11 0x00007f5dd5f32785 in client_command_input (cmd=0x7f5dd79eee70) at imap-client.c:841 client = 0x7f5dd79ee440 command = __FUNCTION__ = "client_command_input" #12 0x00007f5dd5f32aa5 in client_handle_next_command (remove_io_r=, client=0x7f5dd79ee440) at imap-client.c:879 No locals. #13 client_handle_input (client=client at entry=0x7f5dd79ee440) at imap-client.c:891 _data_stack_cur_id = 3 remove_io = false handled_commands = false __FUNCTION__ = "client_handle_input" #14 0x00007f5dd5f32e35 in client_input (client=0x7f5dd79ee440) at imap-client.c:933 cmd = 0x7f5dd79e5ae0 output = 0x7f5dd79eeca0 bytes = 39 __FUNCTION__ = "client_input" #15 0x00007f5dd5791247 in io_loop_call_io (io=0x7f5dd79eed90) at ioloop.c:388 ioloop = 0x7f5dd79d8720 t_id = 2 #16 0x00007f5dd5791fd7 in io_loop_handler_run (ioloop=ioloop at entry=0x7f5dd79d8720) at ioloop-epoll.c:220 ctx = 0x7f5dd79d93b0 list = 0x7f5dd79dade0 io = tv = {tv_sec = 1799, tv_usec = 999178} events_count = msecs = ret = 1 i = 0 call = __FUNCTION__ = "io_loop_handler_run" #17 0x00007f5dd5790de8 in io_loop_run (ioloop=0x7f5dd79d8720) at ioloop.c:412 __FUNCTION__ = "io_loop_run" #18 0x00007f5dd5741c93 in master_service_run (service=0x7f5dd79d85b0, callback=) at master-service.c:566 No locals. #19 0x00007f5dd5f272f4 in main (argc=1, argv=0x7f5dd79d8390) at main.c:400 set_roots = {0x7f5dd614aa00 , 0x0} login_set = {auth_socket_path = 0x7f5dd79d0048 "\001", postlogin_socket_path = 0x0, postlogin_timeout_secs = 60, callback = 0x7f5dd5f3c0f0 , failure_callback = 0x7f5dd5f3be20 , request_auth_token = 1} service_flags = storage_service_flags = username = 0x0 c = ------------------------------------------------------------------------------- root at 8ec9579667fc:/# dovecot -n # 2.2.9: /etc/dovecot/dovecot.conf # OS: Linux 3.13.0-43-generic x86_64 Ubuntu 14.04.3 LTS ext4 auth_debug = yes auth_verbose = yes disable_plaintext_auth = no first_valid_gid = 150 first_valid_uid = 150 last_valid_gid = 150 last_valid_uid = 150 mail_debug = yes mail_gid = vmail mail_location = maildir:/var/vmail/%d/%n/Maildir:INDEX=/var/vmail/%d/%n/Maildir/indexes mail_uid = vmail namespace inbox { inbox = yes location = mailbox Attention-Not-Required { auto = subscribe special_use = \Junk } mailbox Attention { auto = subscribe } mailbox Drafts { auto = subscribe special_use = \Drafts } mailbox Review-Classification { auto = subscribe } mailbox Sent { auto = subscribe special_use = \Sent } mailbox Trash { auto = subscribe special_use = \Trash } prefix = } passdb { args = /etc/dovecot/dovecot-sql.conf.ext driver = sql } plugin { antispam_backend = dspam antispam_debug_target = syslog antispam_dspam_args = --deliver=;--user;%u; antispam_dspam_binary = /usr/bin/dspam antispam_signature = X-DSPAM-Signature antispam_signature_missing = error antispam_spam = Attention-Not-Required antispam_trash = Trash antispam_unsure = Review-Classification antispam_verbose_debug = 1 mail_log_events = delete undelete expunge copy mailbox_delete mailbox_rename mail_log_fields = uid box msgid size sieve = ~/.dovecot.sieve sieve_dir = ~/sieve } postmaster_address = postmaster at example.net protocols = " imap lmtp" service lmtp { inet_listener lmtp { address = 127.0.0.1 port = 24 } user = vmail } ssl_cert = References: <55C341B9.9080203@granch.ru> <55C8A8BD.9090204@mie.utoronto.ca> Message-ID: <55CC2882.6090206@granch.ru> 10/08/15 19:35, Oscar del Rio ?????: > > You forgot to post your "doveconf -n" configuration. > It looks like your attachment storage includes the username in the path. > That won't work. > The path should be global, e.g. /usr/share/mail/attachments, and > readable/writable by the dovecot process. > > SiS deduplicates the attachments regardless of username since the same > attachment might be sent/received by multiple users. SiS will save a > single copy in the global path and transparently link it to the users' > mailboxes. Thank you for reply. I had have understood this already, when spent some time for thinking about which ways available for SiS to deduplicate attachments without access to FS and kernel. Of course, is there the only way - keep attachments in big one pile and link it to user's mail. It is not for me. When I'll set mail_attachment_fs=posix will dovecot stores attachments for each user? I.e. when mail group info aliased to users aaa, bbb and ccc, will dovecot store attachments separately for aaa at granch.ru, bbb at granch.ru, ccc at granch.ru or it again try to keep it in aaa's mail directory and link to another two? -- ? ?????????. ?????? ????? ?????????????? (AchilovRN at granch.ru) ??????? ?????????? ?? ?????? ?????????? ??? ??? "?????", ???: +7 (383) 233-35-12, ???. 107 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2606 bytes Desc: ?????????????????????????????????? ?????????????? S/MIME URL: From AchilovRN at granch.ru Thu Aug 13 05:33:03 2015 From: AchilovRN at granch.ru (=?UTF-8?B?0JDRh9C40LvQvtCyINCg0LDRiNC40LQg0J3Rg9GA0LzRg9GF0LDQvNC10LQ=?= =?UTF-8?B?0L7QstC40Yc=?=) Date: Thu, 13 Aug 2015 11:33:03 +0600 Subject: Convert from Maildir to Mdbox In-Reply-To: <55C9A5A8.9010501@echter-kuechen-elektro.de> References: <55C9A5A8.9010501@echter-kuechen-elektro.de> Message-ID: <55CC2C0F.2010709@granch.ru> 11/08/15 13:35, J. Echter ?????: > Hi, > > i'd like to convert from maildir to mdbox. > > what would be the best way to do this? > > i have read the dsync part of the dovecot wiki but i don't get what i > have to use for my setup. Sometime, when I must converted many user's mailboxes from mbox to mdbox, I had have wrote one script to mass convert mailboxes for local or domain users and can provide it for you. It support mbox and maildir as source formats. -- ? ?????????. ?????? ????? ?????????????? (AchilovRN at granch.ru) ??????? ?????????? ?? ?????? ?????????? ??? ??? "?????", ???: +7 (383) 233-35-12, ???. 107 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2606 bytes Desc: ?????????????????????????????????? ?????????????? S/MIME URL: From AchilovRN at granch.ru Thu Aug 13 05:42:02 2015 From: AchilovRN at granch.ru (=?UTF-8?B?0JDRh9C40LvQvtCyINCg0LDRiNC40LQg0J3Rg9GA0LzRg9GF0LDQvNC10LQ=?= =?UTF-8?B?0L7QstC40Yc=?=) Date: Thu, 13 Aug 2015 11:42:02 +0600 Subject: Error when reading attachment - one byte overhead? Message-ID: <55CC2E2A.7050405@granch.ru> After converting maildir to mdbox mail periodically obtained these messages: Aug 13 11:22:27 mailsrv dovecot: imap(user-aaa): Error: read(/usr/share/smbusers/user-aaa/mail/attachments/a0/8e/a08e3e63e11aae3f37020bdca0da7d-7b39ab07ea73ba559d450000dec08558[base64:18 b/l]) failed: Stream is larger than expected (434427 > 434426, eof=1) Aug 13 11:22:27 mailsrv dovecot: imap(user-aaa): Error: read(attachments-connector(/usr/share/smbusers/user-aaa/mail/storage/m.23)) failed: read(/usr/share/smbusers/user-aaa/mail/attachments/a0/8e/a08e3e63e11aae3f37020bdca0da7d-7b39ab07ea73ba559d450000dec08558[base64:18 b/l]) failed: Stream is larger than expected (434427 > 434426, eof=1) (FETCH BODY[] for mailbox ?????? ????????/??? ?????? UID 20) Aug 13 11:22:27 mailsrv dovecot: imap(user-aaa): Disconnected: FETCH read() failed in=16923 out=497431 How to correct it? Conversion was did with 2.2.8 version, next will installed 2.2.18. -- ? ?????????. ?????? ????? ?????????????? (AchilovRN at granch.ru) ??????? ?????????? ?? ?????? ?????????? ??? ??? "?????", ???: +7 (383) 233-35-12, ???. 107 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2606 bytes Desc: ?????????????????????????????????? ?????????????? S/MIME URL: From vvu at mcra.fr Thu Aug 13 07:45:37 2015 From: vvu at mcra.fr (Vu Ngoc VU) Date: Thu, 13 Aug 2015 09:45:37 +0200 (CEST) Subject: need help debugging deleted mails Message-ID: hello, one of our users complained about the fact that he doesn't receive some messages (randomly) that his teamates have. Grepping for the message-ID in the mail.log, I can see lmtp says it is saved : Aug 12 15:23:23 imap-01 dovecot: lmtp(4943, johndoe at example.net): fFuCO09Hy1VPEwAA2t/0+w: sieve: vsize= msgid=<037e01d0d502$0e3687d0$2aa39770$@example.net>: stored mail into mailbox 'INBOX' The only difference with others recipients is the 'sieve:... stored mail' instead of 'saved mail to' when there is no sieve. But when he asked me, the message was no more in his maildir (I checked with doveadm search). As we have checked and disabled all the filters in his MUA and on the server (sieve) and this issue still continue, I did mkdir a dovecot.rawlog inside his home. With the help of rawlog, I can see that the message can be found in a .out file (by grepping the Message-ID). And by reading this .out log raw log, the entire message were sent to the client. So I guess that : * the message was written to disk * it was sent to the MUA But I need your help because I don't get how this message disappear from his maildir. My assumptions was that the client must do some commands like : * STORE the \Deleted flag to this message UID * and then, later do EXPUNGE but I did : * grep -i expunge * 20150812* => no match * grep -iE '\+flags .*\\deleted' => a few matches but not with the appropriate UID snippet from the 20150812-133858-13317.in file: 8nk1 UID FETCH 208209 (UID FLAGS RFC822.SIZE BODY.PEEK[HEADER] INTERNALDATE) p7uv IDLE DONE 6yk3 NOOP 83g6 IDLE DONE yhbd UID FETCH 208209 (UID FLAGS BODY.PEEK[]) fy16 IDLE DONE snippet from the 20150812-133858-13317.out file: * 56295 FETCH (UID 208209) wn2p OK Fetch completed. * 56295 FETCH (UID 208209 FLAGS (\Deleted \Seen \Recent) RFC822.SIZE 17138 INTERNALDATE "12-Aug-2015 15:23:22 +0200" BODY[HEADER] {2836} Considering these points: * I cannot find any match with 208209 early in raw logs * I don't see any expunge for the whole day * I've grepped for \Deleted (insensitive) and didn't find matching like a interval of messages UID. What did I miss to find out how/when the message got \Deleted flag and was removed from his maildir? A few about my setup : * system is ubuntu 14.04 inside OpenVZ containers (Proxmox) * dovecot version is 2.2.9 from Ubuntu packages * 2 director with Internet and local ip addresses * 2 backends with only local addresses * these dovecot access storage servers by NFSv3 (that's why we have director) * imap/lmtp are proxied Thanks a lot for reading, I'm really disappointed about this problem. For the moment, I'll delete his script even if all rules are disabled (via roundcube, and I checked the rule is commented in the file). And also I'll try to watch what happens with inotify on the storage server. From darix at opensu.se Thu Aug 13 11:38:58 2015 From: darix at opensu.se (Marcus =?UTF-8?B?UsO8Y2tlcnQ=?=) Date: Thu, 13 Aug 2015 13:38:58 +0200 Subject: need help debugging deleted mails In-Reply-To: References: Message-ID: <20150813133858.45d1d8f5@tengu.suse.de> mail_log plugin might help darix -- openSUSE - SUSE Linux is my linux openSUSE is good for you www.opensuse.org From Christian.Schmidt at chemie.uni-hamburg.de Thu Aug 13 13:23:14 2015 From: Christian.Schmidt at chemie.uni-hamburg.de (Christian Schmidt) Date: Thu, 13 Aug 2015 15:23:14 +0200 Subject: backing up IMAP server on a hard drive In-Reply-To: References: <20150806102112.Horde.vSPTHO6e4PuheGsNw3FVyg1@www.vfemail.net> <20150806110613.Horde.Hi0mD3dS1oJ42DKObmZ9SQ8@www.vfemail.net> Message-ID: <55CC9A42.7030508@chemie.uni-hamburg.de> On 06.08.2015 18:11, Kevin Laurie wrote: > I am not sure what the parameter for the -host2 should be(apart from > localhost? ) localhost. Regards, Christian Schmidt -- No signature available. -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 5306 bytes Desc: S/MIME Cryptographic Signature URL: From vvu at mcra.fr Thu Aug 13 14:15:00 2015 From: vvu at mcra.fr (Vu Ngoc VU) Date: Thu, 13 Aug 2015 16:15:00 +0200 (CEST) Subject: need help debugging deleted mails In-Reply-To: <20150813133858.45d1d8f5@tengu.suse.de> References: <20150813133858.45d1d8f5@tengu.suse.de> Message-ID: > Date: Thu, 13 Aug 2015 13:38:58 > From: Marcus R?ckert > To: dovecot at dovecot.org > Subject: Re: need help debugging deleted mails > > mail_log plugin might help > > darix I've seen this plugin, but I fear that it will be enabled for all users. I don't have seen an way to enable it only for 1 user like rawlog permits. Maybe, I'll setup a new container for only users that I want to "debug" with that plugin enabled. From bind at enas.net Thu Aug 13 14:28:54 2015 From: bind at enas.net (Urban Loesch) Date: Thu, 13 Aug 2015 16:28:54 +0200 Subject: need help debugging deleted mails In-Reply-To: References: <20150813133858.45d1d8f5@tengu.suse.de> Message-ID: <55CCA9A6.8060003@enas.net> Am 13.08.2015 um 16:15 schrieb Vu Ngoc VU: >> Date: Thu, 13 Aug 2015 13:38:58 >> From: Marcus R?ckert >> To: dovecot at dovecot.org >> Subject: Re: need help debugging deleted mails >> >> mail_log plugin might help >> >> darix > > I've seen this plugin, but I fear that it will be enabled for all users. > I don't have seen an way to enable it only for 1 user like rawlog permits. We have more than 20k accounts on our server and we have no problem with this plugin at all. It helps us very often, if some users are complaining that they are loosing emails. Every time it was a fault by the users different clients, eg. acces via IMAP and POP3. > > Maybe, I'll setup a new container for only users that I want to "debug" with that plugin enabled. > Reagards Urban From darix at opensu.se Thu Aug 13 14:31:02 2015 From: darix at opensu.se (Marcus Rueckert) Date: Thu, 13 Aug 2015 16:31:02 +0200 Subject: need help debugging deleted mails In-Reply-To: References: <20150813133858.45d1d8f5@tengu.suse.de> Message-ID: <20150813143101.GA2869@nordisch.org> On 2015-08-13 16:15:00 +0200, Vu Ngoc VU wrote: > I've seen this plugin, but I fear that it will be enabled for all users. > I don't have seen an way to enable it only for 1 user like rawlog permits. > > Maybe, I'll setup a new container for only users that I want to "debug" with that plugin enabled. TBH the output from this plugin is very valuable for your support team in any case. so it doesnt harm to have it enabled by default. darix -- openSUSE - SUSE Linux is my linux openSUSE is good for you www.opensuse.org From Jost.Krieger+dovecot at rub.de Thu Aug 13 14:33:24 2015 From: Jost.Krieger+dovecot at rub.de (Jost Krieger) Date: Thu, 13 Aug 2015 16:33:24 +0200 Subject: sieve-filter failure problems Message-ID: <20150813143324.GE46491@ruhr-uni-bochum.de> I use sieve-filter for postprocessing misclassified mail. For false positives I use the following script: require [ "variables", "include", "fileinto" ]; global [ "FORCENOSPAM", "ext", "ext1" ]; set "FORCENOSPAM" "YES"; fileinto "JUNK-PRENOUCE"; if header :matches "Delivered-To" "*+*@*" { set "ext" "${2}"; set "ext1" "${2}"; } include :personal "production"; FORCENOSPAM suppresses the spam handling and the Delivered-To stuff tries to regenerate envelope info. This is my crontab entry: 1-57/4 * * * * cd ~kriegjcb; /usr/local/dovecot/bin/sieve-filter -v -e -W -u kriegjcb -m Spam/U0-failed ~kriegjcb/sieve/nospamfilter.sieve Spam/U0 Most things work fine, but a few of my rules will try to forward mail. This will fail and my idea was I handle this manually from the folder Spam/U0-failed. Unfortunately the mail will arrive there but will also be left in the source folder Spam/U0 and will be processed again every 4 minutes ... This is probably not intentional? Yours Jost Krieger -- | Jost.Krieger+sig at ruhr-uni-bochum.de Please help stamp out spam! | | Postmaster, JAPH, resident answer machine at RUB Comp. Center | | Sincere words are not sweet, sweet words are not sincere. | | Lao Tse, Tao Te King 81 | -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 4784 bytes Desc: not available URL: From vvu at mcra.fr Thu Aug 13 14:42:03 2015 From: vvu at mcra.fr (Vu Ngoc VU) Date: Thu, 13 Aug 2015 16:42:03 +0200 (CEST) Subject: need help debugging deleted mails In-Reply-To: <20150813143101.GA2869@nordisch.org> References: <20150813133858.45d1d8f5@tengu.suse.de> <20150813143101.GA2869@nordisch.org> Message-ID: > Date: Thu, 13 Aug 2015 16:31:02 > From: Marcus Rueckert > To: dovecot at dovecot.org > Subject: Re: need help debugging deleted mails > > On 2015-08-13 16:15:00 +0200, Vu Ngoc VU wrote: >> I've seen this plugin, but I fear that it will be enabled for all users. >> I don't have seen an way to enable it only for 1 user like rawlog permits. >> >> Maybe, I'll setup a new container for only users that I want to "debug" with that plugin enabled. > > TBH the output from this plugin is very valuable for your support team > in any case. so it doesnt harm to have it enabled by default. OK, I'll enable it then. I added some auditd rule on his Maildir/cur on the storage server. But looks like it's just a NFS share, events cannot be intercepted. From vvu at mcra.fr Thu Aug 13 14:43:22 2015 From: vvu at mcra.fr (Vu Ngoc VU) Date: Thu, 13 Aug 2015 16:43:22 +0200 (CEST) Subject: need help debugging deleted mails In-Reply-To: <55CCA9A6.8060003@enas.net> References: <20150813133858.45d1d8f5@tengu.suse.de> <55CCA9A6.8060003@enas.net> Message-ID: > Date: Thu, 13 Aug 2015 16:28:54 > From: Urban Loesch > Reply-To: Dovecot Mailing List > To: dovecot at dovecot.org > Subject: Re: need help debugging deleted mails > >> I've seen this plugin, but I fear that it will be enabled for all users. >> I don't have seen an way to enable it only for 1 user like rawlog permits. > > We have more than 20k accounts on our server and we have no problem with this plugin at all. > It helps us very often, if some users are complaining that they are loosing emails. > Every time it was a fault by the users different clients, eg. acces via IMAP > and POP3. OK, thanks for sharing your experience about this plugin. I'll add it to our dovecot conf. From dravion.smith at gmx.net Fri Aug 14 11:22:56 2015 From: dravion.smith at gmx.net (dravion.smith at gmx.net) Date: Fri, 14 Aug 2015 13:22:56 +0200 Subject: Multidomain / IP Address Setup (Dovevot 2.2.10 on CentOS7 ) is failing: Fatal: Couldn't parse private ssl_key: error:0906D06C:PEM Message-ID: <55CDCF90.8050207@gmx.net> Hi, I want to say hello and here is my big problem ;D Iam trying to archive a Postfix/Dovecot 2.2.10 CentOS7 Multidomain Setup with multiple (valid StartSSL Certs), but iam only able to run a single Domain Cert server only. ps: I need a multiple domainssetup for every customer and it is not an option for me redirecting any email to a single domain server. I really need this setup working. IMHO: I think it SELinux could interfere with multiple Certs in diffrent folders (it is activated in CentOs7 by default and is needed by other apps) Ok, here is my logfile data: systemctl start postfix.service [OK] systemctl start dovecot.service [OK] /var/log/messages *systemd: Stopping Dovecot IMAP/POP3 email server... *systemd: Starting Dovecot IMAP/POP3 email server... *systemd: Started Dovecot IMAP/POP3 email server. /var/log/maillog *dovecot: master: Dovecot v2.2.10 starting up for imap, pop3, lmtp (core dumps disabled) ### This works (Thunderbird, Outlook 2013, Opera Mail ect.) #### local mydomain01.tld { protocol imap { ssl_cert = I have a fairly standard antispam/dspam setup: postfix delivers to dspam, which classifies and tags the mail, and then passes it to postfix which uses sieve to put mail into the appropriate folders based on dspam's classification. I also have the antispam plugin watching mail being moved between folders and re-training dpsam accordingly. If I now add some sieve filter rules which put incoming mail into appropriate folders, possibly overriding what would normally result from dspam's classification, will that be seen by the antispam plugin and used for training dspam? Regards, Andrew McNaughton From kremels at kreme.com Fri Aug 14 14:51:27 2015 From: kremels at kreme.com (@lbutlr) Date: Fri, 14 Aug 2015 08:51:27 -0600 Subject: Error: User initialization failed Message-ID: OK, I obviously forgot something when setting up new user accounts (which I did via postfix admin to add them to the MySQL database): dovecot: lda(chance at example.com): Error: User initialization failed: Namespace '': Ambiguous mail location setting, don't know what to do with it: /usr/local/virtual/chance@ example.com (try prefixing it with mbox: or milder:) I have lots of mysql users defined and working, but do I need to manually create the mail store for new ones? (Sorry, it?s been quite a long time since I had to add accounts). -- You start a conversation you can't even finish it You're talkin' a lot, but you're not sayin' anything When I have nothing to say, my lips are sealed Say something once, why say it again? From cms at balius.com Fri Aug 14 14:52:12 2015 From: cms at balius.com (Chad M Stewart) Date: Fri, 14 Aug 2015 09:52:12 -0500 Subject: multiple messages per second to a single mailbox In-Reply-To: <8D617AEC-0038-4B4F-B393-C95DCD737F96@balius.com> References: <55CB7B42.104@univention.de> <8D617AEC-0038-4B4F-B393-C95DCD737F96@balius.com> Message-ID: <9999C211-0CA7-4F23-851A-CB36DA10B685@balius.com> The problem happened again this morning. Removing fsync calls helped, but I'm not sure about leaving that enabled long term. I still believe the problem is multiple dovecot processes trying to write to a single folder at the same time. (If I could run dtrace I might be able to cobble together a script to prove it.) I tried writing a sieve script to direct the messages to a set of folders, but I'm not able to make the logic work. I was thinking something like: generate random # ($N) between say 1-10, then file message into folder mail$N. But I didn't find a method to do that within sieve. My next thought was to try parsing the message-id header. If the first character is [0-5] then fileinto mail1, etc. Then I could go so far as having 36 subfolders which the messages could be written too. This mailbox only keeps messages for a rolling 1d window. Right now for example it has 260,186 messages in the INBOX. The sieve script I tried (with only about 4 hours of sleep) was require ["fileinto","regex"]; if header :regex "message-id" "^1" { fileinto "mail1"; } else { keep; } If anyone has some suggestions on how I might spread the messages out over multiple folders I'd like to hear your thoughts. Again the servers are configured using maildir, so each folder should have its own index and thus file locking contention should be less, at least so goes the theory in my head. Thank you, Chad From kremels at kreme.com Fri Aug 14 14:56:32 2015 From: kremels at kreme.com (@lbutlr) Date: Fri, 14 Aug 2015 08:56:32 -0600 Subject: Error: User initialization failed In-Reply-To: References: Message-ID: <46885658-010B-4F0C-9996-B5C47B46DDB5@kreme.com> On Aug 14, 2015, at 8:51 AM, @lbutlr wrote: > dovecot: lda(chance at example.com): Error: User initialization failed: Namespace '': Ambiguous mail location setting, don't know what to do with it: /usr/local/virtual/chance@ example.com (try prefixing it with mbox: or milder:) Here?s the dovecot -n output $ dovecot -n # 2.2.18: /usr/local/etc/dovecot/dovecot.conf # OS: FreeBSD 9.2-RELEASE-p3 i386 auth_failure_delay = 5 secs auth_mechanisms = PLAIN LOGIN default_process_limit = 200 disable_plaintext_auth = no first_valid_uid = 89 login_log_format_elements = user=<%u> %r %m %c mail_location = maildir:~/Maildir mail_max_userip_connections = 90 namespace inbox { inbox = yes location = mailbox Drafts { special_use = \Drafts } mailbox Junk { auto = subscribe special_use = \Junk } mailbox NotJunk { auto = subscribe } mailbox Sent { special_use = \Sent } mailbox "Sent Messages" { special_use = \Sent } mailbox Trash { special_use = \Trash } prefix = } passdb { driver = pam } passdb { args = /etc/dovecot/dovecot-sql.conf.ext driver = sql } protocols = imap service auth { unix_listener /var/spool/postfix/private/auth { mode = 0666 } } service imap-login { inet_listener imaps { port = 993 ssl = yes } } ssl_cert = References: <55C9A5A8.9010501@echter-kuechen-elektro.de> <55CC2C0F.2010709@granch.ru> Message-ID: <55CE2675.3070003@echter-kuechen-elektro.de> Am 13.08.2015 um 07:33 schrieb ?????? ????? ??????????????: > 11/08/15 13:35, J. Echter ?????: >> Hi, >> >> i'd like to convert from maildir to mdbox. >> >> what would be the best way to do this? >> >> i have read the dsync part of the dovecot wiki but i don't get what i >> have to use for my setup. > > Sometime, when I must converted many user's mailboxes from mbox to > mdbox, I had have wrote one script to mass convert mailboxes for local > or domain users and can provide it for you. It support mbox and > maildir as source formats. > Hi, thanks for your reply. Would be nice to have a look at this script :) Thanks for your offer. Juergen From ad+lists at uni-x.org Fri Aug 14 17:57:02 2015 From: ad+lists at uni-x.org (Alexander Dalloz) Date: Fri, 14 Aug 2015 19:57:02 +0200 Subject: Multidomain / IP Address Setup (Dovevot 2.2.10 on CentOS7 ) is failing: Fatal: Couldn't parse private ssl_key: error:0906D06C:PEM In-Reply-To: <55CDCF90.8050207@gmx.net> References: <55CDCF90.8050207@gmx.net> Message-ID: <55CE2BEE.7060806@uni-x.org> Am 14.08.2015 um 13:22 schrieb dravion.smith at gmx.net: > Hi, > I want to say hello and here is my big problem ;D > > Iam trying to archive a Postfix/Dovecot 2.2.10 CentOS7 Multidomain Setup > with multiple (valid > StartSSL Certs), but iam only able to run a single Domain Cert server only. > > ps: I need a multiple domainssetup for every customer and it is not an > option for me redirecting any email > to a single domain server. I really need this setup working. > > IMHO: I think it SELinux could interfere with multiple Certs in diffrent > folders (it is activated in CentOs7 by > default and is needed by other apps) What have you done to exclude that SELinux interferes? Run "ausearch -m avc" to check for AVCs. > Ok, here is my logfile data: > > systemctl start postfix.service [OK] > systemctl start dovecot.service [OK] > > /var/log/messages > *systemd: Stopping Dovecot IMAP/POP3 email server... > *systemd: Starting Dovecot IMAP/POP3 email server... > *systemd: Started Dovecot IMAP/POP3 email server. > > /var/log/maillog > *dovecot: master: Dovecot v2.2.10 starting up for imap, pop3, lmtp (core > dumps disabled) > > ### This works (Thunderbird, Outlook 2013, Opera Mail ect.) #### > > local mydomain01.tld { > > protocol imap { > ssl_cert = > ssl_key = > } > > } You are leaving the terrain of your distribution. That's not the intended path. /etc/pki/tls/{certs,private}/ is. > ### this 10-ssl.conf ### --- FAILS (the error occurs after an email > client accesses IMAP Folders) > local mydomain01.tld { > > protocol imap { > ssl_cert = > ssl_key = > } > > } > > local mydomain02.tld { > > protocol imap { > ssl_cert = > ssl_key = > } > > } See above. Why 2 times the same certificate pair files? Make sure the permissions (and not only of the files itself) and the SELinux context is set properly. You gave zero information about that. > /var/log/mailog ### Error log ### > Aug 14 12:50:38 matrix dovecot: imap-login: Fatal: Couldn't parse > private ssl_key: error:0906D06C:PEM routines:PEM_read_bio:no start line: > Aug 14 12:50:38 matrix dovecot: master: Error: service(imap-login): > command startup failed, throttling for 60 secs The key file contains "-----BEGIN PRIVATE KEY-----" as first line and "-----END PRIVATE KEY-----" as last line? > I really dont know why a single domain is no problem but if i enable > multiple domains dovecots > start with any error, even if i set debug verbose leven to extrem high > but if i access dovecot with > Thunderbird my server loggile explodes with something like this Couldn't > parse private ssl_key: error:0906D06C:PEM but the certs are 100% valid > and checked over and over again. > > Any help is greatly appreciated! > > Greetings, > Dravion Alexander From andrzej.filip at gmail.com Fri Aug 14 18:01:04 2015 From: andrzej.filip at gmail.com (Andrzej A. Filip) Date: Fri, 14 Aug 2015 20:01:04 +0200 Subject: multiple messages per second to a single mailbox In-Reply-To: <84C57ADD-1F76-4276-9DDD-B5B347600A80@balius.com> (Chad M. Stewart's message of "Wed, 12 Aug 2015 11:25:56 -0500") References: <84C57ADD-1F76-4276-9DDD-B5B347600A80@balius.com> Message-ID: Chad M Stewart wrote: > On Aug 12, 2015, at 11:04 AM, Andrzej A. Filip wrote: >> >> > > <..snip..> > >> Could you provide the following info: >> a) mailbox type (maildir/mbox/dbox/...) > > maildir > >> [mail_location in dovecot's config] > > /srv/mail/// > >> b) file system type (ext2/ext3/ext4/fat32/...) >> [provided by "df -T" command on my system] > > As I said ext4. > > Since I posted I've changed a couple of things: ulimit -n 8192, and > disabled fsync as in mail_fsync = never. I'm not sure if I'll put it > back in the LMTP section or not. Given all the hardware abstraction > layers. Are docecot and postfix located on the same server? Can postfix access (deliver) directly maildir file directory dovecot uses? -- A. Filip From cms at balius.com Fri Aug 14 21:04:28 2015 From: cms at balius.com (Chad M Stewart) Date: Fri, 14 Aug 2015 16:04:28 -0500 Subject: multiple messages per second to a single mailbox In-Reply-To: References: <84C57ADD-1F76-4276-9DDD-B5B347600A80@balius.com> Message-ID: <83DDA393-5FA9-4BB3-BF7E-56372C107934@balius.com> On Aug 14, 2015, at 1:01 PM, Andrzej A. Filip wrote: > > > Are docecot and postfix located on the same server? > Can postfix access (deliver) directly maildir file directory dovecot uses? > For the moment yes they are on the same server. I designed it to be modular, the various components can be placed on different systems with no configuration changes required, should it be necessary. I'll note that Postfix does not appear to have any problems writing the files to its queue as fast as they are being delivered to it. Postfix and Dovecot are writing to different disks, though both are on the SAN, and both have noatime set. -Chad From andrzej.filip at gmail.com Fri Aug 14 21:20:04 2015 From: andrzej.filip at gmail.com (Andrzej A. Filip) Date: Fri, 14 Aug 2015 23:20:04 +0200 Subject: multiple messages per second to a single mailbox In-Reply-To: <83DDA393-5FA9-4BB3-BF7E-56372C107934@balius.com> (Chad M. Stewart's message of "Fri, 14 Aug 2015 16:04:28 -0500") References: <84C57ADD-1F76-4276-9DDD-B5B347600A80@balius.com> <83DDA393-5FA9-4BB3-BF7E-56372C107934@balius.com> Message-ID: Chad M Stewart wrote: > On Aug 14, 2015, at 1:01 PM, Andrzej A. Filip wrote: > >> >> >> Are docecot and postfix located on the same server? >> Can postfix access (deliver) directly maildir file directory dovecot uses? >> > > For the moment yes they are on the same server. I designed it to be > modular, the various components can be placed on different systems > with no configuration changes required, should it be necessary. > > I'll note that Postfix does not appear to have any problems writing > the files to its queue as fast as they are being delivered to it. > Postfix and Dovecot are writing to different disks, though both are on > the SAN, and both have noatime set. As I understand: * maildir is designed for allowing fast "lockless" parallel deliveries * dovecot's LDA updates some dovecot specific index/cache files even for deliveries to maildir => making postfix deliver to maildir without updating dovecot specific files may solve your performance problems -- A. Filip From mysqlstudent at gmail.com Sat Aug 15 00:08:13 2015 From: mysqlstudent at gmail.com (Alex) Date: Fri, 14 Aug 2015 20:08:13 -0400 Subject: IMAP folder separator issue Message-ID: Hi, I'm using dovecot-2.2.18 on fedora22 and having a problem with Thunderbird. I reported this problem some time ago, and received some tips, but no solution that's worked. I'm now trying to fix this in earnest, and having problems. The problem is sub-folders are created with a caret in them, instead of a new folder below the parent folder. So instead of folder/sub-folder, folder^sub-folder is created. I'm using mbox folders, and it worked under previous versions of fedora and dovecot. I believe it has to do with the separator character that's used, based on this wiki page: http://wiki2.dovecot.org/Plugins/Listescape The problem is that those instructions cause errors: Aug 14 20:00:01 orion dovecot: pop3(cerberus): Error: User initialization failed: namespace configuration error: Duplicate namespace prefix: "" Aug 14 20:00:01 orion dovecot: pop3(cerberus): Error: Invalid user settings. Refer to server log for more information. I've also had to change the "namespace private" to "namespace { type=private }" as that's apparently a deprecated syntax. Perhaps there's an updated page with more instruction and correct information? I'm unsure where this duplicate namespace is that it's talking about... I've included my doveconf output below, in hopes someone can help. I should also point out that "separator = $" is quoted in the original doveconf.conf but apparently doveconf doesn't properly print it. # 2.2.18: /etc/dovecot/dovecot.conf # OS: Linux 4.0.7-300.fc22.x86_64 x86_64 Fedora release 22 (Twenty Two) default_client_limit = 2000 mail_gid = mail mail_location = mbox:~/mail:INBOX=/var/spool/mail/%u mail_plugins = " listescape" mail_privileged_group = mail mbox_write_locks = fcntl namespace { inbox = yes location = prefix = separator = $ type = private } namespace inbox { inbox = yes location = mailbox Drafts { special_use = \Drafts } mailbox Junk { special_use = \Junk } mailbox Sent { special_use = \Sent } mailbox "Sent Messages" { special_use = \Sent } mailbox Trash { special_use = \Trash } prefix = } passdb { driver = pam } passdb { driver = pam } protocols = imap pop3 service auth { unix_listener /var/spool/postfix/private/auth { group = postfix mode = 0660 user = postfix } } service imap-login { inet_listener imap { port = 0 } inet_listener imaps { port = 993 } process_min_avail = 20 service_count = 0 } service pop3-login { inet_listener pop3 { port = 0 } inet_listener pop3s { address = 192.168.1.100 port = 995 } } ssl_ca = References: <55CDCF90.8050207@gmx.net> <55CE2BEE.7060806@uni-x.org> Message-ID: <55CE9467.40004@gmx.net> Hello Alex Am 14.08.2015 um 19:57 schrieb Alexander Dalloz: > What have you done to exclude that SELinux interferes? > Just some sysinfo: CentOS Linux release 7.1.1503 (Core) (i run yum update every day) sestatus: SELinux status: enabled SELinuxfs mount: /sys/fs/selinux SELinux root directory: /etc/selinux Loaded policy name: targeted Current mode: enforcing Mode from config file: enforcing Policy MLS status: enabled Policy deny_unknown status: allowed Max kernel policy version: 28 > > Run "ausearch -m avc" to check for AVCs. There is no indication SELinux is blocking somewhat grep "SELinux is preventing" /var/log/messages grep "denied" /var/log/audit/audit.log ausearch -m avc shows no deniead messages > >> ### This works (Thunderbird, Outlook 2013, Opera Mail ect.) #### >> >> local mydomain01.tld { >> >> protocol imap { >> ssl_cert = >> > ssl_key = >> > } >> >> } >> Sorry the above has some typo errors, forget it. Ok, this works well: Configfile: /etc/dovecot/conf.d/10-ssl.conf protocol imap { ssl_cert = References: <55CDCF90.8050207@gmx.net> <55CE2BEE.7060806@uni-x.org> <55CE9467.40004@gmx.net> Message-ID: <55CE962F.4010500@gmx.net> ### CORRECTION Am 15.08.2015 um 03:22 schrieb dravion.smith at gmx.net: > > #### BUT #### > If i try something like this in /etc/dovecot/conf.d/10-ssl.conf > > local imap.mydomain01.tld { > protocol imap { > ssl_cert = > ssl_key = > } > } > > local imap.mydomain02.tld { > protocol imap { > ssl_cert = > ssl_key = > } > } > > It throws errors like this > "imap-login "parse private ssl_key: error:0906D06C:PEM" > in the logfile /var/maillog From edgar at pettijohn-web.com Sat Aug 15 01:50:31 2015 From: edgar at pettijohn-web.com (Edgar Pettijohn) Date: Fri, 14 Aug 2015 20:50:31 -0500 Subject: Multidomain / IP Address Setup (Dovevot 2.2.10 on CentOS7 ) is failing: Fatal: Couldn't parse private ssl_key: error:0906D06C:PEM In-Reply-To: <55CE962F.4010500@gmx.net> References: <55CDCF90.8050207@gmx.net> <55CE2BEE.7060806@uni-x.org> <55CE9467.40004@gmx.net> <55CE962F.4010500@gmx.net> Message-ID: <55CE9AE7.4020506@pettijohn-web.com> On 08/14/15 20:30, dravion.smith at gmx.net wrote: > > ### CORRECTION > Am 15.08.2015 um 03:22 schrieb dravion.smith at gmx.net: >> >> #### BUT #### >> If i try something like this in /etc/dovecot/conf.d/10-ssl.conf >> local_name imap.mydomain01.tld >> local imap.mydomain01.tld { >> protocol imap { >> ssl_cert = >> > ssl_key = >> > } >> } >> local_name imap.mydomain02.tld >> local imap.mydomain02.tld { >> protocol imap { >> ssl_cert = >> > ssl_key = >> > } >> } >> >> It throws errors like this >> "imap-login "parse private ssl_key: error:0906D06C:PEM" >> in the logfile /var/maillog If they are separate ip's why not try that? local x.x.x.x { etc.. http://wiki2.dovecot.org/SSL/DovecotConfiguration From dravion.smith at gmx.net Sat Aug 15 02:24:44 2015 From: dravion.smith at gmx.net (dravion.smith at gmx.net) Date: Sat, 15 Aug 2015 04:24:44 +0200 Subject: Multidomain / IP Address Setup (Dovevot 2.2.10 on CentOS7 ) is failing: Fatal: Couldn't parse private ssl_key: error:0906D06C:PEM In-Reply-To: <55CE9AE7.4020506@pettijohn-web.com> References: <55CDCF90.8050207@gmx.net> <55CE2BEE.7060806@uni-x.org> <55CE9467.40004@gmx.net> <55CE962F.4010500@gmx.net> <55CE9AE7.4020506@pettijohn-web.com> Message-ID: <55CEA2EC.1080203@gmx.net> Hi Ed, Am 15.08.2015 um 03:50 schrieb Edgar Pettijohn: > If they are separate ip's why not try that? > > local x.x.x.x { etc.. > > http://wiki2.dovecot.org/SSL/DovecotConfiguration This was my main source for configuring it. Yes, there are diffrent IPv6/64 Bit Addresses and dig mx resolves the correct domains to diffent aaaa addresses counterpart (imap.mydomain01.tld resolves to a diffrent IP then imap.mydomain02.tld and the corresponding StartSSL certs are correctly issued to imap.mydomain01.tld and imap.mydomain02.tld) IP as Host resolution and even SNI Hostresolution (for newer E-Mail-Client Programs) are leading allways to the same Error, if i enable multidomain hosting/multi IP Hosting. Only single Domain Hosting works with the its own certificate, thadts why iam confident the Certs are ok. The imap-login process and/or OpenSSL access by imap-login is simply failing if enable multidomain hosting with Couldn't parse private ssl_key: error:0906D06C:PEM routines:PEM_read_bio:no start line: Expecting: ANY PRIVATE KEY As far as my research goes the debian folks have a bug report filed under https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=771334 Drav From edgar at pettijohn-web.com Sat Aug 15 05:08:57 2015 From: edgar at pettijohn-web.com (Edgar Pettijohn) Date: Sat, 15 Aug 2015 00:08:57 -0500 Subject: Multidomain / IP Address Setup (Dovevot 2.2.10 on CentOS7 ) is failing: Fatal: Couldn't parse private ssl_key: error:0906D06C:PEM In-Reply-To: <55CE962F.4010500@gmx.net> References: <55CDCF90.8050207@gmx.net> <55CE2BEE.7060806@uni-x.org> <55CE9467.40004@gmx.net> <55CE962F.4010500@gmx.net> Message-ID: <55CEC969.2010302@pettijohn-web.com> I think you should post doveconf -n output. On 08/14/15 20:30, dravion.smith at gmx.net wrote: > > ### CORRECTION > Am 15.08.2015 um 03:22 schrieb dravion.smith at gmx.net: >> >> #### BUT #### >> If i try something like this in /etc/dovecot/conf.d/10-ssl.conf >> >> local imap.mydomain01.tld { >> protocol imap { >> ssl_cert = >> > ssl_key = >> > } >> } >> >> local imap.mydomain02.tld { >> protocol imap { >> ssl_cert = >> > ssl_key = >> > } >> } >> >> It throws errors like this >> "imap-login "parse private ssl_key: error:0906D06C:PEM" >> in the logfile /var/maillog From superinterstellar at gmail.com Sat Aug 15 05:54:13 2015 From: superinterstellar at gmail.com (Kevin Laurie) Date: Sat, 15 Aug 2015 13:54:13 +0800 Subject: backing up email / saving maildir on external hard drives In-Reply-To: References: <20150809114449.7b5a273f@mydesq2.domain.cxm> <55C9FC7D.4080201@filmakademie.de> <55CA1B25.7080100@chemie.uni-hamburg.de> Message-ID: Dear everyone, I got it taken care of. Just copied the maildir to a standard FAT drive. I dont know why I was making it so complicated. Thanks tho! Appreciate it. On Wed, Aug 12, 2015 at 12:32 AM, G?tz Reinicke wrote: > >> Am 11.08.2015 um 17:56 schrieb Christian Schmidt : >> >> On 11.08.2015 15:45, G?tz Reinicke - IT Koordinator wrote: >>> talking of rsync and compression is may be also a bit misleading. >>> >>> On the destination there will be no compressed files if you transfer >>> with rsync! >> >> ...unless your dovecot is working with compressed maildir files. ;-) >> > > I was talking about the rsync file transfer, which will not compress the output on the backup server?. not about anything dovecote has configured ? may be uhh he uses a filesystem compression too ? Or some rsync piping trough gzip? > > kidding :) > > /G?tz From ml+dovecot at valo.at Sat Aug 15 06:16:00 2015 From: ml+dovecot at valo.at (Christian Kivalo) Date: Sat, 15 Aug 2015 08:16:00 +0200 Subject: Multidomain / IP Address Setup (Dovevot 2.2.10 on CentOS7 ) is failing: Fatal: Couldn't parse private ssl_key: error:0906D06C:PEM In-Reply-To: <55CEC969.2010302@pettijohn-web.com> References: <55CDCF90.8050207@gmx.net> <55CE2BEE.7060806@uni-x.org> <55CE9467.40004@gmx.net> <55CE962F.4010500@gmx.net> <55CEC969.2010302@pettijohn-web.com> Message-ID: <08698357-81F6-465C-B601-D3F2002E7565@valo.at> >I think you should post doveconf -n output. with your dovecot -n output provided this mail would not need to be sent >On 08/14/15 20:30, dravion.smith at gmx.net wrote: >> >> ### CORRECTION >> Am 15.08.2015 um 03:22 schrieb dravion.smith at gmx.net: >>> >>> #### BUT #### >>> If i try something like this in /etc/dovecot/conf.d/10-ssl.conf >>> >>> local imap.mydomain01.tld { >>> protocol imap { >>> ssl_cert = >>> >> ssl_key = >>> >> } >>> } >>> >>> local imap.mydomain02.tld { >>> protocol imap { >>> ssl_cert = >>> >> ssl_key = >>> >> } >>> } >>> >>> It throws errors like this >>> "imap-login "parse private ssl_key: error:0906D06C:PEM" >>> in the logfile /var/maillog from the dovecot from the dovecot ssl wiki page http://wiki2.dovecot.org/SSL/DovecotConfiguration Multiple SSL certificates Different certificates per IP and protocol If you have multiple IPs available, this method is guaranteed to work with all clients. [snip] Note that you will still need a top-level "default"?ssl_key?and?ssl_cert?as well, or you will receive errors. in addition to your two domain specific ssl certs have you also defined a "default" ssl_key and ssl_cert as required as required by the documentation? regards - c From ml+dovecot at valo.at Sat Aug 15 06:34:14 2015 From: ml+dovecot at valo.at (Christian Kivalo) Date: Sat, 15 Aug 2015 08:34:14 +0200 Subject: IMAP folder separator issue In-Reply-To: References: Message-ID: ># 2.2.18: /etc/dovecot/dovecot.conf ># OS: Linux 4.0.7-300.fc22.x86_64 x86_64 Fedora release 22 (Twenty Two) >default_client_limit = 2000 >mail_gid = mail >mail_location = mbox:~/mail:INBOX=/var/spool/mail/%u >mail_plugins = " listescape" >mail_privileged_group = mail >mbox_write_locks = fcntl >namespace { > inbox = yes > location = > prefix = > separator = $ > type = private >} >namespace inbox { > inbox = yes > location = > mailbox Drafts { > special_use = \Drafts > } the namespaces wiki page at http://wiki2.dovecot.org/Namespaces says you should have srt inbox = yes only once: inbox: "yes", if this namespace contains the user's INBOX. There is only one INBOX, so only one namespace can have inbox=yes. you have set inbox=yes twice. - c From dravion.smith at gmx.net Sat Aug 15 06:58:04 2015 From: dravion.smith at gmx.net (dravion.smith at gmx.net) Date: Sat, 15 Aug 2015 08:58:04 +0200 Subject: Multidomain / IP Address Setup (Dovevot 2.2.10 on CentOS7 ) is failing: Fatal: Couldn't parse private ssl_key: error:0906D06C:PEM In-Reply-To: <08698357-81F6-465C-B601-D3F2002E7565@valo.at> References: <55CDCF90.8050207@gmx.net> <55CE2BEE.7060806@uni-x.org> <55CE9467.40004@gmx.net> <55CE962F.4010500@gmx.net> <55CEC969.2010302@pettijohn-web.com> <08698357-81F6-465C-B601-D3F2002E7565@valo.at> Message-ID: <55CEE2FC.3030909@gmx.net> Am 15.08.2015 um 08:16 schrieb Christian Kivalo: > [snip] > > Note that you will still need a top-level "default" ssl_key and ssl_cert as well, or you will receive errors. > > in addition to your two domain specific ssl certs have you also defined a "default" ssl_key and ssl_cert as required as required by the documentation? > > regards > - c Did you really read the wiki and thadt i allready said it was my main source?? 1) Domains works flawless 2) If i change the domainnames and certificates it works flawless but 3) If i try (like described in the Wiki you posted) i get this "imap-login "parse private ssl_key: error:0906D06C:PEM" Its the fucking imap-login process screwing things up without any reason if you try to configure it like described in the damm wiki! From ml+dovecot at valo.at Sat Aug 15 07:04:36 2015 From: ml+dovecot at valo.at (Christian Kivalo) Date: Sat, 15 Aug 2015 09:04:36 +0200 Subject: Multidomain / IP Address Setup (Dovevot 2.2.10 on CentOS7 ) is failing: Fatal: Couldn't parse private ssl_key: error:0906D06C:PEM In-Reply-To: <55CEE2FC.3030909@gmx.net> References: <55CDCF90.8050207@gmx.net> <55CE2BEE.7060806@uni-x.org> <55CE9467.40004@gmx.net> <55CE962F.4010500@gmx.net> <55CEC969.2010302@pettijohn-web.com> <08698357-81F6-465C-B601-D3F2002E7565@valo.at> <55CEE2FC.3030909@gmx.net> Message-ID: Am 15. August 2015 08:58:04 MESZ, schrieb "dravion.smith at gmx.net" : > >Am 15.08.2015 um 08:16 schrieb Christian Kivalo: > >> [snip] >> >> Note that you will still need a top-level "default" ssl_key and >ssl_cert as well, or you will receive errors. >> >> in addition to your two domain specific ssl certs have you also >defined a "default" ssl_key and ssl_cert as required as required by the >documentation? >> >> regards >> - c > >Did you really read the wiki and thadt i allready said it was my main >source?? > >1) Domains works flawless >2) If i change the domainnames and certificates it works flawless >but >3) If i try (like described in the Wiki you posted) i get this > >"imap-login "parse private ssl_key: error:0906D06C:PEM" > >Its the fucking imap-login process screwing things up without any >reason if you try to configure it like >described in the damm wiki! provide your multi ssl doveconf -n output. - c From dravion.smith at gmx.net Sat Aug 15 07:06:44 2015 From: dravion.smith at gmx.net (dravion.smith at gmx.net) Date: Sat, 15 Aug 2015 09:06:44 +0200 Subject: Multidomain / IP Address Setup (Dovevot 2.2.10 on CentOS7 ) is failing: Fatal: Couldn't parse private ssl_key: error:0906D06C:PEM In-Reply-To: References: <55CDCF90.8050207@gmx.net> <55CE2BEE.7060806@uni-x.org> <55CE9467.40004@gmx.net> <55CE962F.4010500@gmx.net> <55CEC969.2010302@pettijohn-web.com> <08698357-81F6-465C-B601-D3F2002E7565@valo.at> <55CEE2FC.3030909@gmx.net> Message-ID: <55CEE504.20606@gmx.net> Am 15.08.2015 um 09:04 schrieb Christian Kivalo: > provide your multi ssl doveconf -n output. - c No. I leave this shit alone and running dovecot in multiinstance mode and now its works. From djk at tobit.co.uk Sat Aug 15 11:19:56 2015 From: djk at tobit.co.uk (djk) Date: Sat, 15 Aug 2015 12:19:56 +0100 Subject: ipv4 imap not accepting connections in mixed ipv4/ipv6 Message-ID: <55CF205C.2060707@tobit.co.uk> I have two installations of dovecot 2.18, apart from the necessary changes to allow them to replicate amongst themselves the configs are identical. Both are running Ubuntu 14.04 and are kept up to date. One is running 64bit and the other 32 bit (for historical reasons). Both have ipv4 and ipv6 addresses. The 64 bit installation allows access on port 143 and 993 on both ipv4 and ipv6. The 32 bit installation does not allow access on 143 on ipv4, but allows ipv4/ipv6 on 993. Both installations happily allow connections to any other ports that dovecot might be using on both ipv4/ipv6, as well as everything else that is running (on other ports) on those machines. There are firewalls, but no warning messages. Switching firewalls off makes no difference. Tcpdump shows the SYN packets arriving on the (correct) interface. Any suggestions? Dirk -------------- next part -------------- # 2.2.18: /etc/dovecot/dovecot.conf # Pigeonhole version 0.4.8 (0c4ae064f307+) # OS: Linux 3.13.0-61-generic i686 Ubuntu 14.04.3 LTS auth_verbose = yes doveadm_password = %Db234A!&., at vc$ first_valid_uid = 2000 imapc_features = rfc822.size fetch-headers imapc_host = post.tobit.co.uk imapc_port = 993 imapc_ssl = imaps imapc_ssl_verify = no imapc_user = %n lda_mailbox_autocreate = yes login_greeting = IMAP ready. mail_attachment_dir = %h/attachments mail_attachment_min_size = 64 k mail_location = mdbox:%h:INBOX=%h/mail mail_plugins = " fts fts_lucene notify replication" mail_prefetch_count = 20 mailbox_list_index = yes managesieve_notify_capability = mailto managesieve_sieve_capability = fileinto reject envelope encoded-character vacation subaddress comparator-i;ascii-numeric relational regex imap4flags copy include variables body enotify environment mailbox date index ihave duplicate mdbox_rotate_size = 10 M namespace inbox { inbox = yes location = mailbox Drafts { special_use = \Drafts } mailbox Junk { special_use = \Junk } mailbox Sent { special_use = \Sent } mailbox "Sent Messages" { special_use = \Sent } mailbox Trash { special_use = \Trash } prefix = } passdb { args = /etc/dovecot/dovecot-sql.conf.ext driver = sql } plugin { fts = lucene fts_lucene = whitespace_chars=@. mail_replica = tcps: sieve = file:~/sieve;active=~/.dovecot.sieve } postmaster_address = postmaster at tobit.co.uk protocols = " imap lmtp sieve sieve" service aggregator { fifo_listener replication-notify-fifo { mode = 0666 user = vmail } unix_listener replication-notify { mode = 0666 user = vmail } } service auth { unix_listener exim-auth { group = Debian-exim mode = 0660 user = Debian-exim } } service config { unix_listener config { mode = 0666 user = vmail } } service doveadm { inet_listener { port = 23999 ssl = yes } } service imap-login { inet_listener imap { port = 143 } inet_listener imaps { port = 993 ssl = yes } } service lmtp { inet_listener lmtp { address = 127.0.0.1 port = } unix_listener lmtp { mode = 0666 } user = vmail } service managesieve-login { inet_listener sieve { port = } process_min_avail = 0 service_count = 1 vsz_limit = 64 M } service managesieve { process_limit = 100 } service replicator { process_min_avail = 1 unix_listener replicator-doveadm { mode = 0600 user = vmail } } ssl_cert = <[key file] ssl_client_ca_dir = /etc/ssl/certs ssl_key = <[key file] ssl_protocols = !SSLv3 !SSLv2 userdb { args = /etc/dovecot/dovecot-sql.conf.ext driver = sql } protocol lmtp { mail_plugins = sieve } From dovecot at knutejohnson.com Sat Aug 15 21:26:21 2015 From: dovecot at knutejohnson.com (Knute Johnson) Date: Sat, 15 Aug 2015 14:26:21 -0700 Subject: Syncing two mail servers? Message-ID: <55CFAE7D.9080302@knutejohnson.com> I currently have a mail server using Dovecot and sendmail. I was thinking I would like to have a backup of my mail and an additional server in case of hardware failure. Is it possible to run two identical or nearly identical server setups and sync the mail files between them? If it is possible, in general terms how would one go about this? Thanks, -- Knute Johnson From eduardo at freedominterface.org Sat Aug 15 23:03:05 2015 From: eduardo at freedominterface.org (Eduardo Ramos) Date: Sat, 15 Aug 2015 20:03:05 -0300 Subject: Syncing two mail servers? In-Reply-To: <55CFAE7D.9080302@knutejohnson.com> References: <55CFAE7D.9080302@knutejohnson.com> Message-ID: <55CFC529.20600@freedominterface.org> For sure Knute. Take a look at this: http://wiki2.dovecot.org/Replication I think a good implementation would be using a dovecot director layer + dovecot replication. On 08/15/2015 06:26 PM, Knute Johnson wrote: > I currently have a mail server using Dovecot and sendmail. I was > thinking I would like to have a backup of my mail and an additional > server in case of hardware failure. Is it possible to run two > identical or nearly identical server setups and sync the mail files > between them? If it is possible, in general terms how would one go > about this? > > Thanks, > From edgar at pettijohn-web.com Sun Aug 16 00:46:24 2015 From: edgar at pettijohn-web.com (Edgar Pettijohn) Date: Sat, 15 Aug 2015 19:46:24 -0500 Subject: Multidomain / IP Address Setup (Dovevot 2.2.10 on CentOS7 ) is failing: Fatal: Couldn't parse private ssl_key: error:0906D06C:PEM In-Reply-To: <55CEE504.20606@gmx.net> References: <55CDCF90.8050207@gmx.net> <55CE2BEE.7060806@uni-x.org> <55CE9467.40004@gmx.net> <55CE962F.4010500@gmx.net> <55CEC969.2010302@pettijohn-web.com> <08698357-81F6-465C-B601-D3F2002E7565@valo.at> <55CEE2FC.3030909@gmx.net> <55CEE504.20606@gmx.net> Message-ID: <55CFDD60.30404@pettijohn-web.com> /etc/ssl $ sudo doveconf -n # 2.2.15: /etc/dovecot/dovecot.conf # OS: OpenBSD 5.7 amd64 ffs auth_mechanisms = plain login default_client_limit = 500 disable_plaintext_auth = no first_valid_uid = 1000 imap_client_workarounds = delay-newmail tb-extra-mailbox-sep tb-lsub-flags mail_location = maildir:/var/vmail/%d/%n/Maildir managesieve_notify_capability = mailto managesieve_sieve_capability = fileinto reject envelope encoded-character vacation subaddress comparator-i;ascii-numeric relational regex imap4flags copy include variables body enotify environment mailbox date ihave duplicate mbox_write_locks = fcntl mmap_disable = yes namespace inbox { inbox = yes location = mailbox Drafts { special_use = \Drafts } mailbox Junk { special_use = \Junk } mailbox Sent { special_use = \Sent } mailbox "Sent Messages" { special_use = \Sent } mailbox Trash { special_use = \Trash } prefix = } passdb { args = scheme=BLF-CRYPT username_format=%n /etc/mail/users driver = passwd-file } plugin { sieve = file:~/sieve;active=~/.dovecot.sieve } pop3_client_workarounds = outlook-no-nuls oe-ns-eoh postmaster_address = postmaster@%d protocols = imap pop3 lmtp sieve sieve service auth { unix_listener auth-userdb { group = _smtpd mode = 0666 user = _smtpd } } service imap-login { inet_listener imap { port = 143 } inet_listener imaps { port = 993 ssl = yes } } service lmtp { unix_listener lmtp { mode = 0666 } } service managesieve-login { inet_listener sieve { port = 4190 } inet_listener sieve_deprecated { port = 2000 } } service pop3-login { inet_listener pop3 { port = 110 } inet_listener pop3s { port = 995 ssl = yes } } ssl_cert = > > Am 15.08.2015 um 09:04 schrieb Christian Kivalo: >> provide your multi ssl doveconf -n output. - c > > No. I leave this shit alone and running dovecot in multiinstance mode > and now its works. From dravion.smith at gmx.net Sun Aug 16 02:33:25 2015 From: dravion.smith at gmx.net (dravion.smith at gmx.net) Date: Sun, 16 Aug 2015 04:33:25 +0200 Subject: Multidomain / IP Address Setup (Dovevot 2.2.10 on CentOS7 ) is failing: Fatal: Couldn't parse private ssl_key: error:0906D06C:PEM In-Reply-To: <55CFDD60.30404@pettijohn-web.com> References: <55CDCF90.8050207@gmx.net> <55CE2BEE.7060806@uni-x.org> <55CE9467.40004@gmx.net> <55CE962F.4010500@gmx.net> <55CEC969.2010302@pettijohn-web.com> <08698357-81F6-465C-B601-D3F2002E7565@valo.at> <55CEE2FC.3030909@gmx.net> <55CEE504.20606@gmx.net> <55CFDD60.30404@pettijohn-web.com> Message-ID: <55CFF675.5070406@gmx.net> Hi Ed, Interresting setup. I like the way you striped it all down to just one single file :-) But can you explain why you use globally: ssl_cert = (dedicated IPv6 Address but had no lock Configinfo: 1) MTA (Postfix 2.10.1) and MDA (Dovecot 2.2.10) configured on IPv6 Addresses 2) The MTA and MDA are connected to MariaDB (the default MySQL replacement on CentOS7 now for virtual domains, users, passwords, aliaes ect 3) Postfix uses Dovecots SASL Implementation and Postfix and Dovecot talking via LMTP and UNIX Sockets. Details: ### yum info postfix ### Name : postfix Arch : x86_64 Epoche : 2 Version : 2.10.1 Release : 6.el7 Size : 12 M >From : installed From Source : debian.n-ix.net_centos_7_os_x86_64_ Summary : Postfix Mail Transport Agent URL : http://www.postfix.org License : IBM and GPLv2+ Description : Postfix is a Mail Transport Agent (MTA), supporting LDAP, SMTP AUTH (SASL), TLS ### yum info dovecot ### Name : dovecot Arch : i686 Epoche : 1 Version : 2.2.10 Ausgabe : 4.el7_0.1 Gr??e : 3.2 M Quelle : debian.n-ix.net_centos_7_os_x86_64_ Summary : Secure imap and pop3 server URL : http://www.dovecot.org/ Lizenz : MIT and LGPLv2 Description : Dovecot is an IMAP server for Linux/UNIX-like systems, written with security : primarily in mind. It also contains a small POP3 server. It supports mail : in either of maildir or mbox formats. Cheers, Drav From ml+dovecot at valo.at Sun Aug 16 05:31:11 2015 From: ml+dovecot at valo.at (Christian Kivalo) Date: Sun, 16 Aug 2015 07:31:11 +0200 Subject: Multidomain / IP Address Setup (Dovevot 2.2.10 on CentOS7 ) is failing: Fatal: Couldn't parse private ssl_key: error:0906D06C:PEM In-Reply-To: References: Message-ID: <24E88059-BF5E-4E21-8B65-35DF8FE38F1D@valo.at> >But can you explain why you use globally: > >ssl_cert = ssl_key = >and certs for any additional Domain each? > >## >local_name mail.pettijohn-web.com { > ssl_cert = ssl_key = } >## he configured the top level "default" ssl_{cert,key} as requested by the documentation i may quote myself >from the dovecot ssl wiki page http://wiki2.dovecot.org/SSL/DovecotConfiguration >Multiple SSL certificates >Different certificates per IP and protocol > >[snip] > >Note that you will still need a top-level "default"?ssl_key?and?ssl_cert?as well, or you will receive errors. i think, but untested, this default ssl_{cert,key} are used for those clients that don't support SNI (as with apache, which uses the cert of the first site). >I configured it the way you do, but within the default >/etc/dovecot/confd structure but >i had no luck. I testet local_name (SNI), local, local >(dedicated >IPv6 Address but had no lock it should make no difference whether you use one large config file or the conf.d structure. your doveconf -n output would really be helpful regards - c >Configinfo: >1) MTA (Postfix 2.10.1) and MDA (Dovecot 2.2.10) configured on IPv6 >Addresses >2) The MTA and MDA are connected to MariaDB (the default MySQL >replacement on CentOS7 now for virtual domains, users, passwords, >aliaes ect >3) Postfix uses Dovecots SASL Implementation and Postfix and Dovecot >talking via LMTP and UNIX Sockets. > >Details: >### yum info postfix ### >Name : postfix >Arch : x86_64 >Epoche : 2 >Version : 2.10.1 >Release : 6.el7 >Size : 12 M >From : installed > From Source : debian.n-ix.net_centos_7_os_x86_64_ >Summary : Postfix Mail Transport Agent >URL : http://www.postfix.org >License : IBM and GPLv2+ >Description : Postfix is a Mail Transport Agent (MTA), supporting LDAP, > >SMTP AUTH (SASL), TLS > >### yum info dovecot ### >Name : dovecot >Arch : i686 >Epoche : 1 >Version : 2.2.10 >Ausgabe : 4.el7_0.1 >Gr??e : 3.2 M >Quelle : debian.n-ix.net_centos_7_os_x86_64_ >Summary : Secure imap and pop3 server >URL : http://www.dovecot.org/ >Lizenz : MIT and LGPLv2 >Description : Dovecot is an IMAP server for Linux/UNIX-like systems, >written with security > : primarily in mind. It also contains a small POP3 >server. It supports mail > : in either of maildir or mbox formats. > >Cheers, >Drav From stephan at rename-it.nl Sun Aug 16 17:01:46 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Sun, 16 Aug 2015 19:01:46 +0200 Subject: sieve-filter failure problems In-Reply-To: <20150813143324.GE46491@ruhr-uni-bochum.de> References: <20150813143324.GE46491@ruhr-uni-bochum.de> Message-ID: <55D0C1FA.60201@rename-it.nl> Op 8/13/2015 om 4:33 PM schreef Jost Krieger: > I use sieve-filter for postprocessing misclassified mail. > For false positives I use the following script: > > require [ "variables", "include", "fileinto" ]; > > global [ "FORCENOSPAM", "ext", "ext1" ]; > > set "FORCENOSPAM" "YES"; > > fileinto "JUNK-PRENOUCE"; > > if header :matches "Delivered-To" "*+*@*" { > set "ext" "${2}"; > set "ext1" "${2}"; > } > > include :personal "production"; > > FORCENOSPAM suppresses the spam handling and the Delivered-To stuff > tries to regenerate envelope info. > > This is my crontab entry: > 1-57/4 * * * * cd ~kriegjcb; /usr/local/dovecot/bin/sieve-filter -v -e -W -u kriegjcb -m Spam/U0-failed ~kriegjcb/sieve/nospamfilter.sieve Spam/U0 > > Most things work fine, but a few of my rules will try to forward mail. > This will fail and my idea was I handle this manually from the folder > Spam/U0-failed. > > Unfortunately the mail will arrive there but will also be left in the source folder > Spam/U0 and will be processed again every 4 minutes ... > > This is probably not intentional? Fixed: http://hg.rename-it.nl/dovecot-2.2-pigeonhole/rev/469f9794408b Regards, Stephan. From skdovecot at smail.inf.fh-brs.de Mon Aug 17 06:54:19 2015 From: skdovecot at smail.inf.fh-brs.de (Steffen Kaiser) Date: Mon, 17 Aug 2015 08:54:19 +0200 (CEST) Subject: Error: User initialization failed In-Reply-To: <46885658-010B-4F0C-9996-B5C47B46DDB5@kreme.com> References: <46885658-010B-4F0C-9996-B5C47B46DDB5@kreme.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, 14 Aug 2015, @lbutlr wrote: > On Aug 14, 2015, at 8:51 AM, @lbutlr wrote: >> dovecot: lda(chance at example.com): Error: User initialization failed: Namespace '': Ambiguous mail location setting, don't know what to do with it: /usr/local/virtual/chance@ example.com (try prefixing it with mbox: or milder:) > > Here?s the dovecot -n output > > $ dovecot -n > # 2.2.18: /usr/local/etc/dovecot/dovecot.conf > # OS: FreeBSD 9.2-RELEASE-p3 i386 > auth_failure_delay = 5 secs > auth_mechanisms = PLAIN LOGIN > default_process_limit = 200 > disable_plaintext_auth = no > first_valid_uid = 89 > login_log_format_elements = user=<%u> %r %m %c > mail_location = maildir:~/Maildir > userdb { > args = /etc/dovecot/dovecot-sql.conf.ext > default_fields = uid=vpopmail gid=vchkpw mail_location=/usr/local/virtual/%u mail=maildir:/usr/local/virtual/%u looks like this mail_location is confusing Dovecot. If you wanted to specify "home" here, be aware that home and mail location should (aka must) be separate directories. > driver = sql > } > > - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVdGFG3z1H7kL/d9rAQKL3ggAiRRLG51ri5yRZdhRW+gW3A4Q9vnxH5Pp pOQV/VAI2gh/T0giNfUJ/K2OnMbEPJHBySFCBgcAYdKxJiSb30RO9alnu18/5q/3 XdpXAvpR9obzXaWTb6rGF8SII7VV5tmG953EEnzuqLOyQG6I/LbFznP+U9Wmenw0 +lDYbYbavcyeQBAz5IkRs20AeK6a6nK0JVClR2yvMwmm8YuYjG7UiUmk95p4rgme qKdypCaDPsq5xJjWEYtodAVimVTEjhsbdtSGruBPycIwSP501QpGJMGlGvJMhfd6 ObafIqfyWUeVelTC9NU89e+V9PwB/i5E6ZZE2xjrzMgCgnkQWpgcAA== =vvM8 -----END PGP SIGNATURE----- From alessio at skye.it Mon Aug 17 07:47:58 2015 From: alessio at skye.it (Alessio Cecchi) Date: Mon, 17 Aug 2015 09:47:58 +0200 Subject: Dovecot 2.2.18 Panic: file index-mail-binary.c In-Reply-To: <55B75128.6070207@borgelt.org> References: <20150728091215.Horde.gmc7i3jGADyFKBjMowoMTOj@hermes.dg4yfa.org> <55B75128.6070207@borgelt.org> Message-ID: <55D191AE.9080308@skye.it> Il 28/07/2015 11:53, Michael Borgelt ha scritto: > Hi. > I have tried this email also with thunderbird imap which works. So this is > as special Problem with the imap-module of the Horde framework. I will do > some more debugging this evening. > > My Horde framework has the following Versions: > Horde Groupware Webmail Edition 5.2.7 > Horde (horde) 5.2.6 > Horde_Imap_Client 2.29.1 Hi, I have the same error with Roundcube, I'm running dovecot 2.2.18 with Maildir/ and Director: Aug 17 09:34:35 imap(info at domain.com): Panic: file index-mail-binary.c: line 354 (blocks_count_lines): assertion failed: (ret == -1) Aug 17 09:34:35 imap(info at domain.com): Error: Raw backtrace: /usr/lib64/dovecot/libdovecot.so.0(+0x7b53a) [0x7f259013553a] -> /usr/lib64/dovecot/libdovecot.so.0(+0x7b5a6) [0x7f25901355a6] -> /usr/lib64/dovecot/libdovecot.so.0(+0x281ea) [0x7f25900e21ea] -> /usr/lib64/dovecot/libdovecot-storage.so.0(+0xad78e) [0x7f259045978e] -> /usr/lib64/dovecot/libdovecot-storage.so.0(index_mail_get_binary_stream+0x18e) [0x7f2590459a7e] -> /usr/lib64/dovecot/libdovecot-storage.so.0(mail_get_binary_stream+0x63) [0x7f25903e4923] -> /usr/lib64/dovecot/libdovecot-storage.so.0(imap_msgpart_open+0xb0) [0x7f259048c660] -> dovecot/imap [info at domain.com 192.168.1.15 UID FETCH]() [0x41af06] -> dovecot/imap [info at domain.com 192.168.1.15 UID FETCH]() [0x419274] -> dovecot/imap [info at domain.com 192.168.1.15 UID FETCH](imap_fetch_more+0x34) [0x4194b4] -> dovecot/imap [info at domain.com 192.168.1.15 UID FETCH](cmd_fetch+0x4b0) [0x40dda0] -> dovecot/imap [info at domain.com 192.168.1.15 UID FETCH](command_exec+0x4d) [0x4178ed] -> dovecot/imap [info at domain.com 192.168.1.15 UID FETCH]() [0x4168d0] -> dovecot/imap [info at domain.com 192.168.1.15 UID FETCH]() [0x416966] -> dovecot/imap [info at domain.com 192.168.1.15 UID FETCH](client_handle_input+0x11d) [0x416c5d] -> dovecot/imap [info at domain.com 192.168.1.15 UID FETCH](client_input+0x6f) [0x416fdf] -> /usr/lib64/dovecot/libdovecot.so.0(io_loop_call_io+0x5b) [0x7f2590147fbb] -> /usr/lib64/dovecot/libdovecot.so.0(io_loop_handler_run_internal+0xaf) [0x7f259014918f] -> /usr/lib64/dovecot/libdovecot.so.0(io_loop_handler_run+0x9) [0x7f2590148049] -> /usr/lib64/dovecot/libdovecot.so.0(io_loop_run+0x38) [0x7f25901480c8] -> /usr/lib64/dovecot/libdovecot.so.0(master_service_run+0x13) [0x7f25900e7043] -> dovecot/imap [info at domain.com 192.168.1.15 UID FETCH](main+0x2b1) [0x4206c1] -> /lib64/libc.so.6(__libc_start_main+0xfd) [0x7f258fd44d5d] -> dovecot/imap [info at domain.com 192.168.1.15 UID FETCH]() [0x40b199] Aug 17 09:34:35 imap(info at domain.com): Fatal: master: service(imap): child 13918 killed with signal 6 (core dumps disabled) -- Alessio Cecchi http://www.linkedin.com/in/alessice From tom at whyscream.net Mon Aug 17 07:53:48 2015 From: tom at whyscream.net (Tom Hendrikx) Date: Mon, 17 Aug 2015 09:53:48 +0200 Subject: sieve and antispam In-Reply-To: <55CDE85D.2070406@mcnaughty.com> References: <55CDE85D.2070406@mcnaughty.com> Message-ID: <55D1930C.2060401@whyscream.net> On 14-08-15 15:08, Andrew McN wrote: > > I have a fairly standard antispam/dspam setup: postfix delivers to > dspam, which classifies and tags the mail, and then passes it to postfix > which uses sieve to put mail into the appropriate folders based on > dspam's classification. I also have the antispam plugin watching mail > being moved between folders and re-training dpsam accordingly. > > If I now add some sieve filter rules which put incoming mail into > appropriate folders, possibly overriding what would normally result from > dspam's classification, will that be seen by the antispam plugin and > used for training dspam? > > Regards, > Andrew McNaughton > Normally, you'd enable the antispam plugin only for the imap backend, meaning that message delivery is not seen by the plugin. So no issues there. Tom From nmilas at noa.gr Mon Aug 17 11:26:30 2015 From: nmilas at noa.gr (Nikolaos Milas) Date: Mon, 17 Aug 2015 14:26:30 +0300 Subject: Duplicate mails with pop3 + dsync replication In-Reply-To: <7D04A2F7-D467-4A80-B64B-7B3FC8C76CB7@wogri.com> References: <7D04A2F7-D467-4A80-B64B-7B3FC8C76CB7@wogri.com> Message-ID: <55D1C4E6.6070600@noa.gr> On 22/6/2015 9:49 am, Wolfgang Hennerbichler wrote: > It turns out that if I enable this option: > pop3_deleted_flag = "$POP3Deleted? > > The issue no longer persists. I have to manually expunge the kept mails that have been deleted via pop3 though: > doveadm expunge mailbox INBOX KEYWORD '$POP3Deleted' -A Hello, I was wondering if the same behavior persists with latest version(s), for example 2.2.18. If so, how do you overcome this issue? What is the suggested practice? Can someone please provide administration guidelines to handle it? I am seeing this behavior on v2.2.13; I am now planning an upgrade to 2.2.18. Please advise! My setup is: Master 1: dsync_remote_cmd = ssh -l root master2.example.com doveadm dsync-server -u%u replication_dsync_parameters = -d -N -l 30 -U service aggregator { fifo_listener replication-notify-fifo { user = vmail } unix_listener replication-notify { user = vmail } } service replicator { unix_listener replicator-doveadm { mode = 0600 } } Master 2: dsync_remote_cmd = ssh -l root master1.example.com doveadm dsync-server -u%u replication_dsync_parameters = -d -N -l 30 -U service aggregator { fifo_listener replication-notify-fifo { user = vmail } unix_listener replication-notify { user = vmail } } service replicator { unix_listener replicator-doveadm { mode = 0600 } } Replication plugin is enabled on protocols pop3, imap, lda. Thanks in advance, Nick From Jost.Krieger+dovecot at rub.de Mon Aug 17 11:29:59 2015 From: Jost.Krieger+dovecot at rub.de (Jost Krieger) Date: Mon, 17 Aug 2015 13:29:59 +0200 Subject: sieve-filter failure problems In-Reply-To: <55D0C1FA.60201@rename-it.nl> References: <20150813143324.GE46491@ruhr-uni-bochum.de> <55D0C1FA.60201@rename-it.nl> Message-ID: <20150817112959.GC878@ruhr-uni-bochum.de> On Sun Aug 16 19:01:46 2015, Stephan Bosch wrote: > Op 8/13/2015 om 4:33 PM schreef Jost Krieger: > > ... > > Unfortunately the mail will arrive there but will also be left in the source folder > > Spam/U0 and will be processed again every 4 minutes ... > > > > This is probably not intentional? > > Fixed: > > http://hg.rename-it.nl/dovecot-2.2-pigeonhole/rev/469f9794408b Thanks a lot, works like a charm. Yours Jost Krieger -- | Jost.Krieger+sig at ruhr-uni-bochum.de Please help stamp out spam! | | Postmaster, JAPH, resident answer machine at RUB Comp. Center | | Sincere words are not sweet, sweet words are not sincere. | | Lao Tse, Tao Te King 81 | -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 4784 bytes Desc: not available URL: From andy at andybev.com Mon Aug 17 12:05:18 2015 From: andy at andybev.com (Andrew Beverley) Date: Mon, 17 Aug 2015 13:05:18 +0100 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <55C36AF2.5090202@rename-it.nl> References: <1437780774.7496.7.camel@andybev.com> <55C308CD.5070308@rename-it.nl> <1438846556.14063.7.camel@andybev.com> <55C3118F.3050805@rename-it.nl> <1438848811.18074.3.camel@andybev.com> <55C36AF2.5090202@rename-it.nl> Message-ID: <1439813118.9225.111.camel@andybev.com> On Thu, 2015-08-06 at 16:10 +0200, Stephan Bosch wrote: > Since the extdata plugin is not part of the normal Debian packages (not > afaik anyway), you likely compiled and installed the extdata plugin > manually in the past while using Dovecot 2.1. Do that again, but use > this repository: http://hg.rename-it.nl/pigeonhole-0.4-sieve-extdata/ I'm just trying to install this now, but am getting the following compilation error: make[2]: Entering directory '/usr/src/pigeonhole-0-4-sieve-extdata-4ce3912ee200/src' /bin/bash ../libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I.. -I.. -I/usr/include/dovecot -I/usr/include/dovecot/sieve -DPKG_RUNDIR=\"""\" -std=gnu99 -g -O2 -fPIE -fstack-protector-strong -Wformat -Werror=format-security -Wall -W -Wmissing-prototypes -Wmissing-declarations -Wpointer-arith -Wchar-subscripts -Wformat=2 -Wbad-function-cast -fno-builtin-strftime -Wstrict-aliasing=2 -MT sieve-extdata-plugin.lo -MD -MP -MF .deps/sieve-extdata-plugin.Tpo -c -o sieve-extdata-plugin.lo sieve-extdata-plugin.c libtool: compile: gcc -DHAVE_CONFIG_H -I. -I.. -I.. -I/usr/include/dovecot -I/usr/include/dovecot/sieve -DPKG_RUNDIR=\"\" -std=gnu99 -g -O2 -fstack-protector-strong -Wformat -Werror=format-security -Wall -W -Wmissing-prototypes -Wmissing-declarations -Wpointer-arith -Wchar-subscripts -Wformat=2 -Wbad-function-cast -fno-builtin-strftime -Wstrict-aliasing=2 -MT sieve-extdata-plugin.lo -MD -MP -MF .deps/sieve-extdata-plugin.Tpo -c sieve-extdata-plugin.c -fPIC -DPIC -o .libs/sieve-extdata-plugin.o sieve-extdata-plugin.c:20:44: error: 'PIGEONHOLE_ABI_VERSION' undeclared here (not in a function) const char *sieve_extdate_plugin_version = PIGEONHOLE_ABI_VERSION; This is on Debian Jessie using: ./configure --with-dovecot=/usr/lib/dovecot/ --with-pigeonhole=/usr/include/dovecot/sieve/ The headers seem to be found correctly when using configure, it's just when compiling that I get an error. pigeonhole-0-4-sieve-extdata-4ce3912ee200 Any ideas? Thanks, Andy From stephan at rename-it.nl Mon Aug 17 12:13:53 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Mon, 17 Aug 2015 14:13:53 +0200 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <1439813118.9225.111.camel@andybev.com> References: <1437780774.7496.7.camel@andybev.com> <55C308CD.5070308@rename-it.nl> <1438846556.14063.7.camel@andybev.com> <55C3118F.3050805@rename-it.nl> <1438848811.18074.3.camel@andybev.com> <55C36AF2.5090202@rename-it.nl> <1439813118.9225.111.camel@andybev.com> Message-ID: <55D1D001.9050602@rename-it.nl> Andrew Beverley schreef op 17-8-2015 om 14:05: > On Thu, 2015-08-06 at 16:10 +0200, Stephan Bosch wrote: >> Since the extdata plugin is not part of the normal Debian packages (not >> afaik anyway), you likely compiled and installed the extdata plugin >> manually in the past while using Dovecot 2.1. Do that again, but use >> this repository: http://hg.rename-it.nl/pigeonhole-0.4-sieve-extdata/ > I'm just trying to install this now, but am getting the following compilation > error: > > make[2]: Entering directory '/usr/src/pigeonhole-0-4-sieve-extdata-4ce3912ee200/src' > /bin/bash ../libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I.. -I.. -I/usr/include/dovecot -I/usr/include/dovecot/sieve -DPKG_RUNDIR=\"""\" -std=gnu99 -g -O2 -fPIE -fstack-protector-strong -Wformat -Werror=format-security -Wall -W -Wmissing-prototypes -Wmissing-declarations -Wpointer-arith -Wchar-subscripts -Wformat=2 -Wbad-function-cast -fno-builtin-strftime -Wstrict-aliasing=2 -MT sieve-extdata-plugin.lo -MD -MP -MF .deps/sieve-extdata-plugin.Tpo -c -o sieve-extdata-plugin.lo sieve-extdata-plugin.c > libtool: compile: gcc -DHAVE_CONFIG_H -I. -I.. -I.. -I/usr/include/dovecot -I/usr/include/dovecot/sieve -DPKG_RUNDIR=\"\" -std=gnu99 -g -O2 -fstack-protector-strong -Wformat -Werror=format-security -Wall -W -Wmissing-prototypes -Wmissing-declarations -Wpointer-arith -Wchar-subscripts -Wformat=2 -Wbad-function-cast -fno-builtin-strftime -Wstrict-aliasing=2 -MT sieve-extdata-plugin.lo -MD -MP -MF .deps/sieve-extdata-plugin.Tpo -c sieve-extdata-plugin.c -fPIC -DPIC -o .libs/sieve-extdata-plugin.o > sieve-extdata-plugin.c:20:44: error: 'PIGEONHOLE_ABI_VERSION' undeclared here (not in a function) > const char *sieve_extdate_plugin_version = PIGEONHOLE_ABI_VERSION; > > This is on Debian Jessie using: > > ./configure --with-dovecot=/usr/lib/dovecot/ --with-pigeonhole=/usr/include/dovecot/sieve/ > > The headers seem to be found correctly when using configure, it's just when > compiling that I get an error. > > pigeonhole-0-4-sieve-extdata-4ce3912ee200 > > Any ideas? Yes. I prepared for preventing your problem by adding ABI version support (much like Dovecot itself already has for its plugins). This is the last change in the pigeonhole-0.4-sieve-extdata plugin repository. There is no Pigeonhole release for this yet (and adoption in Debian will likely take some time), but until then you can avoid this by using the the exdata hg revision before tip (57c8d3e6b562). Regards, Stephan. From andy at andybev.com Mon Aug 17 12:35:36 2015 From: andy at andybev.com (Andrew Beverley) Date: Mon, 17 Aug 2015 13:35:36 +0100 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <55D1D001.9050602@rename-it.nl> References: <1437780774.7496.7.camel@andybev.com> <55C308CD.5070308@rename-it.nl> <1438846556.14063.7.camel@andybev.com> <55C3118F.3050805@rename-it.nl> <1438848811.18074.3.camel@andybev.com> <55C36AF2.5090202@rename-it.nl> <1439813118.9225.111.camel@andybev.com> <55D1D001.9050602@rename-it.nl> Message-ID: <1439814936.9225.119.camel@andybev.com> On Mon, 2015-08-17 at 14:13 +0200, Stephan Bosch wrote: > but until then you can avoid this by using the the exdata hg revision before tip > (57c8d3e6b562). Great, thanks for the quick reply, that fixed the compilation problem. I'm still getting a segfault though. This time the backtrace is: Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1". Core was generated by `dovecot/lmtp'. Program terminated with signal SIGSEGV, Segmentation fault. #0 sieve_validator_find_command_registration (valdtr=0xffffffff, command=0x7ff53d612cd8 "extdata") at sieve-validator.c:309 309 sieve-validator.c: No such file or directory. (gdb) bt full #0 sieve_validator_find_command_registration (valdtr=0xffffffff, command=0x7ff53d612cd8 "extdata") at sieve-validator.c:309 No locals. #1 sieve_validator_register_command (valdtr=0xffffffff, ext=0x7ff53fb5a280, cmd_def=0x7ff53d813400 ) at sieve-validator.c:331 cmd_reg = #2 0x00007ff53d6124e6 in ext_extdata_validator_load (ext=0x7ff53fb5a280, valdtr=0xffffffff) at ext-extdata.c:46 No locals. #3 0x00007ff53e08c1a4 in _sieve_extension_unload (ext=) at sieve-extensions.c:316 No locals. #4 sieve_extension_unregister (ext=) at sieve-extensions.c:426 ext_reg = ext_id = #5 0x00007ff53e08c50e in sieve_plugins_unload (svinst=0x7ff53fb599b0) at sieve-plugins.c:166 module = 0x7ff53fb5e500 unload_func = plugin = 0x7ff53fb5a268 __FUNCTION__ = "sieve_plugins_unload" #6 0x00007ff53e0918dc in sieve_deinit (svinst=svinst at entry=0x7fff1893f7e0) at sieve.c:136 No locals. #7 0x00007ff53e2da948 in lda_sieve_deliver_mail (mdctx=, storage_r=0x7fff1893f998) at lda-sieve-plugin.c:948 srctx = {svinst = 0x7ff53fb599b0, mdctx = 0x7fff1893f9c0, home_dir = 0x7ff53fb55f78 "/var/mail/vhosts/xxx/xxx", scripts = 0x7ff53fb1a088, script_count = 0, user_script = 0x0, main_script = 0x0, msgdata = 0x0, scriptenv = 0x0, user_ehandler = 0x0, master_ehandler = 0x7ff53fb5e6c0, userlog = 0x0} debug = svenv = {hostname = 0x7ff53fb35b98 "xxx.com", domainname = 0x0, base_dir = 0x7ff53fb558c0 "/var/run/dovecot", username = 0x7ff53fb54da8 "xxx at xxx.com", home_dir = 0x7ff53fb55f78 "/var/mail/vhosts/xxx/xxx", flags = SIEVE_FLAG_HOME_RELATIVE, location = SIEVE_ENV_LOCATION_MDA, delivery_phase = SIEVE_DELIVERY_PHASE_DURING} i = ret = #8 0x00007ff53f08dd69 in mail_deliver (ctx=ctx at entry=0x7fff1893f9c0, storage_r=storage_r at entry=0x7fff1893f998) at mail-deliver.c:400 ret = #9 0x00007ff53f4bd171 in client_deliver (session=0x7ff53fb53060, src_mail=0x7ff53fb4e840, rcpt=0x7ff53fb277d0, client=0x7ff53fb26f50) at commands.c:689 lda_set = 0x7ff53fb35b28 ns = set_parser = line = storage = 0x0 sets = mail_error = 1068655520 ret = dctx = {pool = 0x7ff53fb53040, set = 0x7ff53fb35b28, session = 0x7ff53fb53060, dup_ctx = 0x7ff53fb598d0, session_id = 0x7ff53fb27760 "WhTfIZfS0VVBKAAAChoztw", src_mail = 0x7ff53fb4e840, src_envelope_sender = 0x7ff53fb27778 "xxx at hotmail.com", dest_user = 0x7ff53fb54cd0, dest_addr = 0x7ff53fb27dd0 "xxx at xxx.com", final_dest_addr = 0x7ff53fb27dd0 "xxx at xxx.com", dest_mailbox_name = 0x7ff53f4bfad0 "INBOX", dest_mail = 0x0, var_expand_table = 0x0, tempfail_error = 0x0, tried_default_save = false, saved_mail = false, save_dest_mail = false, mailbox_full = false, dsn = false} input = mail_set = error = username = ---Type to continue, or q to quit--- #10 client_deliver_next (session=0x7ff53fb53060, src_mail=0x7ff53fb4e840, client=0x7ff53fb26f50) at commands.c:732 count = #11 client_input_data_write_local (input=, client=0x7ff53fb26f50) at commands.c:827 src_mail = 0x7ff53fb4e840 first_uid = 4294967295 session = 0x7ff53fb53060 old_uid = 0 #12 client_input_data_write (client=0x7ff53fb26f50) at commands.c:939 input = 0x7ff53fb36c50 ret = true #13 client_input_data_handle (client=0x7ff53fb26f50) at commands.c:1033 data = size = 936 ret = #14 0x00007ff53eb07d0f in io_loop_call_io (io=0x7ff53fb263a0) at ioloop.c:441 ioloop = 0x7ff53fb21730 t_id = 2 __FUNCTION__ = "io_loop_call_io" #15 0x00007ff53eb08d09 in io_loop_handler_run_internal (ioloop=ioloop at entry=0x7ff53fb21730) at ioloop-epoll.c:220 ctx = 0x7ff53fb24550 io = tv = {tv_sec = 299, tv_usec = 981328} events_count = msecs = ret = 1 i = 0 j = call = __FUNCTION__ = "io_loop_handler_run_internal" #16 0x00007ff53eb07d79 in io_loop_handler_run (ioloop=ioloop at entry=0x7ff53fb21730) at ioloop.c:488 No locals. #17 0x00007ff53eb07df8 in io_loop_run (ioloop=0x7ff53fb21730) at ioloop.c:465 __FUNCTION__ = "io_loop_run" #18 0x00007ff53eab2dc3 in master_service_run (service=0x7ff53fb215c0, callback=) at master-service.c:566 No locals. #19 0x00007ff53f4bb9b5 in main (argc=1, argv=0x7ff53fb21390) at main.c:122 set_roots = {0x7ff53f2914c0 , 0x7ff53f6c16c0 , 0x0} service_flags = storage_service_flags = c = From nmilas at noa.gr Mon Aug 17 12:42:08 2015 From: nmilas at noa.gr (Nikolaos Milas) Date: Mon, 17 Aug 2015 15:42:08 +0300 Subject: Duplicate mails with pop3 + dsync replication In-Reply-To: References: <7D04A2F7-D467-4A80-B64B-7B3FC8C76CB7@wogri.com> <55D1C4E6.6070600@noa.gr> Message-ID: <55D1D6A0.6010202@noa.gr> On 17/8/2015 2:29 ??, Wolfgang Hennerbichler wrote: >> I was wondering if the same behavior persists with latest version(s), for example 2.2.18. > I don?t know, I have to deal with the packages that come with my distribution?s repository. > >> >If so, how do you overcome this issue? What is the suggested practice? > All I can say is that it works for me with the above setting and a cron-job that runs every minute which expunges mails for the pop3 users. > Thank you for your reply. I am on CentOS 5 and 6, but I am building RPMs for our test and production servers. I recently built v2.2.18 and I decided to give it a try. (I run it with postfix - also just upgraded to 2.11.6.) The issue seems to not occur any more. I had already tested on two test servers, and I hoped someone could provide independent feedback before I decided to upgrade production servers. However, I now also upgraded our two production servers and things seem to be running smoothly. All the best, Nick From stephan at rename-it.nl Mon Aug 17 13:23:19 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Mon, 17 Aug 2015 15:23:19 +0200 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <1439814936.9225.119.camel@andybev.com> References: <1437780774.7496.7.camel@andybev.com> <55C308CD.5070308@rename-it.nl> <1438846556.14063.7.camel@andybev.com> <55C3118F.3050805@rename-it.nl> <1438848811.18074.3.camel@andybev.com> <55C36AF2.5090202@rename-it.nl> <1439813118.9225.111.camel@andybev.com> <55D1D001.9050602@rename-it.nl> <1439814936.9225.119.camel@andybev.com> Message-ID: <55D1E047.4040404@rename-it.nl> Andrew Beverley schreef op 17-8-2015 om 14:35: > On Mon, 2015-08-17 at 14:13 +0200, Stephan Bosch wrote: >> but until then you can avoid this by using the the exdata hg revision before tip >> (57c8d3e6b562). > Great, thanks for the quick reply, that fixed the compilation problem. > > I'm still getting a segfault though. This time the backtrace is: Still looks like the executed extdata plugin was compiled against a different pigeonhole. That validator_load function is called inappropriately from extension_unload. Are you sure that the plugin is installed and the old one is not present anymore? Regards, Stephan. From gfinch at ldmltd.ca Mon Aug 17 14:30:00 2015 From: gfinch at ldmltd.ca (Gregory Finch) Date: Mon, 17 Aug 2015 07:30:00 -0700 Subject: multiple messages per second to a single mailbox In-Reply-To: <9999C211-0CA7-4F23-851A-CB36DA10B685@balius.com> References: <55CB7B42.104@univention.de> <8D617AEC-0038-4B4F-B393-C95DCD737F96@balius.com> <9999C211-0CA7-4F23-851A-CB36DA10B685@balius.com> Message-ID: <55D1EFE8.8030906@ldmltd.ca> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2015-08-14 7:52 AM, Chad M Stewart wrote: > > The problem happened again this morning. Removing fsync calls helped, but I'm not sure about leaving that enabled long term. > > I still believe the problem is multiple dovecot processes trying to write to a single folder at the same time. (If I could run dtrace I might be able to cobble together a script to prove it.) > > I tried writing a sieve script to direct the messages to a set of folders, but I'm not able to make the logic work. I was thinking something like: generate random # ($N) between say 1-10, then file message into folder mail$N. But I didn't find a method to do that within sieve. > > My next thought was to try parsing the message-id header. If the first character is [0-5] then fileinto mail1, etc. Then I could go so far as having 36 subfolders which the messages could be written too. This mailbox only keeps messages for a rolling 1d window. Right now for example it has 260,186 messages in the INBOX. > > The sieve script I tried (with only about 4 hours of sleep) was > > require ["fileinto","regex"]; > > if header :regex "message-id" "^1" { > fileinto "mail1"; > } else { > keep; > } > > > If anyone has some suggestions on how I might spread the messages out over multiple folders I'd like to hear your thoughts. Again the servers are configured using maildir, so each folder should have its own index and thus file locking contention should be less, at least so goes the theory in my head. > > > Thank you, > Chad > With that many messages, it may be the index updates that are slowing things down. Depending on your needs or how you use this folder (it sounds automated to me), you may not need the indexes on disk at all. If you set the indexes to RAM, it may help. mail_location = maildir:~/Maildir:INDEX=MEMORY as per http://wiki2.dovecot.org/MailLocation#Index_files If indexes in RAM isn't an option, maybe just put them on a separate spindle / SSD would help. Please note I'm shooting from the hip with this advice, I don't have a system that pushes dovecot that hard. Greg -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBCAAGBQJV0e/oAAoJECD7Htp+IT9eUXsQALgtSocaTUXWjMzp1oIEZTsT IOBzc25q4LZtNWeQpa/czmc9TBQfmqcRzXspyqu/dTxdesJ05WL3PiD/c66Bsmb8 0652a+6l7rLgMroIgsYCQZNBaoU+7FgIakWQlyHA/Ods7DTFNNRvAVNhZ6e6NcvR pkIM5EtLaHzc7318ahR2mC4tRHbmlSxZT/fnWuCu185VyuG6GiHum0piyplEIRqw ngbHkMROgD/OJTKScngRP73umLznxOfReVrM50tIlgpONY+ngTLvD6rnkbLO1t1Y KdVAc49iEQPTRuRVDDS+kf2haUx9Rh8Zz0R4VRdoiBC9bcRmeuQDNU4mg4fBCgQF JlkIkuYBvZe/y7TcxJ5rJXNtEn6g4Uew7mIt+sMnGx7Ro55kl0X61Qg/IWu+F8m5 aO9Y7jMjptsjm/04YWlaeCNUPolgUQrFLDjiO5uycEZh1NReNZEH5AFM0cYWMhBK PWzDRXwv4XEypNA+y+uDLdN3mX/KcWMUv90ipJT5PdrLPjvgjTbyU/Wr391H01Kf 85ShLVvBsV2NGvXY0ZRAxuF5Ndq04TNrRYDvVgzb4XlCH++l2WeZJ2CZ8b1KL5AC BydVyp5tWoP5TIzelG6NcMcqNjD05rtmcH2QCT1BLMksY9+7YkbU6IX+W5Dc5sOX krjSPNAsLOFJJCxBIZ/S =AYpS -----END PGP SIGNATURE----- From andy at andybev.com Mon Aug 17 14:50:10 2015 From: andy at andybev.com (Andrew Beverley) Date: Mon, 17 Aug 2015 15:50:10 +0100 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <55D1E047.4040404@rename-it.nl> References: <1437780774.7496.7.camel@andybev.com> <55C308CD.5070308@rename-it.nl> <1438846556.14063.7.camel@andybev.com> <55C3118F.3050805@rename-it.nl> <1438848811.18074.3.camel@andybev.com> <55C36AF2.5090202@rename-it.nl> <1439813118.9225.111.camel@andybev.com> <55D1D001.9050602@rename-it.nl> <1439814936.9225.119.camel@andybev.com> <55D1E047.4040404@rename-it.nl> Message-ID: <1439823010.32417.5.camel@andybev.com> On Mon, 2015-08-17 at 15:23 +0200, Stephan Bosch wrote: > Still looks like the executed extdata plugin was compiled against a > different pigeonhole. Got it. There was an old version in /usr/lib/dovecot/modules/sieve/, whereas the new one was installed into /usr/local/lib/dovecot/sieve/. That all works now, and no segfault. Thanks for all your help and patience. Unfortunately my vacation rule no longer works though. I'm getting "sieve: user's script dict:proxy::sieve;name=active doesn't exist (trying default script location instead)" I have a dict:proxy configured (/etc/dovecot/pigeonhole-sieve.dict): connect = host=127.0.0.1 dbname=postfix user=postfix password=xxx map { pattern = priv/sieve/name/$script_name table = virtual_users username_field = email value_field = sieve_id fields { sieve_active = $script_name } } And that added to dovecot.conf: dict { sieve = mysql:/etc/dovecot/pigeonhole-sieve.dict } And that in 90-sieve.conf: plugin { sieve = dict:proxy::sieve;name=active } I guess I'm doing something stupid. Any ideas? Thanks, Andy From dovecot at knutejohnson.com Mon Aug 17 15:25:48 2015 From: dovecot at knutejohnson.com (Knute Johnson) Date: Mon, 17 Aug 2015 08:25:48 -0700 Subject: Syncing two mail servers? In-Reply-To: <55CFC529.20600@freedominterface.org> References: <55CFAE7D.9080302@knutejohnson.com> <55CFC529.20600@freedominterface.org> Message-ID: <55D1FCFC.6040301@knutejohnson.com> Thanks I'll look into that. knute... On 8/15/2015 16:03, Eduardo Ramos wrote: > For sure Knute. > > Take a look at this: http://wiki2.dovecot.org/Replication > > I think a good implementation would be using a dovecot director layer > + dovecot replication. > > > On 08/15/2015 06:26 PM, Knute Johnson wrote: >> I currently have a mail server using Dovecot and sendmail. I was >> thinking I would like to have a backup of my mail and an additional >> server in case of hardware failure. Is it possible to run two >> identical or nearly identical server setups and sync the mail files >> between them? If it is possible, in general terms how would one go >> about this? >> >> Thanks, >> -- Knute Johnson From stephan at rename-it.nl Mon Aug 17 15:37:15 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Mon, 17 Aug 2015 17:37:15 +0200 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <1439823010.32417.5.camel@andybev.com> References: <1437780774.7496.7.camel@andybev.com> <55C308CD.5070308@rename-it.nl> <1438846556.14063.7.camel@andybev.com> <55C3118F.3050805@rename-it.nl> <1438848811.18074.3.camel@andybev.com> <55C36AF2.5090202@rename-it.nl> <1439813118.9225.111.camel@andybev.com> <55D1D001.9050602@rename-it.nl> <1439814936.9225.119.camel@andybev.com> <55D1E047.4040404@rename-it.nl> <1439823010.32417.5.camel@andybev.com> Message-ID: <55D1FFAB.7000304@rename-it.nl> Andrew Beverley schreef op 17-8-2015 om 16:50: > On Mon, 2015-08-17 at 15:23 +0200, Stephan Bosch wrote: >> Still looks like the executed extdata plugin was compiled against a >> different pigeonhole. > Got it. There was an old version in /usr/lib/dovecot/modules/sieve/, whereas the > new one was installed into /usr/local/lib/dovecot/sieve/. That all works now, and > no segfault. Thanks for all your help and patience. > > Unfortunately my vacation rule no longer works though. I'm getting "sieve: user's > script dict:proxy::sieve;name=active doesn't exist (trying default script location > instead)" [...] > I guess I'm doing something stupid. Any ideas? Could you enable mail_debug and look at your log output? That should show some more details of what it is doing. Regards, Stephan. From cms at balius.com Mon Aug 17 15:38:38 2015 From: cms at balius.com (Chad M Stewart) Date: Mon, 17 Aug 2015 10:38:38 -0500 Subject: multiple messages per second to a single mailbox In-Reply-To: <55D1EFE8.8030906@ldmltd.ca> References: <55CB7B42.104@univention.de> <8D617AEC-0038-4B4F-B393-C95DCD737F96@balius.com> <9999C211-0CA7-4F23-851A-CB36DA10B685@balius.com> <55D1EFE8.8030906@ldmltd.ca> Message-ID: <5B4A2141-F792-450D-B494-9AAE2B6AEEA4@balius.com> On Aug 17, 2015, at 9:30 AM, Gregory Finch wrote: <..snip..> > With that many messages, it may be the index updates that are slowing > things down. Depending on your needs or how you use this folder (it > sounds automated to me), you may not need the indexes on disk at all. If > you set the indexes to RAM, it may help. > > mail_location = maildir:~/Maildir:INDEX=MEMORY > > as per http://wiki2.dovecot.org/MailLocation#Index_files > > If indexes in RAM isn't an option, maybe just put them on a separate > spindle / SSD would help. > > Please note I'm shooting from the hip with this advice, I don't have a > system that pushes dovecot that hard. > > Greg > > This particular mailbox is unique. This mailbox holds a copy of member to member communications. The mailbox also receives a lot of unfiltered spam. You're right, it is the index updates that were limiting the message insertion rate. I wrote a sieve script that divides the messages based on the first character of the message-id header. For now this appears to be enough to spread out the index updates. The performance metrics indicate I'm now seeing double the IOPS as before the sieve script. With no noticeable increase in IO wait times. If the sieve script ever stops being enough, then I'll look into moving this mailboxes index files to RAM. Thank you for the tip! Regards, -Chad From chris.laif at googlemail.com Mon Aug 17 16:11:52 2015 From: chris.laif at googlemail.com (Chris Laif) Date: Mon, 17 Aug 2015 18:11:52 +0200 Subject: Store ACL files within shared mailbox / rename shared folder problem Message-ID: Hi. If $otheruser shares a mailbox to $myuser, my private index is stored in *my* Maildir, according to this template (from the dovecot docs): mail_location = maildir:%%h/Maildir:INDEXPVT=%h/Maildir/shared/%%u If $otheruser renames the shared mailbox, all my private mail flags (Seen) are lost because my mailclient couldn't find the private index anymore. Is there any way to store the private index within the shared mailbox? If $otheruser renames the mailbox, the index will be 'migrated' as well. Something like this: /data/mails/$otheruser/Maildir/.MagicSharedFolder/shared/$myuser.dovecot.index.pvt.log I see no way on how to do craft a INDEXPVT for this. Kind regards, Chris From andy at andybev.com Tue Aug 18 00:04:17 2015 From: andy at andybev.com (Andrew Beverley) Date: Tue, 18 Aug 2015 01:04:17 +0100 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <55D1FFAB.7000304@rename-it.nl> References: <1437780774.7496.7.camel@andybev.com> <55C308CD.5070308@rename-it.nl> <1438846556.14063.7.camel@andybev.com> <55C3118F.3050805@rename-it.nl> <1438848811.18074.3.camel@andybev.com> <55C36AF2.5090202@rename-it.nl> <1439813118.9225.111.camel@andybev.com> <55D1D001.9050602@rename-it.nl> <1439814936.9225.119.camel@andybev.com> <55D1E047.4040404@rename-it.nl> <1439823010.32417.5.camel@andybev.com> <55D1FFAB.7000304@rename-it.nl> Message-ID: <1439856257.1183.22.camel@andybev.com> On Mon, 2015-08-17 at 17:37 +0200, Stephan Bosch wrote: > > Unfortunately my vacation rule no longer works though. I'm getting "sieve: > > user's script dict:proxy::sieve;name=active doesn't exist (trying default > > script location instead)" > [...] > > I guess I'm doing something stupid. Any ideas? > > Could you enable mail_debug and look at your log output? That should > show some more details of what it is doing. I've fixed this by changing the name of the sieve script in the database from "active" to "main script". I'm a bit confused by this though. I could see this in the log: Aug 17 23:28:54 fieri dovecot: lmtp(10285, x at x.com): Debug: yxlyGzZu0lUtKAAAChoztw: sieve: sieve dict backend: user=x at x.com, uri=proxy::sieve, script=main script Aug 17 23:28:54 fieri dovecot: lmtp(10285, x at x.com): Debug: yxlyGzZu0lUtKAAAChoztw: sieve: sieve dict backend: script `main script' not found at path priv/sieve/name/main script Aug 17 23:28:54 fieri dovecot: lmtp(10285, x at x.com): Debug: yxlyGzZu0lUtKAAAChoztw: sieve: user's script dict:proxy::sieve;name=active doesn't exist (trying default script location instead) Aug 17 23:28:54 fieri dovecot: lmtp(10285, x at x.com): Debug: yxlyGzZu0lUtKAAAChoztw: sieve: no default script configured for user Aug 17 23:28:54 fieri dovecot: lmtp(10285, x at x.com): Debug: yxlyGzZu0lUtKAAAChoztw: sieve: user has no valid location for a personal script Aug 17 23:28:54 fieri dovecot: lmtp(10285, x at x.com): Debug: yxlyGzZu0lUtKAAAChoztw: sieve: no scripts to execute: reverting to default delivery. So it's looking first for "main script", which is why it now works. But it was then looking for "dict:proxy::sieve;name=active". Why didn't that work when I have this in my sieve config? plugin { sieve = dict:proxy::sieve;name=active } I also don't understand why it is looking for "main script" rather than "active" at all, and why things stopped working when I upgraded. Sorry for all the questions - I'll be happy to update the wiki once I've understood things better. Thanks for all your help, Andy From stephan at rename-it.nl Tue Aug 18 06:53:57 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Tue, 18 Aug 2015 08:53:57 +0200 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <1439856257.1183.22.camel@andybev.com> References: <1437780774.7496.7.camel@andybev.com> <55C308CD.5070308@rename-it.nl> <1438846556.14063.7.camel@andybev.com> <55C3118F.3050805@rename-it.nl> <1438848811.18074.3.camel@andybev.com> <55C36AF2.5090202@rename-it.nl> <1439813118.9225.111.camel@andybev.com> <55D1D001.9050602@rename-it.nl> <1439814936.9225.119.camel@andybev.com> <55D1E047.4040404@rename-it.nl> <1439823010.32417.5.camel@andybev.com> <55D1FFAB.7000304@rename-it.nl> <1439856257.1183.22.camel@andybev.com> Message-ID: <55D2D685.2070304@rename-it.nl> Op 8/18/2015 om 2:04 AM schreef Andrew Beverley: > On Mon, 2015-08-17 at 17:37 +0200, Stephan Bosch wrote: >>> Unfortunately my vacation rule no longer works though. I'm getting "sieve: >>> user's script dict:proxy::sieve;name=active doesn't exist (trying default >>> script location instead)" >> [...] >>> I guess I'm doing something stupid. Any ideas? >> Could you enable mail_debug and look at your log output? That should >> show some more details of what it is doing. > I've fixed this by changing the name of the sieve script in the database from > "active" to "main script". I'm a bit confused by this though. > > I could see this in the log: > > Aug 17 23:28:54 fieri dovecot: lmtp(10285, x at x.com): Debug: > yxlyGzZu0lUtKAAAChoztw: sieve: sieve dict backend: user=x at x.com, uri=proxy::sieve, > script=main script > Aug 17 23:28:54 fieri dovecot: lmtp(10285, x at x.com): Debug: > yxlyGzZu0lUtKAAAChoztw: sieve: sieve dict backend: script `main script' not found > at path priv/sieve/name/main script > Aug 17 23:28:54 fieri dovecot: lmtp(10285, x at x.com): Debug: > yxlyGzZu0lUtKAAAChoztw: sieve: user's script dict:proxy::sieve;name=active doesn't > exist (trying default script location instead) > Aug 17 23:28:54 fieri dovecot: lmtp(10285, x at x.com): Debug: > yxlyGzZu0lUtKAAAChoztw: sieve: no default script configured for user > Aug 17 23:28:54 fieri dovecot: lmtp(10285, x at x.com): Debug: > yxlyGzZu0lUtKAAAChoztw: sieve: user has no valid location for a personal script > Aug 17 23:28:54 fieri dovecot: lmtp(10285, x at x.com): Debug: > yxlyGzZu0lUtKAAAChoztw: sieve: no scripts to execute: reverting to default > delivery. > > So it's looking first for "main script", which is why it now works. But it was > then looking for "dict:proxy::sieve;name=active". Why didn't that work when I have > this in my sieve config? > > plugin { > sieve = dict:proxy::sieve;name=active > } > > I also don't understand why it is looking for "main script" rather than "active" > at all, and why things stopped working when I upgraded. > > Sorry for all the questions - I'll be happy to update the wiki once I've > understood things better. Unfortunately, Jessie is stuck at a very old version of Pigeonhole which has this problem. Regards, Stephan. From Christian.Schmidt at chemie.uni-hamburg.de Tue Aug 18 08:28:57 2015 From: Christian.Schmidt at chemie.uni-hamburg.de (Christian Schmidt) Date: Tue, 18 Aug 2015 10:28:57 +0200 Subject: multiple messages per second to a single mailbox In-Reply-To: References: <84C57ADD-1F76-4276-9DDD-B5B347600A80@balius.com> <83DDA393-5FA9-4BB3-BF7E-56372C107934@balius.com> Message-ID: <55D2ECC9.3050904@chemie.uni-hamburg.de> On 14.08.2015 23:20, Andrzej A. Filip wrote: > As I understand: > * maildir is designed for allowing fast "lockless" parallel deliveries > * dovecot's LDA updates some dovecot specific index/cache files even for > deliveries to maildir > => making postfix deliver to maildir without updating dovecot specific files > may solve your performance problems ...with disabling dovecot to update its indexes, sieve filters and so on. I would not configure the system this way. Kind Regards, Christian -- No signature available. -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 5306 bytes Desc: S/MIME Cryptographic Signature URL: From andy at andybev.com Tue Aug 18 09:02:29 2015 From: andy at andybev.com (Andrew Beverley) Date: Tue, 18 Aug 2015 10:02:29 +0100 Subject: Segfaults after upgrade to Debian Jessie In-Reply-To: <55D2D685.2070304@rename-it.nl> References: <1437780774.7496.7.camel@andybev.com> <55C308CD.5070308@rename-it.nl> <1438846556.14063.7.camel@andybev.com> <55C3118F.3050805@rename-it.nl> <1438848811.18074.3.camel@andybev.com> <55C36AF2.5090202@rename-it.nl> <1439813118.9225.111.camel@andybev.com> <55D1D001.9050602@rename-it.nl> <1439814936.9225.119.camel@andybev.com> <55D1E047.4040404@rename-it.nl> <1439823010.32417.5.camel@andybev.com> <55D1FFAB.7000304@rename-it.nl> <1439856257.1183.22.camel@andybev.com> <55D2D685.2070304@rename-it.nl> Message-ID: <1439888549.4205.16.camel@andybev.com> On Tue, 2015-08-18 at 08:53 +0200, Stephan Bosch wrote: > Unfortunately, Jessie is stuck at a very old version of Pigeonhole which > has this problem. Thanks Stephan. I've heard from the Debian packager that he's planning to package a newer version soon, so that will help. Andy From benjamin.cimerman at gmail.com Tue Aug 18 10:35:19 2015 From: benjamin.cimerman at gmail.com (Benjamin Cimerman) Date: Tue, 18 Aug 2015 12:35:19 +0200 Subject: pop3-login: Disconnected (tried to use disallowed plaintext auth) Message-ID: Hi I have problem with dovecot 2.2.10 - plaintext auth without SSL is not working. I'm upgrading old server and want that settings "stay same" as on old server. SSL has to be optional. /var/log/maillog *pop3-login: Disconnected (tried to use disallowed plaintext auth): user=<>, rip=_._._._, lip=_._._._, session=* my settings: [root at mail dovecot]# grep ssl * -R -n *conf.d/10-ssl.conf:8:ssl = yes* *conf.d/10-ssl.conf:40:ssl_verify_client_cert = no* *conf.d/10-auth.conf:90:auth_ssl_require_client_cert = no* *dovecot.conf:7:ssl = yes* [root at mail dovecot]# grep plain * -R -n *conf.d/10-auth.conf:11:disable_plaintext_auth = no* *conf.d/10-auth.conf:101:auth_mechanisms = plain login* Thanx you for any help. From alessio at skye.it Tue Aug 18 13:42:12 2015 From: alessio at skye.it (Alessio Cecchi) Date: Tue, 18 Aug 2015 15:42:12 +0200 Subject: NFSv4 delegation In-Reply-To: <20150709150840.GI22874@tiscali.com> References: <20150611140312.GE21542@tiscali.com> <557AAE1B.2080906@skye.it> <20150612110254.GE21520@tiscali.com> <557ABEEB.1080402@skye.it> <20150709150840.GI22874@tiscali.com> Message-ID: <55D33634.3070507@skye.it> Hi, last day I tried to enable "nfsv4 delegation" on my cluster (enable nfsv4.0 and read delegation on Netapp and remount the volume on Linux Centos 6.7 clients with Dovecot+Director). It was a failure, after some minutes the load on clients was high, and in dovecot.log I found these errors (repeated continuously): Aug 17 20:28:17 pop01eeh dovecot: imap(info at domain.com): Error: mail_index_wait_lock_fd() failed with file /home/domains/domain.com/info/Maildir/dovecot.index.log: Input/output error Aug 17 20:28:21 pop01eeh dovecot: imap(info at domain.com): Error: fcntl(/home/domains/domain.com/info/Maildir/dovecot.index.cache, write-lock, F_SETLKW) locking failed: Input/output error Is really impossible to run NFSv4 with delegation with dovecot? If it were possible the number of NFS operations would drop so much with many benefits on performance. Il 09/07/2015 17:08, brd ha scritto: > just a quick update: > > no issues with NFSv4.0 (load is slowly growing, currently ~7k mailboxes) > > instead, bad news from delegation front, we enabled it for a couple of > days but we ran in ugly issues: processes went in "uninterruptible > sleep" state, load average gets huge, reboot was the only escape :-( > > -brd > > Alessio Cecchi wrote: >> Il 12/06/2015 13:02, brd ha scritto: >>> Alessio Cecchi wrote: >>> [...] >>>> Have you already try to run NFSv4? >>> it's in place on a (very) small sample of mailboxes in dbox format, no issues >>> up to now (Debian Wheezy mainline kernel) >> >> Good to know, let me know what will be when the load will grow. >> >>>> When we switch to netapp and nfsv4 we had many problems (lock problems >>>> and instability) and we had to go immediately to NFSv3. I don't know if >>>> was a netapp problem or nfs client (Debian with 2.6 kernel), now we are >>>> using Centos 6 as NFS client and we should re-try to mount mailbox as nfsv4. >>> >>> when you've had problems, the "delegation feature" was active on Netapp >>> filers? (AFAIK it is disabled by default) >> >> Never enabled "delegation feature" on my netapp. >> >> Probably my issue was on client side. I had to switch immediately to NFSv3 >> without being able to investigate the problem. >> >> Ciao >> -- >> Alessio Cecchi >> http://www.linkedin.com/in/alessice -- Alessio Cecchi http://www.linkedin.com/in/alessice From sami.ketola at dovecot.fi Tue Aug 18 14:20:36 2015 From: sami.ketola at dovecot.fi (Sami Ketola) Date: Tue, 18 Aug 2015 16:20:36 +0200 Subject: NFSv4 delegation In-Reply-To: <55D33634.3070507@skye.it> References: <20150611140312.GE21542@tiscali.com> <557AAE1B.2080906@skye.it> <20150612110254.GE21520@tiscali.com> <557ABEEB.1080402@skye.it> <20150709150840.GI22874@tiscali.com> <55D33634.3070507@skye.it> Message-ID: Hi, Just out of curiosity what is in nfsv4 delegation that you think would give a benefit on your configuration? If I read back the thread you seem to have dovecot configured with director ring in front of the backends. In that case Dovecot already manages storage in a way that only one of the backends is accessing each users data at a time. So I can?t see anything but problems form enabling delegations. Sorry but I have zero experience in running dovegot with nfsv4 delegations since in general it is not needed. Sami > On 18 Aug 2015, at 15:42, Alessio Cecchi wrote: > > Hi, > > last day I tried to enable "nfsv4 delegation" on my cluster (enable nfsv4.0 and read delegation on Netapp and remount the volume on Linux Centos 6.7 clients with Dovecot+Director). > > It was a failure, after some minutes the load on clients was high, and in dovecot.log I found these errors (repeated continuously): > > Aug 17 20:28:17 pop01eeh dovecot: imap(info at domain.com): Error: mail_index_wait_lock_fd() failed with file /home/domains/domain.com/info/Maildir/dovecot.index.log: Input/output error > Aug 17 20:28:21 pop01eeh dovecot: imap(info at domain.com): Error: fcntl(/home/domains/domain.com/info/Maildir/dovecot.index.cache, write-lock, F_SETLKW) locking failed: Input/output error > > Is really impossible to run NFSv4 with delegation with dovecot? If it were possible the number of NFS operations would drop so much with many benefits on performance. > > Il 09/07/2015 17:08, brd ha scritto: >> just a quick update: >> >> no issues with NFSv4.0 (load is slowly growing, currently ~7k mailboxes) >> >> instead, bad news from delegation front, we enabled it for a couple of >> days but we ran in ugly issues: processes went in "uninterruptible >> sleep" state, load average gets huge, reboot was the only escape :-( >> >> -brd >> >> Alessio Cecchi wrote: >>> Il 12/06/2015 13:02, brd ha scritto: >>>> Alessio Cecchi wrote: >>>> [...] >>>>> Have you already try to run NFSv4? >>>> it's in place on a (very) small sample of mailboxes in dbox format, no issues >>>> up to now (Debian Wheezy mainline kernel) >>> >>> Good to know, let me know what will be when the load will grow. >>> >>>>> When we switch to netapp and nfsv4 we had many problems (lock problems >>>>> and instability) and we had to go immediately to NFSv3. I don't know if >>>>> was a netapp problem or nfs client (Debian with 2.6 kernel), now we are >>>>> using Centos 6 as NFS client and we should re-try to mount mailbox as nfsv4. >>>> >>>> when you've had problems, the "delegation feature" was active on Netapp >>>> filers? (AFAIK it is disabled by default) >>> >>> Never enabled "delegation feature" on my netapp. >>> >>> Probably my issue was on client side. I had to switch immediately to NFSv3 >>> without being able to investigate the problem. >>> >>> Ciao >>> -- >>> Alessio Cecchi >>> http://www.linkedin.com/in/alessice > > -- > Alessio Cecchi > http://www.linkedin.com/in/alessice From alessio at skye.it Tue Aug 18 14:46:32 2015 From: alessio at skye.it (Alessio Cecchi) Date: Tue, 18 Aug 2015 16:46:32 +0200 Subject: NFSv4 delegation In-Reply-To: References: <20150611140312.GE21542@tiscali.com> <557AAE1B.2080906@skye.it> <20150612110254.GE21520@tiscali.com> <557ABEEB.1080402@skye.it> <20150709150840.GI22874@tiscali.com> <55D33634.3070507@skye.it> Message-ID: <55D34548.1090906@skye.it> Hi, in this tipical setup (Dovecot/Director thate share Maildir via NFS) on your NFS Server you have (about) 90% of read operations and only 10% of write operations. If you see detailed stats for NFS operations you have 40-50% of GETATTR, this means that NFS/Dovecot clients are caching data (mainly dovecot index files) but they have to revalidate the cache frequently asking to the NFS Server if the file is changed (via GETATTR) but the file never change because only this client opens it. So the NFS Server is wasting operations with (unnecessary) GETATTR request. With NFSv4 and delegation you can practically eliminate these (GETATTR) requests and speed up your NFS server (instead of buy SSD disk). This because with delegation a client open file and since is the only client to open these file (that is true with Director) the NFS server delegate the client to manage the file without check if was changed (via GETATTR) on the NFS share. You can find more information (on NFSv4 and delegation) here: http://www.fsl.cs.stonybrook.edu/docs/nfs4perf/nfs4perf-login.pdf It would be very useful if NFSv4 Delegation if it worked fine with dovecot. Please talks about this with Timo. Il 18/08/2015 16:20, Sami Ketola ha scritto: > Hi, > > Just out of curiosity what is in nfsv4 delegation that you think would give a benefit on your configuration? > > If I read back the thread you seem to have dovecot configured with director ring in front of the backends. In that case Dovecot already manages storage in a way that only one of the backends is accessing each users data at a time. So I can?t see anything but problems form enabling delegations. > > Sorry but I have zero experience in running dovegot with nfsv4 delegations since in general it is not needed. > > Sami > >> On 18 Aug 2015, at 15:42, Alessio Cecchi wrote: >> >> Hi, >> >> last day I tried to enable "nfsv4 delegation" on my cluster (enable nfsv4.0 and read delegation on Netapp and remount the volume on Linux Centos 6.7 clients with Dovecot+Director). >> >> It was a failure, after some minutes the load on clients was high, and in dovecot.log I found these errors (repeated continuously): >> >> Aug 17 20:28:17 pop01eeh dovecot: imap(info at domain.com): Error: mail_index_wait_lock_fd() failed with file /home/domains/domain.com/info/Maildir/dovecot.index.log: Input/output error >> Aug 17 20:28:21 pop01eeh dovecot: imap(info at domain.com): Error: fcntl(/home/domains/domain.com/info/Maildir/dovecot.index.cache, write-lock, F_SETLKW) locking failed: Input/output error >> >> Is really impossible to run NFSv4 with delegation with dovecot? If it were possible the number of NFS operations would drop so much with many benefits on performance. >> >> Il 09/07/2015 17:08, brd ha scritto: >>> just a quick update: >>> >>> no issues with NFSv4.0 (load is slowly growing, currently ~7k mailboxes) >>> >>> instead, bad news from delegation front, we enabled it for a couple of >>> days but we ran in ugly issues: processes went in "uninterruptible >>> sleep" state, load average gets huge, reboot was the only escape :-( >>> >>> -brd >>> >>> Alessio Cecchi wrote: >>>> Il 12/06/2015 13:02, brd ha scritto: >>>>> Alessio Cecchi wrote: >>>>> [...] >>>>>> Have you already try to run NFSv4? >>>>> it's in place on a (very) small sample of mailboxes in dbox format, no issues >>>>> up to now (Debian Wheezy mainline kernel) >>>> >>>> Good to know, let me know what will be when the load will grow. >>>> >>>>>> When we switch to netapp and nfsv4 we had many problems (lock problems >>>>>> and instability) and we had to go immediately to NFSv3. I don't know if >>>>>> was a netapp problem or nfs client (Debian with 2.6 kernel), now we are >>>>>> using Centos 6 as NFS client and we should re-try to mount mailbox as nfsv4. >>>>> >>>>> when you've had problems, the "delegation feature" was active on Netapp >>>>> filers? (AFAIK it is disabled by default) >>>> >>>> Never enabled "delegation feature" on my netapp. >>>> >>>> Probably my issue was on client side. I had to switch immediately to NFSv3 >>>> without being able to investigate the problem. >>>> >>>> Ciao >>>> -- >>>> Alessio Cecchi >>>> http://www.linkedin.com/in/alessice >> >> -- >> Alessio Cecchi >> http://www.linkedin.com/in/alessice > -- Alessio Cecchi http://www.linkedin.com/in/alessice From tom at talpey.com Tue Aug 18 15:25:37 2015 From: tom at talpey.com (Tom Talpey) Date: Tue, 18 Aug 2015 11:25:37 -0400 Subject: NFSv4 delegation In-Reply-To: <55D34548.1090906@skye.it> References: <20150611140312.GE21542@tiscali.com> <557AAE1B.2080906@skye.it> <20150612110254.GE21520@tiscali.com> <557ABEEB.1080402@skye.it> <20150709150840.GI22874@tiscali.com> <55D33634.3070507@skye.it> <55D34548.1090906@skye.it> Message-ID: <55D34E71.1010808@talpey.com> On 8/18/2015 10:46 AM, Alessio Cecchi wrote: > Hi, > > in this tipical setup (Dovecot/Director thate share Maildir via NFS) on > your NFS Server you have (about) 90% of read operations and only 10% of > write operations. > > If you see detailed stats for NFS operations you have 40-50% of GETATTR, > this means that NFS/Dovecot clients are caching data (mainly dovecot > index files) but they have to revalidate the cache frequently asking to > the NFS Server if the file is changed (via GETATTR) but the file never > change because only this client opens it. > > So the NFS Server is wasting operations with (unnecessary) GETATTR request. > > With NFSv4 and delegation you can practically eliminate these (GETATTR) > requests and speed up your NFS server (instead of buy SSD disk). > > This because with delegation a client open file and since is the only > client to open these file (that is true with Director) the NFS server > delegate the client to manage the file without check if was changed (via > GETATTR) on the NFS share. Enabling delegations on Maildir is going to scale very badly. The NFS client will end up requesting one on every message open, and because an NFS server can only support a limited number of active delegations, it will be forced to constantly recall them, only to issue new short-lived ones to the next message open. While it's a decent idea for indexes, the client has no way to request delegations selectively for them. Perhaps if you used mdbox where the file count is lower, but even then, there will be many users and many mailboxes so it is important to be sure there are enough delegations available at the NFS server. That's a NetApp question in your case, and not a Dovecot one. > > You can find more information (on NFSv4 and delegation) here: > http://www.fsl.cs.stonybrook.edu/docs/nfs4perf/nfs4perf-login.pdf > > It would be very useful if NFSv4 Delegation if it worked fine with dovecot. > > Please talks about this with Timo. > > Il 18/08/2015 16:20, Sami Ketola ha scritto: >> Hi, >> >> Just out of curiosity what is in nfsv4 delegation that you think would >> give a benefit on your configuration? >> >> If I read back the thread you seem to have dovecot configured with >> director ring in front of the backends. In that case Dovecot already >> manages storage in a way that only one of the backends is accessing >> each users data at a time. So I can?t see anything but problems form >> enabling delegations. >> >> Sorry but I have zero experience in running dovegot with nfsv4 >> delegations since in general it is not needed. >> >> Sami >> >>> On 18 Aug 2015, at 15:42, Alessio Cecchi wrote: >>> >>> Hi, >>> >>> last day I tried to enable "nfsv4 delegation" on my cluster (enable >>> nfsv4.0 and read delegation on Netapp and remount the volume on Linux >>> Centos 6.7 clients with Dovecot+Director). >>> >>> It was a failure, after some minutes the load on clients was high, >>> and in dovecot.log I found these errors (repeated continuously): >>> >>> Aug 17 20:28:17 pop01eeh dovecot: imap(info at domain.com): Error: >>> mail_index_wait_lock_fd() failed with file >>> /home/domains/domain.com/info/Maildir/dovecot.index.log: Input/output >>> error >>> Aug 17 20:28:21 pop01eeh dovecot: imap(info at domain.com): Error: >>> fcntl(/home/domains/domain.com/info/Maildir/dovecot.index.cache, >>> write-lock, F_SETLKW) locking failed: Input/output error >>> >>> Is really impossible to run NFSv4 with delegation with dovecot? If it >>> were possible the number of NFS operations would drop so much with >>> many benefits on performance. >>> >>> Il 09/07/2015 17:08, brd ha scritto: >>>> just a quick update: >>>> >>>> no issues with NFSv4.0 (load is slowly growing, currently ~7k >>>> mailboxes) >>>> >>>> instead, bad news from delegation front, we enabled it for a couple of >>>> days but we ran in ugly issues: processes went in "uninterruptible >>>> sleep" state, load average gets huge, reboot was the only escape :-( >>>> >>>> -brd >>>> >>>> Alessio Cecchi wrote: >>>>> Il 12/06/2015 13:02, brd ha scritto: >>>>>> Alessio Cecchi wrote: >>>>>> [...] >>>>>>> Have you already try to run NFSv4? >>>>>> it's in place on a (very) small sample of mailboxes in dbox >>>>>> format, no issues >>>>>> up to now (Debian Wheezy mainline kernel) >>>>> >>>>> Good to know, let me know what will be when the load will grow. >>>>> >>>>>>> When we switch to netapp and nfsv4 we had many problems (lock >>>>>>> problems >>>>>>> and instability) and we had to go immediately to NFSv3. I don't >>>>>>> know if >>>>>>> was a netapp problem or nfs client (Debian with 2.6 kernel), now >>>>>>> we are >>>>>>> using Centos 6 as NFS client and we should re-try to mount >>>>>>> mailbox as nfsv4. >>>>>> >>>>>> when you've had problems, the "delegation feature" was active on >>>>>> Netapp >>>>>> filers? (AFAIK it is disabled by default) >>>>> >>>>> Never enabled "delegation feature" on my netapp. >>>>> >>>>> Probably my issue was on client side. I had to switch immediately >>>>> to NFSv3 >>>>> without being able to investigate the problem. >>>>> >>>>> Ciao >>>>> -- >>>>> Alessio Cecchi >>>>> http://www.linkedin.com/in/alessice >>> >>> -- >>> Alessio Cecchi >>> http://www.linkedin.com/in/alessice >> > From alessio at skye.it Tue Aug 18 15:37:45 2015 From: alessio at skye.it (Alessio Cecchi) Date: Tue, 18 Aug 2015 17:37:45 +0200 Subject: NFSv4 delegation In-Reply-To: <55D34E71.1010808@talpey.com> References: <20150611140312.GE21542@tiscali.com> <557AAE1B.2080906@skye.it> <20150612110254.GE21520@tiscali.com> <557ABEEB.1080402@skye.it> <20150709150840.GI22874@tiscali.com> <55D33634.3070507@skye.it> <55D34548.1090906@skye.it> <55D34E71.1010808@talpey.com> Message-ID: <55D35149.8050904@skye.it> Il 18/08/2015 17:25, Tom Talpey ha scritto: > On 8/18/2015 10:46 AM, Alessio Cecchi wrote: >> Hi, >> >> in this tipical setup (Dovecot/Director thate share Maildir via NFS) on >> your NFS Server you have (about) 90% of read operations and only 10% of >> write operations. >> >> If you see detailed stats for NFS operations you have 40-50% of GETATTR, >> this means that NFS/Dovecot clients are caching data (mainly dovecot >> index files) but they have to revalidate the cache frequently asking to >> the NFS Server if the file is changed (via GETATTR) but the file never >> change because only this client opens it. >> >> So the NFS Server is wasting operations with (unnecessary) GETATTR >> request. >> >> With NFSv4 and delegation you can practically eliminate these (GETATTR) >> requests and speed up your NFS server (instead of buy SSD disk). >> >> This because with delegation a client open file and since is the only >> client to open these file (that is true with Director) the NFS server >> delegate the client to manage the file without check if was changed (via >> GETATTR) on the NFS share. > > Enabling delegations on Maildir is going to scale very badly. The NFS > client will end up requesting one on every message open, and because an > NFS server can only support a limited number of active delegations, it > will be forced to constantly recall them, only to issue new short-lived > ones to the next message open. Sure, but in my test there wasn't a scale problem, only few users was online but the "lock" error in dovecot.log come immediately. > While it's a decent idea for indexes, the client has no way to request > delegations selectively for them. Perhaps if you used mdbox where the > file count is lower, but even then, there will be many users and many > mailboxes so it is important to be sure there are enough delegations > available at the NFS server. That's a NetApp question in your case, > and not a Dovecot one. An idea could be to have a NFS share for index only and enable delegation on it. -- Alessio Cecchi http://www.linkedin.com/in/alessice From tom at talpey.com Tue Aug 18 19:23:06 2015 From: tom at talpey.com (Tom Talpey) Date: Tue, 18 Aug 2015 15:23:06 -0400 Subject: NFSv4 delegation In-Reply-To: <55D35149.8050904@skye.it> References: <20150611140312.GE21542@tiscali.com> <557AAE1B.2080906@skye.it> <20150612110254.GE21520@tiscali.com> <557ABEEB.1080402@skye.it> <20150709150840.GI22874@tiscali.com> <55D33634.3070507@skye.it> <55D34548.1090906@skye.it> <55D34E71.1010808@talpey.com> <55D35149.8050904@skye.it> Message-ID: <55D3861A.8050002@talpey.com> On 8/18/2015 11:37 AM, Alessio Cecchi wrote: > > > Il 18/08/2015 17:25, Tom Talpey ha scritto: >> On 8/18/2015 10:46 AM, Alessio Cecchi wrote: >>> Hi, >>> >>> in this tipical setup (Dovecot/Director thate share Maildir via NFS) on >>> your NFS Server you have (about) 90% of read operations and only 10% of >>> write operations. >>> >>> If you see detailed stats for NFS operations you have 40-50% of GETATTR, >>> this means that NFS/Dovecot clients are caching data (mainly dovecot >>> index files) but they have to revalidate the cache frequently asking to >>> the NFS Server if the file is changed (via GETATTR) but the file never >>> change because only this client opens it. >>> >>> So the NFS Server is wasting operations with (unnecessary) GETATTR >>> request. >>> >>> With NFSv4 and delegation you can practically eliminate these (GETATTR) >>> requests and speed up your NFS server (instead of buy SSD disk). >>> >>> This because with delegation a client open file and since is the only >>> client to open these file (that is true with Director) the NFS server >>> delegate the client to manage the file without check if was changed (via >>> GETATTR) on the NFS share. >> >> Enabling delegations on Maildir is going to scale very badly. The NFS >> client will end up requesting one on every message open, and because an >> NFS server can only support a limited number of active delegations, it >> will be forced to constantly recall them, only to issue new short-lived >> ones to the next message open. > > Sure, but in my test there wasn't a scale problem, only few users was > online but the "lock" error in dovecot.log come immediately. Ok, but I don't see how this is a Dovecot problem. Centos is returning EIO to the lock request, you need to track down why that's happening. NFS v4.0 delegations require a callback port to be open on the client, have you verified that it's set up properly? > >> While it's a decent idea for indexes, the client has no way to request >> delegations selectively for them. Perhaps if you used mdbox where the >> file count is lower, but even then, there will be many users and many >> mailboxes so it is important to be sure there are enough delegations >> available at the NFS server. That's a NetApp question in your case, >> and not a Dovecot one. > > An idea could be to have a NFS share for index only and enable > delegation on it. From noam at maccentricsolutions.com Tue Aug 18 22:11:02 2015 From: noam at maccentricsolutions.com (Noam Birnbaum) Date: Tue, 18 Aug 2015 15:11:02 -0700 Subject: logging into user mailboxes with admin credentials Message-ID: <55D3AD76.4090600@maccentricsolutions.com> Hey folks, We are planning a migration away from Dovecot (sorry) to Google Apps and are trying to figure out how to migrate the contents of user mailboxes to the new provider. Normally we would use an IMAP migration tool to log into each user's mailbox on the old and new providers and transfer the data that way. However, there is no central directory service; all accounts are local. We have a large number of users and for both management and security reasons it's not feasible to compile a list of all IMAP passwords. Is there a way to log into user IMAP accounts on Dovecot using a central administrative credential? Thanks! Warmest and bestest, Noam 510-332-3828 cell Noam Birnbaum Mac Daddy http://www.maccentricsolutions.com 877.luv.macs x666 tweet @noamb Tech support ---> 877.luv.macs or support at maccentricsolutions.com Luvvv us? _Yelp us!_ From larryrtx at gmail.com Tue Aug 18 22:13:35 2015 From: larryrtx at gmail.com (Larry Rosenman) Date: Tue, 18 Aug 2015 17:13:35 -0500 Subject: logging into user mailboxes with admin credentials In-Reply-To: <55D3AD76.4090600@maccentricsolutions.com> References: <55D3AD76.4090600@maccentricsolutions.com> Message-ID: look at master_users http://wiki2.dovecot.org/Authentication/MasterUsers On Tue, Aug 18, 2015 at 5:11 PM, Noam Birnbaum wrote: > Hey folks, > > We are planning a migration away from Dovecot (sorry) to Google Apps and > are trying to figure out how to migrate the contents of user mailboxes to > the new provider. Normally we would use an IMAP migration tool to log into > each user's mailbox on the old and new providers and transfer the data that > way. > > However, there is no central directory service; all accounts are local. We > have a large number of users and for both management and security reasons > it's not feasible to compile a list of all IMAP passwords. > > Is there a way to log into user IMAP accounts on Dovecot using a central > administrative credential? > > Thanks! > > Warmest and bestest, > Noam > 510-332-3828 cell > > > > Noam Birnbaum > Mac Daddy > http://www.maccentricsolutions.com > 877.luv.macs x666 > tweet @noamb > > Tech support ---> 877.luv.macs or support at maccentricsolutions.com > > Luvvv us? _Yelp us!_ < > http://www.yelp.com/biz/maccentric-solutions-san-francisco> > -- Larry Rosenman http://www.lerctr.org/~ler Phone: +1 214-642-9640 (c) E-Mail: larryrtx at gmail.com US Mail: 7011 W Parmer Ln, Apt 1115, Austin, TX 78729-6961 From AchilovRN at granch.ru Wed Aug 19 04:17:47 2015 From: AchilovRN at granch.ru (=?UTF-8?B?0JDRh9C40LvQvtCyINCg0LDRiNC40LQg0J3Rg9GA0LzRg9GF0LDQvNC10LQ=?= =?UTF-8?B?0L7QstC40Yc=?=) Date: Wed, 19 Aug 2015 10:17:47 +0600 Subject: Convert from Maildir to Mdbox In-Reply-To: <55CE2675.3070003@echter-kuechen-elektro.de> References: <55C9A5A8.9010501@echter-kuechen-elektro.de> <55CC2C0F.2010709@granch.ru> <55CE2675.3070003@echter-kuechen-elektro.de> Message-ID: <55D4036B.7070706@granch.ru> 14/08/15 23:33, J. Echter ?????: >> > Hi, > > thanks for your reply. > > Would be nice to have a look at this script :) > > Thanks for your offer. > > Juergen > http://www.sheltonsoft.ru/fileZ/other/convmbx.tar.bz2 When you converting maildir-based mail boxes, you must grant write access to anyone to root mailbox folder, elsewhere conveting failed (dsync runs in user access space). I.e. when your layout is /usr/folder/mailbox1, /usr/folder/mailbox2 etc. you must grant access to /usr/folder. Script detects user home directory and placed mail onto it. commonlib.sh and colorprint.sh placed anywhere on PATH, /usr/bin is good place. Message "INBOX exist..." you can safely ignored -- ? ?????????. ?????? ????? ?????????????? (AchilovRN at granch.ru) ??????? ?????????? ?? ?????? ?????????? ??? ??? "?????", ???: +7 (383) 233-35-12, ???. 107 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2606 bytes Desc: ?????????????????????????????????? ?????????????? S/MIME URL: From nmilas at noa.gr Wed Aug 19 14:36:28 2015 From: nmilas at noa.gr (Nikolaos Milas) Date: Wed, 19 Aug 2015 17:36:28 +0300 Subject: PROXY protocol In-Reply-To: References: <54B01495.6040300@wheres5.com> Message-ID: <55D4946C.1080204@noa.gr> On 12/1/2015 10:29 ??, Francisco Wagner C. Freire wrote: > +1 > > On Fri, Jan 9, 2015 at 3:49 PM, Hoggins! wrote: > >> Hello folks, >> >> Any plans on implementing the PROXY protocol to allow Dovecot being >> behind a TCP proxy, and still logging the real IP address of the users ? >> See : http://blog.haproxy.com/haproxy/proxy-protocol/ >> >> Thanks ! >> >> Any news on that? In a lot of scenarios the haproxy PROXY protocol will be important. Please let us know if it is in the roadmap. Can you please let us know of any alternative solution(s) to pass to dovecot real client info through a haproxy server (services imap, imaps, pop3, pop3s)? Thanks in advance, Nick From stephan at rename-it.nl Wed Aug 19 14:43:41 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Wed, 19 Aug 2015 16:43:41 +0200 Subject: PROXY protocol In-Reply-To: <55D4946C.1080204@noa.gr> References: <54B01495.6040300@wheres5.com> <55D4946C.1080204@noa.gr> Message-ID: <55D4961D.2030304@rename-it.nl> Op 19-8-2015 om 16:36 schreef Nikolaos Milas: > On 12/1/2015 10:29 ??, Francisco Wagner C. Freire wrote: > >> +1 >> >> On Fri, Jan 9, 2015 at 3:49 PM, Hoggins! wrote: >> >>> Hello folks, >>> >>> Any plans on implementing the PROXY protocol to allow Dovecot being >>> behind a TCP proxy, and still logging the real IP address of the >>> users ? >>> See : http://blog.haproxy.com/haproxy/proxy-protocol/ >>> >>> Thanks ! >>> >>> > > Any news on that? > > In a lot of scenarios the haproxy PROXY protocol will be important. > > Please let us know if it is in the roadmap. > > Can you please let us know of any alternative solution(s) to pass to > dovecot real client info through a haproxy server (services imap, > imaps, pop3, pop3s)? Well... http://hg.dovecot.org/dovecot-2.2/rev/4d7a83ddb644 Regards, Stephan. From kremels at kreme.com Wed Aug 19 15:39:41 2015 From: kremels at kreme.com (@lbutlr) Date: Wed, 19 Aug 2015 09:39:41 -0600 Subject: Error: User initialization failed In-Reply-To: References: <46885658-010B-4F0C-9996-B5C47B46DDB5@kreme.com> Message-ID: > On 17 Aug 2015, at 00:54, Steffen Kaiser wrote: > >> userdb { >> args = /etc/dovecot/dovecot-sql.conf.ext >> default_fields = uid=vpopmail gid=vchkpw mail_location=/usr/local/virtual/%u mail=maildir:/usr/local/virtual/%u > > looks like this mail_location is confusing Dovecot. If you wanted to specify "home" here, be aware that home and mail location should (aka must) be separate directories. No, the problem is that dovecot does not create the root directory for virtual users. Virtual users do not have a ?home? directory. So, when adding a virtual user I have to remember to go to /usr/local/virtual and maildirmake user at domain.tld for each user. -- I listen to the wind, to the wind of my soul From paulo at matos-sorge.com Wed Aug 19 20:45:00 2015 From: paulo at matos-sorge.com (Paulo Matos) Date: Wed, 19 Aug 2015 21:45:00 +0100 Subject: Delivering email to Maildir PreINBOX Message-ID: Hi all, I have a running dovecot with: $ dovecot -n # 2.2.18: /etc/dovecot/dovecot.conf # OS: Linux 4.1.0-x86_64-linode59 x86_64 Fedora release 22 (Twenty Two) ext4 auth_debug = yes auth_mechanisms = plain login digest-md5 cram-md5 auth_verbose = yes auth_verbose_passwords = yes default_internal_user = root imap_client_workarounds = delay-newmail tb-extra-mailbox-sep mail_debug = yes mail_location = maildir:/home/vmail/%d/%n/Maildir maildir_very_dirty_syncs = yes mbox_write_locks = fcntl namespace { inbox = yes location = prefix = INBOX. separator = . type = private } namespace inbox { location = mailbox Drafts { special_use = \Drafts } mailbox Junk { special_use = \Junk } mailbox Sent { special_use = \Sent } mailbox "Sent Messages" { special_use = \Sent } mailbox Trash { special_use = \Trash } prefix = } passdb { args = /etc/dovecot/dovecot-sql.conf.ext driver = sql } postmaster_address = pmatos protocols = imap quota_full_tempfail = yes service auth { unix_listener /var/spool/postfix/private/auth { group = postfix mode = 0666 user = postfix } unix_listener auth-master { mode = 0600 user = vmail } user = $default_internal_user } ssl = required ssl_cert = References: Message-ID: <55D4F0ED.6040009@localhost.localdomain.org> On 08/19/2015 10:45 PM, Paulo Matos wrote: > > Hi all, > > I have a running dovecot with: > $ dovecot -n > # 2.2.18: /etc/dovecot/dovecot.conf > [...] > I started to filter my email with imapfilter running on the same host as > dovecot. I would therefore like to deliver email to a folder called > PreINBOX so that imapfilter then sorts the email and ends up delivering > only the useful email to INBOX. > > How can I change the name of the inbox dovecot delivers to? > > Best, > You can't change the name of the INBOX - the INBOX is always the INBOX. But you can choose the name of the target mailbox. I don't know "imapfilter". I'd use for mail filtering. Regards, Pascal From paulo at matos-sorge.com Wed Aug 19 21:42:34 2015 From: paulo at matos-sorge.com (Paulo Matos) Date: Wed, 19 Aug 2015 22:42:34 +0100 Subject: Delivering email to Maildir PreINBOX In-Reply-To: <55D4F0ED.6040009@localhost.localdomain.org> References: <55D4F0ED.6040009@localhost.localdomain.org> Message-ID: On 19/08/15 22:11, Pascal Volk wrote: > On 08/19/2015 10:45 PM, Paulo Matos wrote: >> >> Hi all, >> >> I have a running dovecot with: >> $ dovecot -n >> # 2.2.18: /etc/dovecot/dovecot.conf >> [...] >> I started to filter my email with imapfilter running on the same host as >> dovecot. I would therefore like to deliver email to a folder called >> PreINBOX so that imapfilter then sorts the email and ends up delivering >> only the useful email to INBOX. >> >> How can I change the name of the inbox dovecot delivers to? >> >> Best, >> > > You can't change the name of the INBOX - the INBOX is always the INBOX. > But you can choose the name of the target mailbox. > How do I do that? > I don't know "imapfilter". > I'd use for mail filtering. > Thanks, I will take a look at Sieve. Regards, Paulo > > Regards, > Pascal > From bluewind at xinu.at Wed Aug 19 22:37:39 2015 From: bluewind at xinu.at (Florian Pritz) Date: Thu, 20 Aug 2015 00:37:39 +0200 Subject: segfault when using zlib module with glibc 2.22 Message-ID: <20150820003739.5e3e3897@marin.server-speed.net> Hi, I'm seeing the following segfault when I run `doveadm search mailbox inbox` with the zlib module enabled with dovecot 2.18 and glibc 2.22 on Arch Linux. It does not happen with glibc 2.21. I've noticed that when I remove "zlib" from the mail_plugins line at the beginning of the config file the segfault goes away. I've also tested hg tip and the problem does not appear to be fixed yet. The segfault also happens with other commands like `doveadm purge`. Do you need any more information to reproduce and fix this? Florian backtrace: > 0x00007ffff5eaf1e0 in pthread_mutex_unlock () from /usr/lib/libpthread.so.0 > #0 0x00007ffff5eaf1e0 in pthread_mutex_unlock () from /usr/lib/libpthread.so.0 > #1 0x00007ffff7def753 in _dl_close_worker () from /lib64/ld-linux-x86-64.so.2 > #2 0x00007ffff7defede in _dl_close () from /lib64/ld-linux-x86-64.so.2 > #3 0x00007ffff7dea124 in _dl_catch_error () from /lib64/ld-linux-x86-64.so.2 > #4 0x00007ffff6d235e1 in ?? () from /usr/lib/libdl.so.2 > #5 0x00007ffff6d230bf in dlclose () from /usr/lib/libdl.so.2 > #6 0x00007ffff739326a in module_free (module=0x69da20) at module-dir.c:67 > #7 0x00007ffff7394882 in module_dir_unload (modules=0x68b680 ) at module-dir.c:593 > #8 0x00007ffff765376b in mail_storage_service_deinit (_ctx=0x6a05d0) at mail-storage-service.c:1475 > #9 0x0000000000425ac2 in doveadm_mail_cmd (cmd=0x69f780, argc=3, argv=0x6943a0) at doveadm-mail.c:631 > #10 0x0000000000425dcb in doveadm_mail_try_run (cmd_name=0x6943c9 "search", argc=3, argv=0x694398) at doveadm-mail.c:711 > #11 0x0000000000433e1f in main (argc=3, argv=0x694398) at doveadm.c:346 dovecot.conf: > # 2.0.0: dovecot.conf.orig > # OS: Linux 2.6.35-ARCH x86_64 Arch Linux > auth_mechanisms = plain login > disable_plaintext_auth = yes > mail_location = mdbox:~/.mdbox > mdbox_rotate_size = 10m > mail_plugins = $mail_plugins zlib notify mail_log > > # enable imap notify > mailbox_list_index=yes > > # remove domain part from username and lowercase it > # needed for lmtp delivery to system users > auth_username_format = %Ln > > namespace inbox { > hidden = no > inbox = yes > list = yes > location = > prefix = > separator = . > type = private > mailbox Trash { > auto = subscribe > special_use = \Trash > } > mailbox Drafts { > auto = subscribe > special_use = \Drafts > } > mailbox Sent { > auto = subscribe # autocreate, autosubscribe > special_use = \Sent > } > mailbox Junk { > auto = subscribe > special_use = \Junk > } > } > > passdb { > driver = pam > } > plugin { > sieve = ~/.dovecot.sieve > sieve_dir = ~/.sieve > sieve_global_dir = /etc/dovecot/sieve/global/ > sieve_global_path = /etc/dovecot/sieve/default.sieve > > mail_log_events = delete undelete expunge copy mailbox_delete mailbox_rename > mail_log_fields = uid box msgid size > } > protocols = imap pop3 sieve lmtp > service auth { > unix_listener auth-client { > group = postfix > mode = 0660 > user = postfix > } > user = root > } > service imap-login { > process_limit = 400 > process_min_avail = 5 > #executable = imap-login -D > #chroot = > } > > service managesieve-login { > inet_listener sieve { > #port = 4190 > } > inet_listener sieve-obsolete { > port = 2000 > } > } > service managesieve { > } > service lmtp { > unix_listener /var/spool/postfix/private/dovecot-lmtp { > group = postfix > mode = 0660 > user = postfix > } > } > > #service imap { > # executable = imap postlogin > #} > # > #service postlogin { > # executable = script-login -d rawlog > # unix_listener postlogin { > # } > #} > > login_log_format_elements = "user=<%u> method=%m rip=%r lip=%l mpid=%e %c %k" > imap_id_log = * > > ssl_cert = ssl_key = ssl_prefer_server_ciphers = yes > > ssl_protocols = !SSLv2 !SSLv3 > ssl_options = no_compression > > #ssl_cipher_list = DHE-ECDSA-AES256-GCM-SHA384:DHE-RSA-AES256-GCM-SHA384:ECDHE-ECDSA-AES256-GCM-SHA384:ECDHE-RSA-AES256-GCM-SHA384:DHE-ECDSA-AES256-SHA384:DHE-RSA-AES256-SHA384:ECDHE-ECDSA-AES256-SHA384:ECDHE-RSA-AES256-SHA384:DHE-ECDSA-AES128-GCM-SHA256:DHE-RSA-AES128-GCM-SHA256:ECDHE-ECDSA-AES128-GCM-SHA256:ECDHE-RSA-AES128-GCM-SHA256:DHE-RSA-AES256-SHA256:ECDHE-RSA-AES256-SHA256:DHE-ECDSA-AES128-SHA256:DHE-RSA-AES128-SHA256:ECDHE-ECDSA-AES128-SHA256:ECDHE-RSA-AES128-SHA256:DHE-RSA-AES256-SHA:ECDHE-RSA-AES256-SHA:DHE-RSA-AES128-SHA:EDH-RSA-DES-CBC3-SHA:DHE-RSA-AES256-SHA:DHE-RSA-AES128-SHA:ALL:!LOW:!SSLv2:!EXP:!aNULL > ssl_cipher_list = EDH+CAMELLIA:EDH+aRSA:EECDH+aRSA+AESGCM:EECDH+aRSA+SHA384:EECDH+aRSA+SHA256:EECDH:+CAMELLIA256:+AES256:+CAMELLIA128:+AES128:+SSLv3:!aNULL:!eNULL:!LOW:!3DES:!MD5:!EXP:!PSK:!DSS:!RC4:!SEED:!ECDSA:CAMELLIA256-SHA:AES256-SHA:CAMELLIA128-SHA:AES128-SHA > > #ssl_dh_parameters_length = 4096 > > userdb { > driver = passwd > } > protocol imap { > imap_client_workarounds = tb-extra-mailbox-sep > mail_max_userip_connections = 30 > mail_plugins = $mail_plugins imap_zlib > } > protocol lda { > mail_plugins = $mail_plugins sieve > postmaster_address = postmaster at server-speed.net > } > protocol lmtp { > postmaster_address = postmaster at server-speed.net > mail_plugins = $mail_plugins sieve > } > protocol sieve { > managesieve_logout_format = bytes ( in=%i : out=%o ) > } > plugin { > sieve = ~/.dovecot.sieve > sieve_dir = ~/.sieve > zlib_save_level = 6 > zlib_save = gz > #acl = vfile > } > #plugin { > #fts = lucene > ## Lucene-specific settings, good ones are: > #fts_lucene = whitespace_chars=@. > #fts_autoindex=yes > #} > > auth_failure_delay = 2 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 801 bytes Desc: OpenPGP digital signature URL: From nmilas at noa.gr Thu Aug 20 06:55:42 2015 From: nmilas at noa.gr (Nikolaos Milas) Date: Thu, 20 Aug 2015 09:55:42 +0300 Subject: PROXY protocol In-Reply-To: <55D4961D.2030304@rename-it.nl> References: <54B01495.6040300@wheres5.com> <55D4946C.1080204@noa.gr> <55D4961D.2030304@rename-it.nl> Message-ID: <55D579EE.50605@noa.gr> On 19/8/2015 5:43 ??, Stephan Bosch wrote: > Well... > > http://hg.dovecot.org/dovecot-2.2/rev/4d7a83ddb644 > > Regards, > > Stephan. That was impressive! Thank you Timo and Stephan. You are superb! I hope you will be able to provide some basic guidelines on how to enable/use the new functionality. (I am not very code-literate.) Looking forward to it! Thanks again! All the best, Nick From tim at timg.ws Thu Aug 20 07:35:40 2015 From: tim at timg.ws (Tim Groeneveld) Date: Thu, 20 Aug 2015 17:35:40 +1000 Subject: PROXY protocol In-Reply-To: <55D579EE.50605@noa.gr> References: <54B01495.6040300@wheres5.com> <55D4946C.1080204@noa.gr> <55D4961D.2030304@rename-it.nl> <55D579EE.50605@noa.gr> Message-ID: <14f4a08e2a6.ddff6e08234627.1116309388216282884@timg.ws> Hey Niko, ---- On Thu, 20 Aug 2015 16:55:42 +1000 Nikolaos Milas wrote ---- > I hope you will be able to provide some basic guidelines on how to > enable/use the new functionality. (I am not very code-literate.) Looking through the code, the functionality should not be too hard to enable using the configuration: # This is a list of trusted networks... ips are seperated by ", " # default, empty haproxy_trusted_networks = 10.1.2.0/24, 10.2.1.0/24 # This is the timeout... in seconds. # default, 3 # haproxy_timeout = 3 # modify your inet listener's to include haproxy=yes inet_listener { haproxy = yes } As for HAProxy, the configuration would look something like this: listen smtp :25 mode tcp option tcplog option smtpchk balance roundrobin server smtp1 ip.of.server1:25 check-send-proxy check inter 10s send-proxy server smtp2 ip.of.server2:25 check-send-proxy check inter 10s send-proxy Regards, Tim From skdovecot at smail.inf.fh-brs.de Thu Aug 20 07:43:56 2015 From: skdovecot at smail.inf.fh-brs.de (Steffen Kaiser) Date: Thu, 20 Aug 2015 09:43:56 +0200 (CEST) Subject: Delivering email to Maildir PreINBOX In-Reply-To: References: <55D4F0ED.6040009@localhost.localdomain.org> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wed, 19 Aug 2015, Paulo Matos wrote: >> You can't change the name of the INBOX - the INBOX is always the INBOX. >> But you can choose the name of the target mailbox. > > How do I do that? Depends on your way to deliver messages. With Dovecot LDA you could use the -m option. > >> I don't know "imapfilter". >> I'd use for mail filtering. > > Thanks, I will take a look at Sieve. With Sieve: require ["fileinto"]; fileinto "PreINBOX"; - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVdWFPHz1H7kL/d9rAQJjKQf/csSrfXGBuYcwoFla6gICePndFZpa3HkB MszL6m3FKu3SVL+Y8EB/AR+zR8C1zfRmZPQjwnGX4DRASTk4ftWX92jDnfcmPaAV 0W07BsgZWsKk1epVP6xNy1155tI73MMJSDqTNEUvFgKWa/EjSEdICkt0x9AxWI9H oI7WIPl37xqx2QRFCHH9NbUn6Z2pMNrneUBl96dPT33QZzM59Nxfvmo7oYI8SwU/ zChmv277GpF6fwgrWXryiOGr+LfXoY9uNB5YZKuK+gEOBC11/j+ypVubvds53pyV 8S430nhnY/h4j7YpihaUpjXUKFFUIhbJ/wNZpzX/9Yeumg/+JAjgGw== =cSxB -----END PGP SIGNATURE----- From nmilas at noa.gr Thu Aug 20 20:09:43 2015 From: nmilas at noa.gr (Nikolaos Milas) Date: Thu, 20 Aug 2015 23:09:43 +0300 Subject: PROXY protocol In-Reply-To: <14f4a08e2a6.ddff6e08234627.1116309388216282884@timg.ws> References: <54B01495.6040300@wheres5.com> <55D4946C.1080204@noa.gr> <55D4961D.2030304@rename-it.nl> <55D579EE.50605@noa.gr> <14f4a08e2a6.ddff6e08234627.1116309388216282884@timg.ws> Message-ID: <55D63407.10008@noa.gr> On 20/8/2015 10:35 ??, Tim Groeneveld wrote: > # This is a list of trusted networks... ips are seperated by ", " > # default, empty > haproxy_trusted_networks = 10.1.2.0/24, 10.2.1.0/24 > > # This is the timeout... in seconds. > # default, 3 > # haproxy_timeout = 3 > > # modify your inet listener's to include haproxy=yes > inet_listener { > haproxy = yes > } Thank you Tim, As soon as I manage to re-build Dovecot with the latest snapshot, I'll test it! All the best, Nick From guillaume-dovecot at lakano.org Fri Aug 21 09:49:39 2015 From: guillaume-dovecot at lakano.org (Guillaume Lakano - DOVECOT) Date: Fri, 21 Aug 2015 11:49:39 +0200 Subject: plugin antispam sa-learn run program failed with exit code -1 Message-ID: <55D6F433.3070501@lakano.org> Hello ! I've installed iRedMail on an Ubuntu Server 14.04.1 I'm not the only one on the iRedMail forum to have the problem, but the developpers of iRedMail doesn't found the problem and advice me to ask you directly. When I want to move an email directly to /Junk/ this not works. In Thunderbird, the error is : "[SERVERBUG] failed to send mail" In RoundCube, the error is : "Server Error: UID MOVE: failed to send mail ." But, if we firstly delete the email, then move the email from /Trash/ to /Junk/ this works. Here is the package versions: dovecot 2.2.9-1ubuntu2.1 dovecot-antispam 2.0+20130822-2build1 spamassassin 3.4.0-1ubuntu2.1 I have enabled antispam-debug in syslog, and the sa-learn seems to return an error: Aug 20 20:21:13 mail imap: antispam: plugin initialising (2.0-notgit) Aug 20 20:21:13 mail imap: antispam: verbose debug enabled Aug 20 20:21:13 mail imap: antispam: "Junk" is exact match spam folder Aug 20 20:21:13 mail imap: antispam: no unsure folders Aug 20 20:21:13 mail imap: antispam: "Trash" is exact match trash folder Aug 20 20:21:13 mail imap: antispam: pipe backend spam argument = --spam Aug 20 20:21:13 mail imap: antispam: pipe backend not-spam argument = --ham Aug 20 20:21:13 mail imap: antispam: pipe backend program = /usr/bin/sa-learn Aug 20 20:21:13 mail imap: antispam: pipe backend program arg[0] = -D --username=guillaume.XXXX at nydomain.com Aug 20 20:21:13 mail imap: antispam: pipe backend tmpdir /tmp Aug 20 20:21:14 mail imap: antispam: mailbox_is_unsure(Junk): 0 Aug 20 20:21:14 mail imap: antispam: mailbox_is_trash(INBOX): 0 Aug 20 20:21:14 mail imap: antispam: mailbox_is_trash(Junk): 0 Aug 20 20:21:14 mail imap: antispam: mail copy: from trash: 0, to trash: 0 Aug 20 20:21:14 mail imap: antispam: mailbox_is_spam(INBOX): 0 Aug 20 20:21:14 mail imap: antispam: mailbox_is_spam(Junk): 1 Aug 20 20:21:14 mail imap: antispam: mailbox_is_unsure(INBOX): 0 Aug 20 20:21:14 mail imap: antispam: mail copy: src spam: 0, dst spam: 1, src unsure: 0 Aug 20 20:21:14 mail imap: antispam: running mailtrain backend program /usr/bin/sa-learn Aug 20 20:21:14 mail imap: antispam: running mailtrain backend program /usr/bin/sa-learn Aug 20 20:21:14 mail imap: antispam: running mailtrain backend program parameter 1 -D --username=guillaume.XXXX at mydomain.com Aug 20 20:21:14 mail imap: antispam: running mailtrain backend program parameter 2 --spam Aug 20 20:21:16 mail imap: antispam: run program failed with exit code -1 There is no other log in /var/log/dovecot.log or /var/log/mail.log Here is my dovecot configuration: # dovecot -n # 2.2.9: /etc/dovecot/dovecot.conf # OS: Linux 3.13.0-62-generic x86_64 Ubuntu 14.04.1 LTS auth_master_user_separator = * auth_mechanisms = PLAIN LOGIN dict { acl = mysql:/etc/dovecot/dovecot-share-folder.conf quotadict = mysql:/etc/dovecot/dovecot-used-quota.conf } first_valid_uid = 2000 last_valid_uid = 2000 listen = * [::] log_path = /var/log/dovecot.log mail_gid = 2000 mail_location = maildir:%Lh/Maildir/:INDEX=%Lh/Maildir/ mail_plugins = quota mailbox_alias acl mail_uid = 2000 managesieve_notify_capability = mailto managesieve_sieve_capability = fileinto reject envelope encoded-character vacation subaddress comparator-i;ascii-numeric relational regex imap4flags copy include variables body enotify environment mailbox date ihave namespace { inbox = yes location = mailbox Drafts { auto = subscribe special_use = \Drafts } mailbox Junk { auto = subscribe special_use = \Junk } mailbox Sent { auto = subscribe special_use = \Sent } mailbox "Sent Messages" { auto = no special_use = \Sent } mailbox Spam { auto = no special_use = \Junk } mailbox Trash { auto = subscribe special_use = \Trash } prefix = separator = / type = private } namespace { list = children location = maildir:%%Lh/Maildir/:INDEX=%%Lh/Maildir/Shared/%%u prefix = Shared/%%u/ separator = / subscriptions = yes type = shared } passdb { args = /etc/dovecot/dovecot-mysql.conf driver = sql } passdb { args = /etc/dovecot/dovecot-master-users driver = passwd-file master = yes } plugin { acl = vfile acl_shared_dict = proxy::acl antispam_backend = pipe antispam_debug_target = syslog antispam_pipe_program = /usr/bin/sa-learn antispam_pipe_program_args = --username=%Lu antispam_pipe_program_notspam_arg = --ham antispam_pipe_program_spam_arg = --spam antispam_spam = Junk antispam_trash = Trash antispam_verbose_debug = 1 auth_socket_path = /var/run/dovecot/auth-master mailbox_alias_new = Sent Messages mailbox_alias_new2 = Sent Items mailbox_alias_old = Sent mailbox_alias_old2 = Sent quota = dict:user::proxy::quotadict quota_rule = *:storage=1G quota_warning = storage=85%% quota-warning 85 %u quota_warning2 = storage=90%% quota-warning 90 %u quota_warning3 = storage=95%% quota-warning 95 %u sieve = %Lh/sieve/dovecot.sieve sieve_before = /var/vmail/sieve/dovecot.sieve sieve_dir = %Lh/sieve sieve_execute_bin_dir = /usr/lib/dovecot/sieve-execute sieve_filter_bin_dir = /usr/lib/dovecot/sieve-filter sieve_global_dir = /var/vmail/sieve sieve_global_extensions = +vnd.dovecot.pipe +vnd.dovecot.filter +vnd.dovecot.execute sieve_pipe_bin_dir = /usr/lib/dovecot/sieve-pipe sieve_plugins = sieve_extprograms } protocols = pop3 imap sieve lmtp service auth { unix_listener /var/spool/postfix/private/dovecot-auth { group = postfix mode = 0666 user = postfix } unix_listener auth-master { group = vmail mode = 0666 user = vmail } unix_listener auth-userdb { group = vmail mode = 0660 user = vmail } } service dict { unix_listener dict { group = vmail mode = 0660 user = vmail } } service imap-login { process_limit = 500 service_count = 1 } service lmtp { executable = lmtp -L inet_listener lmtp { address = 127.0.0.1 port = 24 } process_min_avail = 5 unix_listener /var/spool/postfix/private/dovecot-lmtp { group = postfix mode = 0600 user = postfix } user = vmail } service managesieve-login { inet_listener sieve { address = 127.0.0.1 port = 4190 } } service pop3-login { service_count = 1 } service quota-warning { executable = script /usr/local/bin/dovecot-quota-warning.sh unix_listener quota-warning { group = vmail mode = 0660 user = vmail } } ssl = required ssl_cert = References: <54B01495.6040300@wheres5.com> <55D4946C.1080204@noa.gr> <55D4961D.2030304@rename-it.nl> <55D579EE.50605@noa.gr> <14f4a08e2a6.ddff6e08234627.1116309388216282884@timg.ws> <55D63407.10008@noa.gr> Message-ID: <55D70C15.2020204@noa.gr> On 20/8/2015 11:09 ??, Nikolaos Milas wrote: > As soon as I manage to re-build Dovecot with the latest snapshot, I'll > test it! Hello, I've built dovecot with a today snapshot from hg (dovecot-2-2-9f815e781beb) and I am trying to enable haproxy. I configured as follows (lines added compared to initial config are marked with +): + haproxy_trusted_networks = 62.217.xxx.xxx/29, 2001:648:xxx:xxx::/64 service auth { + inet_listener { + haproxy = yes + } unix_listener /var/spool/postfix/private/auth { group = postfix mode = 0660 user = postfix } unix_listener auth-master { group = vmail mode = 0660 user = vmail } user = root } service imap-login { service_count = 1 vsz_limit = 128 M } service pop3-login { service_count = 1 vsz_limit = 128 M } Dovecot starts OK and accepts connections successfully as usual, but when I add the 'send-proxy' directive on haproxy server nodes (in haproxy.cfg), clients cannot login. With pop3s, imaps, I get errors of the form: Aug 21 13:30:04 vdev dovecot: pop3-login: Disconnected (no auth attempts in 0 secs): user=<>, rip={haproxy-server-ip-address}, lip={local-dovecot-server-ip-address}, TLS handshaking: SSL_accept() failed: error:140760FC:SSL routines:SSL23_GET_CLIENT_HELLO:unknown protocol, session= Aug 21 13:30:14 vdev dovecot: imap-login: Disconnected (disconnected before auth was ready, waited 0 secs): user=<>, rip={haproxy-server-ip-address}, lip={local-dovecot-server-ip-address}, TLS handshaking: SSL_accept() failed: error:140760FC:SSL routines:SSL23_GET_CLIENT_HELLO:unknown protocol, session= Aug 21 13:30:15 vdev dovecot: imap-login: Disconnected (no auth attempts in 0 secs): user=<>, rip={haproxy-server-ip-address}, lip={local-dovecot-server-ip-address}, TLS handshaking: SSL_accept() failed: error:140760FC:SSL routines:SSL23_GET_CLIENT_HELLO:unknown protocol, session= With pop3, imap, I get failed auth messages: Aug 21 14:18:12 vdev dovecot: pop3-login: Disconnected (auth failed, 1 attempts in 14 secs): user=, method=PLAIN, rip=62.217.124.4, lip=195.251.204.232, session= Aug 21 14:20:33 vdev dovecot: auth: plain(?,{haproxy-server-ip-address},): Invalid base64 data in continued response Aug 21 14:20:38 vdev dovecot: auth: plain(?,{haproxy-server-ip-address},): Invalid base64 data in continued response Aug 21 14:20:38 vdev dovecot: imap-login: Disconnected (auth failed, 1 attempts in 0 secs): user=<>, method=PLAIN, rip={haproxy-server-ip-address}, lip={local-dovecot-server-ip-address}, session= Note: I have replaced real IP addresses with {haproxy-server-ip-address} and {local-dovecot-server-ip-address}. Should I configure things differently? Please advise. Thanks, Nick From andrew at mcnaughty.com Fri Aug 21 13:51:26 2015 From: andrew at mcnaughty.com (Andrew McN) Date: Fri, 21 Aug 2015 23:51:26 +1000 Subject: plugin antispam sa-learn run program failed with exit code -1 In-Reply-To: <55D6F433.3070501@lakano.org> References: <55D6F433.3070501@lakano.org> Message-ID: <55D72CDE.9050805@mcnaughty.com> On 21/08/15 19:49, Guillaume Lakano - DOVECOT wrote: > Aug 20 20:21:14 mail imap: antispam: running mailtrain backend program parameter 1 -D --username=guillaume.XXXX at mydomain.com > Aug 20 20:21:14 mail imap: antispam: running mailtrain backend program parameter 2 --spam That looks like it should be three parameters, not two? Also, read this: http://wiki2.dovecot.org/Plugins/Antispam#Backend_-_mailtrain_realtime_sa-learn Andrew From skdovecot at smail.inf.fh-brs.de Fri Aug 21 14:21:39 2015 From: skdovecot at smail.inf.fh-brs.de (Steffen Kaiser) Date: Fri, 21 Aug 2015 16:21:39 +0200 (CEST) Subject: plugin antispam sa-learn run program failed with exit code -1 In-Reply-To: <55D6F433.3070501@lakano.org> References: <55D6F433.3070501@lakano.org> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, 21 Aug 2015, Guillaume Lakano - DOVECOT wrote: > Aug 20 20:21:14 mail imap: antispam: running mailtrain backend program > /usr/bin/sa-learn > Aug 20 20:21:14 mail imap: antispam: running mailtrain backend program > /usr/bin/sa-learn > Aug 20 20:21:14 mail imap: antispam: running mailtrain backend program > parameter 1 -D --username=guillaume.XXXX at mydomain.com > Aug 20 20:21:14 mail imap: antispam: running mailtrain backend program > parameter 2 --spam > Aug 20 20:21:16 mail imap: antispam: run program failed with exit code -1 - -1 usually means exec() failed, that is: /usr/bin/sa-learn does not exist or the process thinks so at least. Do you have SELinux active? http://wiki2.dovecot.org/WhyDoesItNotWork near the end. If this does not help, add a wrapper script in your config and check if the script is called at least - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVdcz83z1H7kL/d9rAQKTywf/dhX5NOxUTjLQXIVAI+iOyyvLuncrG4q1 B3HMhGhRjAfGX5nvJnzSI+YFn/PTaZyrs1B8gjaNz4VyW54GcSqUiycG3NZGAv6x uLd1cjIY9UFVl2VB1uyUitVWmioNrsXzq07I5iKf9XT6h4B0XyIVCKHCWt7t9vCc HesPN+Eu6fFSRXNeP1u8Qded/iyAD4Qd/AodIBwcO1gSNgjMhrDF9phYJZHFudLZ HlMvC2SOPB3TDNUdYqdK2ZDS8v7zN6wsKrxEQorWGoqv1vRDRBczm7h+oTSmI0T9 sMo/lHtrWY19As4t9ghr3sBjpwHr+fcQnV3gAkk+x9qHCCKxjT68gw== =Hfaj -----END PGP SIGNATURE----- From guillaume-dovecot at lakano.org Fri Aug 21 14:37:32 2015 From: guillaume-dovecot at lakano.org (Guillaume Lakano - DOVECOT) Date: Fri, 21 Aug 2015 16:37:32 +0200 Subject: plugin antispam sa-learn run program failed with exit code -1 In-Reply-To: References: <55D6F433.3070501@lakano.org> Message-ID: <55D737AC.4090700@lakano.org> Le 21/08/2015 16:21, Steffen Kaiser a ?crit : > > - -1 usually means exec() failed, that is: > > /usr/bin/sa-learn does not exist or the process thinks so at least. Hello Steffen, thanks for your reply. I've tried to use an external script sa-learn-pipe.sh : > #!/bin/bash > echo /usr/bin/sa-learn $* /tmp/sendmail-msg-$$.txt echo "$$-start > ($*)" >> /tmp/sa-learn-pipe.log > echo $* > /tmp/sendmail-parms.txt cat<&0 >> /tmp/sendmail-msg-$$.txt > /usr/bin/sa-learn $* /tmp/sendmail-msg-$$.txt && rm -f > /tmp/sendmail-msg-$$.txt & > echo "$$-end" >> /tmp/sa-learn-pipe.log > exit 0 And to update my configuration like that: > > antispam_debug_target = syslog > antispam_verbose_debug = 1 > > antispam_backend = pipe > antispam_pipe_program_spam_arg = --spam > antispam_pipe_program_notspam_arg = --ham > antispam_pipe_program = /usr/local/bin/sa-learn-pipe.sh > antispam_pipe_program_args = --username=%Lu Now, this seems better, no error, the email is correctly moved in /Junk/ and I have no error > Aug 21 16:31:05 mail imap: antispam: mailbox_is_unsure(Junk): 0 > Aug 21 16:31:05 mail imap: antispam: mailbox_is_trash(INBOX): 0 > Aug 21 16:31:05 mail imap: antispam: mailbox_is_trash(Junk): 0 > Aug 21 16:31:05 mail imap: antispam: mail copy: from trash: 0, to trash: 0 > Aug 21 16:31:05 mail imap: antispam: mailbox_is_spam(INBOX): 0 > Aug 21 16:31:05 mail imap: antispam: mailbox_is_spam(Junk): 0 > Aug 21 16:31:05 mail imap: antispam: mailbox_is_unsure(INBOX): 0 > Aug 21 16:31:05 mail imap: antispam: mail copy: src spam: 0, dst spam: > 0, src unsure: 0 BUT... the sa-learn-pipe.sh doesn't seems to be called ! No files created in /tmp/ :-( I also tried to use the "mailtrain" backend but Dovecot doesn't accept it ( I suppose I need to set "antispam_backend" to "mailtrain" but this doesn't works ). > > Do you have SELinux active? http://wiki2.dovecot.org/WhyDoesItNotWork > near the end. sestatus doesn't exists, package not installed ( policycoreutils ) and no file /var/log/audit* So I suppose iRedMail doesn't install SELinux components. > > If this does not help, add a wrapper script in your config and check > if the script is called at least > It's not called at all :-/ From guillaume-dovecot at lakano.org Fri Aug 21 14:38:48 2015 From: guillaume-dovecot at lakano.org (Guillaume Lakano - DOVECOT) Date: Fri, 21 Aug 2015 16:38:48 +0200 Subject: plugin antispam sa-learn run program failed with exit code -1 In-Reply-To: <55D72CDE.9050805@mcnaughty.com> References: <55D6F433.3070501@lakano.org> <55D72CDE.9050805@mcnaughty.com> Message-ID: <55D737F8.6010206@lakano.org> Le 21/08/2015 15:51, Andrew McN a ?crit : > On 21/08/15 19:49, Guillaume Lakano - DOVECOT wrote: >> Aug 20 20:21:14 mail imap: antispam: running mailtrain backend program parameter 1 -D --username=guillaume.XXXX at mydomain.com >> Aug 20 20:21:14 mail imap: antispam: running mailtrain backend program parameter 2 --spam > That looks like it should be three parameters, not two? Hello Andrew, thanks for your answer :) Sorry about the previous copy/paste, this was a wrong previous test ( I've tested to inject others arguments to see if dovecot still report the error or no... ) Here is the the same logs : > Aug 21 16:04:37 mail imap: antispam: plugin initialising (2.0-notgit) > Aug 21 16:04:37 mail imap: antispam: verbose debug enabled > Aug 21 16:04:37 mail imap: antispam: "Junk" is exact match spam folder > Aug 21 16:04:37 mail imap: antispam: no unsure folders > Aug 21 16:04:37 mail imap: antispam: "Trash" is exact match trash folder > Aug 21 16:04:37 mail imap: antispam: pipe backend spam argument = --spam > Aug 21 16:04:37 mail imap: antispam: pipe backend not-spam argument = > --ham > Aug 21 16:04:37 mail imap: antispam: pipe backend program = > /usr/bin/sa-learn > Aug 21 16:04:37 mail imap: antispam: pipe backend program arg[0] = > --username=guillaume.XXXX at mydomain.com > Aug 21 16:04:37 mail imap: antispam: pipe backend tmpdir /tmp > Aug 21 16:04:37 mail imap: antispam: mailbox_is_unsure(Junk): 0 > Aug 21 16:04:37 mail imap: antispam: mailbox_is_trash(INBOX): 0 > Aug 21 16:04:37 mail imap: antispam: mailbox_is_trash(Junk): 0 > Aug 21 16:04:37 mail imap: antispam: mail copy: from trash: 0, to trash: 0 > Aug 21 16:04:37 mail imap: antispam: mailbox_is_spam(INBOX): 0 > Aug 21 16:04:37 mail imap: antispam: mailbox_is_spam(Junk): 1 > Aug 21 16:04:37 mail imap: antispam: mailbox_is_unsure(INBOX): 0 > Aug 21 16:04:37 mail imap: antispam: mail copy: src spam: 0, dst spam: > 1, src unsure: 0 > Aug 21 16:04:37 mail imap: antispam: running mailtrain backend program > /usr/bin/sa-learn > Aug 21 16:04:37 mail imap: antispam: running mailtrain backend program > /usr/bin/sa-learn > Aug 21 16:04:37 mail imap: antispam: running mailtrain backend program > parameter 1 --username=guillaume.XXXX at mydomain.com > Aug 21 16:04:37 mail imap: antispam: running mailtrain backend program > parameter 2 --spam > Aug 21 16:04:39 mail imap: antispam: run program failed with exit code -1 > Aug 21 16:04:46 mail imap: antispam: plugin initialising (2.0-notgit) > Aug 21 16:04:46 mail imap: antispam: verbose debug enabled > Aug 21 16:04:46 mail imap: antispam: "Junk" is exact match spam folder > Aug 21 16:04:46 mail imap: antispam: no unsure folders > Aug 21 16:04:46 mail imap: antispam: "Trash" is exact match trash folder > Aug 21 16:04:46 mail imap: antispam: pipe backend spam argument = --spam > Aug 21 16:04:46 mail imap: antispam: pipe backend not-spam argument = > --ham > Aug 21 16:04:46 mail imap: antispam: pipe backend program = > /usr/bin/sa-learn > Aug 21 16:04:46 mail imap: antispam: pipe backend program arg[0] = > --username=guillaume.XXXX at mydomain.com > Aug 21 16:04:46 mail imap: antispam: pipe backend tmpdir /tmp > Aug 21 16:04:46 mail imap: antispam: plugin initialising (2.0-notgit) > Aug 21 16:04:46 mail imap: antispam: verbose debug enabled > Aug 21 16:04:46 mail imap: antispam: "Junk" is exact match spam folder > Aug 21 16:04:46 mail imap: antispam: no unsure folders > Aug 21 16:04:46 mail imap: antispam: "Trash" is exact match trash folder > Aug 21 16:04:46 mail imap: antispam: pipe backend spam argument = --spam > Aug 21 16:04:46 mail imap: antispam: pipe backend not-spam argument = > --ham > Aug 21 16:04:46 mail imap: antispam: pipe backend program = > /usr/bin/sa-learn > Aug 21 16:04:46 mail imap: antispam: pipe backend program arg[0] = > --username=guillaume.XXXX at mydomain.com > Aug 21 16:04:46 mail imap: antispam: pipe backend tmpdir /tmp > Also, read this: > > > http://wiki2.dovecot.org/Plugins/Antispam#Backend_-_mailtrain_realtime_sa-learn Ok I'll try to use an external pipe to script Thanks -- ------------------------------------------------------------------------ *NOX DIGITAL* Guillaume Lecanu From j.echter at echter-kuechen-elektro.de Fri Aug 21 17:37:42 2015 From: j.echter at echter-kuechen-elektro.de (J. Echter) Date: Fri, 21 Aug 2015 19:37:42 +0200 Subject: plugin antispam sa-learn run program failed with exit code -1 In-Reply-To: <55D737AC.4090700@lakano.org> References: <55D6F433.3070501@lakano.org> <55D737AC.4090700@lakano.org> Message-ID: <55D761E6.8030501@echter-kuechen-elektro.de> Am 21.08.2015 um 16:37 schrieb Guillaume Lakano - DOVECOT: > Le 21/08/2015 16:21, Steffen Kaiser a ?crit : >> >> - -1 usually means exec() failed, that is: >> >> /usr/bin/sa-learn does not exist or the process thinks so at least. > > Hello Steffen, thanks for your reply. > > I've tried to use an external script sa-learn-pipe.sh : > >> #!/bin/bash >> echo /usr/bin/sa-learn $* /tmp/sendmail-msg-$$.txt echo "$$-start >> ($*)" >> /tmp/sa-learn-pipe.log >> echo $* > /tmp/sendmail-parms.txt cat<&0 >> /tmp/sendmail-msg-$$.txt >> /usr/bin/sa-learn $* /tmp/sendmail-msg-$$.txt && rm -f >> /tmp/sendmail-msg-$$.txt & >> echo "$$-end" >> /tmp/sa-learn-pipe.log >> exit 0 > > And to update my configuration like that: >> >> antispam_debug_target = syslog >> antispam_verbose_debug = 1 >> >> antispam_backend = pipe >> antispam_pipe_program_spam_arg = --spam >> antispam_pipe_program_notspam_arg = --ham >> antispam_pipe_program = /usr/local/bin/sa-learn-pipe.sh >> antispam_pipe_program_args = --username=%Lu > > Now, this seems better, no error, the email is correctly moved in > /Junk/ and I have no error > >> Aug 21 16:31:05 mail imap: antispam: mailbox_is_unsure(Junk): 0 >> Aug 21 16:31:05 mail imap: antispam: mailbox_is_trash(INBOX): 0 >> Aug 21 16:31:05 mail imap: antispam: mailbox_is_trash(Junk): 0 >> Aug 21 16:31:05 mail imap: antispam: mail copy: from trash: 0, to >> trash: 0 >> Aug 21 16:31:05 mail imap: antispam: mailbox_is_spam(INBOX): 0 >> Aug 21 16:31:05 mail imap: antispam: mailbox_is_spam(Junk): 0 >> Aug 21 16:31:05 mail imap: antispam: mailbox_is_unsure(INBOX): 0 >> Aug 21 16:31:05 mail imap: antispam: mail copy: src spam: 0, dst >> spam: 0, src unsure: 0 > > > BUT... the sa-learn-pipe.sh doesn't seems to be called ! > No files created in /tmp/ > > :-( > > I also tried to use the "mailtrain" backend but Dovecot doesn't accept > it ( I suppose I need to set "antispam_backend" to "mailtrain" but > this doesn't works ). > > > >> >> Do you have SELinux active? http://wiki2.dovecot.org/WhyDoesItNotWork >> near the end. > > sestatus doesn't exists, package not installed ( policycoreutils ) > and no file /var/log/audit* > > So I suppose iRedMail doesn't install SELinux components. > > >> >> If this does not help, add a wrapper script in your config and check >> if the script is called at least >> > > It's not called at all :-/ imho you can call sa-learn directly, no pipe neccessary. http://spamassassin.1065346.n5.nabble.com/Piping-to-sa-learn-td105920.html cheers juergen From paulo at matos-sorge.com Fri Aug 21 21:25:42 2015 From: paulo at matos-sorge.com (Paulo Matos) Date: Fri, 21 Aug 2015 22:25:42 +0100 Subject: Delivering email to Maildir PreINBOX In-Reply-To: References: <55D4F0ED.6040009@localhost.localdomain.org> Message-ID: <1440192345019-0047d6a1-bb74142e-a12fb3a2@matos-sorge.com> On Thu, Aug 20, 2015 at 8:41 am, Steffen Kaiser < skdovecot at smail.inf.fh-brs.de [skdovecot at smail.inf.fh-brs.de] > wrote: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wed, 19 Aug 2015, Paulo Matos wrote: >> You can't change the name of the INBOX - the INBOX is always the INBOX. >> But you can choose the name of the target mailbox. > > How do I do that? Depends on your way to deliver messages. With Dovecot LDA you could use the -m option. Thanks Steffen, How do you pass an option to dovecot lda. I can't really find out where it is called. There's a 15-lda file but doesnt have options to call the lda with. Thanks. > >> I don't know "imapfilter". >> I'd use for mail filtering. > > Thanks, I will take a look at Sieve. With Sieve: require ["fileinto"]; fileinto "PreINBOX"; - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVdWFPHz1H7kL/d9rAQJjKQf/csSrfXGBuYcwoFla6gICePndFZpa3HkB MszL6m3FKu3SVL+Y8EB/AR+zR8C1zfRmZPQjwnGX4DRASTk4ftWX92jDnfcmPaAV 0W07BsgZWsKk1epVP6xNy1155tI73MMJSDqTNEUvFgKWa/EjSEdICkt0x9AxWI9H oI7WIPl37xqx2QRFCHH9NbUn6Z2pMNrneUBl96dPT33QZzM59Nxfvmo7oYI8SwU/ zChmv277GpF6fwgrWXryiOGr+LfXoY9uNB5YZKuK+gEOBC11/j+ypVubvds53pyV 8S430nhnY/h4j7YpihaUpjXUKFFUIhbJ/wNZpzX/9Yeumg/+JAjgGw== =cSxB -----END PGP SIGNATURE----- From larryrtx at gmail.com Fri Aug 21 21:27:35 2015 From: larryrtx at gmail.com (Larry Rosenman) Date: Fri, 21 Aug 2015 16:27:35 -0500 Subject: Delivering email to Maildir PreINBOX In-Reply-To: <1440192345019-0047d6a1-bb74142e-a12fb3a2@matos-sorge.com> References: <55D4F0ED.6040009@localhost.localdomain.org> <1440192345019-0047d6a1-bb74142e-a12fb3a2@matos-sorge.com> Message-ID: http://wiki2.dovecot.org/LDA On Fri, Aug 21, 2015 at 4:25 PM, Paulo Matos wrote: > On Thu, Aug 20, 2015 at 8:41 am, Steffen Kaiser < > skdovecot at smail.inf.fh-brs.de [skdovecot at smail.inf.fh-brs.de] > wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On Wed, 19 Aug 2015, Paulo Matos wrote: > > >> You can't change the name of the INBOX - the INBOX is always the INBOX. > >> But you can choose the name of the target mailbox. > > > > How do I do that? > > Depends on your way to deliver messages. With Dovecot LDA you could use > the -m option. > > > > Thanks Steffen, > How do you pass an option to dovecot lda. I can't really find out where it > is > called. There's a 15-lda file but doesnt have options to call the lda with. > Thanks. > > > > > >> I don't know "imapfilter". > >> I'd use for mail filtering. > > > > Thanks, I will take a look at Sieve. > > With Sieve: > > require ["fileinto"]; > fileinto "PreINBOX"; > > - -- > Steffen Kaiser > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1 > > iQEVAwUBVdWFPHz1H7kL/d9rAQJjKQf/csSrfXGBuYcwoFla6gICePndFZpa3HkB > MszL6m3FKu3SVL+Y8EB/AR+zR8C1zfRmZPQjwnGX4DRASTk4ftWX92jDnfcmPaAV > 0W07BsgZWsKk1epVP6xNy1155tI73MMJSDqTNEUvFgKWa/EjSEdICkt0x9AxWI9H > oI7WIPl37xqx2QRFCHH9NbUn6Z2pMNrneUBl96dPT33QZzM59Nxfvmo7oYI8SwU/ > zChmv277GpF6fwgrWXryiOGr+LfXoY9uNB5YZKuK+gEOBC11/j+ypVubvds53pyV > 8S430nhnY/h4j7YpihaUpjXUKFFUIhbJ/wNZpzX/9Yeumg/+JAjgGw== > =cSxB > -----END PGP SIGNATURE----- > -- Larry Rosenman http://www.lerctr.org/~ler Phone: +1 214-642-9640 (c) E-Mail: larryrtx at gmail.com US Mail: 7011 W Parmer Ln, Apt 1115, Austin, TX 78729-6961 From stephan at rename-it.nl Sat Aug 22 07:38:27 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Sat, 22 Aug 2015 09:38:27 +0200 Subject: PROXY protocol In-Reply-To: <55D70C15.2020204@noa.gr> References: <54B01495.6040300@wheres5.com> <55D4946C.1080204@noa.gr> <55D4961D.2030304@rename-it.nl> <55D579EE.50605@noa.gr> <14f4a08e2a6.ddff6e08234627.1116309388216282884@timg.ws> <55D63407.10008@noa.gr> <55D70C15.2020204@noa.gr> Message-ID: <55D826F3.8070905@rename-it.nl> Op 8/21/2015 om 1:31 PM schreef Nikolaos Milas: > On 20/8/2015 11:09 ??, Nikolaos Milas wrote: > >> As soon as I manage to re-build Dovecot with the latest snapshot, >> I'll test it! > > Hello, > > I've built dovecot with a today snapshot from hg > (dovecot-2-2-9f815e781beb) and I am trying to enable haproxy. > > I configured as follows (lines added compared to initial config are > marked with +): > > service auth { > + inet_listener { > + haproxy = yes > + } > unix_listener /var/spool/postfix/private/auth { > group = postfix > mode = 0660 > user = postfix > } > unix_listener auth-master { > group = vmail > mode = 0660 > user = vmail > } > user = root > } Why are you putting this in the auth service? That makes no sense. This needs to be put in the services that you want to enable the haproxy protocol for. For pop and imap those are the login sevices pop3-login and imap-login. So, something like: service imap-login { service_count = 1 vsz_limit = 128 M inet_listener imap { haproxy = yes port = 143 ssl = no } } Note that this will prevent normal clients from connecting to port 143, since the server is expecting the PROXY header. It will drop the connection if it is absent. If you need to retain normal client access, e.g. for a webmail client, the haproxy listener can be put on a different port. Regards, Stephan. From cjhamono at internode.on.net Sat Aug 22 08:12:26 2015 From: cjhamono at internode.on.net (chris) Date: Sat, 22 Aug 2015 17:42:26 +0930 Subject: migrating servers Message-ID: <00cd01d0dcb2$48a50c90$d9ef25b0$@internode.on.net> Hi I am trying to migrate from a Centos dovecot 1.0.7 install to a debian dovecot 2.1.7 install. I thought simply replicating the maildirs would be fine (using rsync) but when I fetch mail from the new server using IMAP thunderbird wants to download all the messages again as duplicates. I searched the wiki over to see if I could find a solution and while it appeared I could use doveadm backup that seems to only work pushing changes from the new server to the old (wrong direction) and that program does not exist on the earlier version of dovecot on the old server. If I can be very blunt the documentation on the wiki assumes far too much knowledge from the reader. It is often hard to decipher exactly what is meant by many of the obtuse instructions. But that is an aside can anyone give me pointers on how to do this migration for all my users so they don't need to sort through hundreds of duplicated emails? Chris From nmilas at noa.gr Sat Aug 22 11:56:17 2015 From: nmilas at noa.gr (Nikolaos Milas) Date: Sat, 22 Aug 2015 14:56:17 +0300 Subject: PROXY protocol In-Reply-To: <55D826F3.8070905@rename-it.nl> References: <54B01495.6040300@wheres5.com> <55D4946C.1080204@noa.gr> <55D4961D.2030304@rename-it.nl> <55D579EE.50605@noa.gr> <14f4a08e2a6.ddff6e08234627.1116309388216282884@timg.ws> <55D63407.10008@noa.gr> <55D70C15.2020204@noa.gr> <55D826F3.8070905@rename-it.nl> Message-ID: <55D86361.1010300@noa.gr> On 22/8/2015 10:38 ??, Stephan Bosch wrote: > Why are you putting this in the auth service? That makes no sense. Sorry for my ignorance. > This needs to be put in the services that you want to enable the haproxy > protocol for. For pop and imap those are the login sevices pop3-login > and imap-login. > > ... > > Note that this will prevent normal clients from connecting to port 143, > since the server is expecting the PROXY header. It will drop the > connection if it is absent. If you need to retain normal client access, > e.g. for a webmail client, the haproxy listener can be put on a > different port. Thank you Stephan. Following your advice, I configured as follows: service imap-login { service_count = 1 vsz_limit = 128 M inet_listener { haproxy = yes port = 23001 ssl = no } inet_listener { haproxy = yes port = 23003 ssl = yes } } service pop3-login { service_count = 1 vsz_limit = 128 M inet_listener { haproxy = yes port = 23002 ssl = no } inet_listener { haproxy = yes port = 23004 ssl = yes } } ...and everything seems to be working fine. Obviously, the proxy sends requests to different ports. For example, requests received by the proxy at port 993 are sent to dovecot's port 23003 etc. If you think the config could be better/cleaner, I would appreciate your advice. One final (I hope) question: I would like to see in dovecot logs which requests come from the proxy. Can we "stamp" log entries originating from different listeners with a configurable label? Many thanks, Nick From nmilas at noa.gr Sat Aug 22 15:25:37 2015 From: nmilas at noa.gr (Nikolaos Milas) Date: Sat, 22 Aug 2015 18:25:37 +0300 Subject: PROXY protocol In-Reply-To: <55D86361.1010300@noa.gr> References: <54B01495.6040300@wheres5.com> <55D4946C.1080204@noa.gr> <55D4961D.2030304@rename-it.nl> <55D579EE.50605@noa.gr> <14f4a08e2a6.ddff6e08234627.1116309388216282884@timg.ws> <55D63407.10008@noa.gr> <55D70C15.2020204@noa.gr> <55D826F3.8070905@rename-it.nl> <55D86361.1010300@noa.gr> Message-ID: <55D89471.6040505@noa.gr> On 22/8/2015 2:56 ??, Nikolaos Milas wrote: > One final (I hope) question: I would like to see in dovecot logs which > requests come from the proxy. Can we "stamp" log entries originating > from different listeners with a configurable label? I noticed that dovecot log entries retain the proxy ip address as the "lip", so they are easily distinguishable. Yet, if it is possible to add a label per listener, it would still be useful to identify them all at once, because there may be many proxies and therefore different lip's. All the best, Nick From cjhamono at internode.on.net Sun Aug 23 06:35:37 2015 From: cjhamono at internode.on.net (Chris) Date: Sun, 23 Aug 2015 16:05:37 +0930 Subject: migrating servers In-Reply-To: <00cd01d0dcb2$48a50c90$d9ef25b0$@internode.on.net> References: <00cd01d0dcb2$48a50c90$d9ef25b0$@internode.on.net> Message-ID: <55D969B9.2000401@internode.on.net> I worked out what the problem was Because I had been planning this move for several months I was syncing home user directories including Maildir for several months unfortunately I forgot the --delete command to rsync so while users were deleting emails on the original server they were often being copied to the destination server before being deleted. Therefore there where many more emails in the folder than expected. many quite old. by resyncing with the --delete command I was able to bring the source and destination into proper sync and all the extra emails were cleaned up Of course the pop email accounts still had the duplicates I have had to advise the user to clean out there email folders and apologised for the inconvenience. Chris On 22/08/2015 5:42 PM, chris wrote: > Hi > > > > I am trying to migrate from a Centos dovecot 1.0.7 install to a debian > dovecot 2.1.7 install. > > > > I thought simply replicating the maildirs would be fine (using rsync) but > when I fetch mail from the new server using IMAP thunderbird wants to > download all the messages again as duplicates. > > > > I searched the wiki over to see if I could find a solution and while it > appeared I could use doveadm backup that seems to only work pushing changes > from the new server to the old (wrong direction) and that program does not > exist on the earlier version of dovecot on the old server. > > > > If I can be very blunt the documentation on the wiki assumes far too much > knowledge from the reader. It is often hard to decipher exactly what is > meant by many of the obtuse instructions. But that is an aside can anyone > give me pointers on how to do this migration for all my users so they don't > need to sort through hundreds of duplicated emails? > > > > Chris From paulo at matos-sorge.com Mon Aug 10 20:11:32 2015 From: paulo at matos-sorge.com (Paulo Matos) Date: Mon, 10 Aug 2015 21:11:32 +0100 Subject: Delivering email to Maildir PreINBOX Message-ID: Hi all, I have a running dovecot with: $ dovecot -n # 2.2.18: /etc/dovecot/dovecot.conf # OS: Linux 4.1.0-x86_64-linode59 x86_64 Fedora release 22 (Twenty Two) ext4 auth_debug = yes auth_mechanisms = plain login digest-md5 cram-md5 auth_verbose = yes auth_verbose_passwords = yes default_internal_user = root imap_client_workarounds = delay-newmail tb-extra-mailbox-sep mail_debug = yes mail_location = maildir:/home/vmail/%d/%n/Maildir maildir_very_dirty_syncs = yes mbox_write_locks = fcntl namespace { inbox = yes location = prefix = INBOX. separator = . type = private } namespace inbox { location = mailbox Drafts { special_use = \Drafts } mailbox Junk { special_use = \Junk } mailbox Sent { special_use = \Sent } mailbox "Sent Messages" { special_use = \Sent } mailbox Trash { special_use = \Trash } prefix = } passdb { args = /etc/dovecot/dovecot-sql.conf.ext driver = sql } postmaster_address = pmatos protocols = imap quota_full_tempfail = yes service auth { unix_listener /var/spool/postfix/private/auth { group = postfix mode = 0666 user = postfix } unix_listener auth-master { mode = 0600 user = vmail } user = $default_internal_user } ssl = required ssl_cert = Hi, I am trying to deal with the problem, that under regular mbox format a path can lead to either a file or a directory, but not both, while under IMAP, a path is usually both, a message folder containing subfolders. I found dovecot's approach to this problem, the Maildir++ layout described under http://wiki2.dovecot.org/MboxChildFolders , and was astonished to see, what problems come with Maildir++ and how complicated dovecot tries to deal with them. I wonder why dovecot is not using a much easier approach and just uses a configurable prefix and/or suffix like .dir or ____ for directory names, e.g. INBOX -> ~/mail/inbox abc -> ~/mail/abc abc/def -> ~/mail/abc.dir/def or ~/mail/___abc___/def which would make all that trouble with .subscriptions and .index much more easier. In my eyes the Maildir++ layout is overcomplicated because of the heap of workarounds to the self-created problems. regards Hadmut From mysqlstudent at gmail.com Sun Aug 23 14:37:41 2015 From: mysqlstudent at gmail.com (Alex) Date: Sun, 23 Aug 2015 10:37:41 -0400 Subject: IMAP Folder Mail Separator In-Reply-To: <55ABE981.3050506@pettijohn-web.com> References: <55ABE981.3050506@pettijohn-web.com> Message-ID: Hi, >>>> I have a fedora21 install with dovecot-2.2.18, and when creating a new >>>> subfolder with thunderbird, the actual folder name is >>>> "Folder1^Folder2" with a carat as a separator, instead of creating >>>> Folder2 in the Folder1 directory. >>>> >>>> Is this a thunderbird problem? I don't recall having this problem >>>> prior to upgrading to fedora21, so thought maybe it's a dovecot >>>> problem. >>>> >>>> Maybe the separator character is customizable? >>> >>> >>> please post your old and the current configuration: >>> >>> doveconf -n >>> >>> http://wiki2.dovecot.org/MailboxFormat/Maildir >>> and >>> http://wiki2.dovecot.org/Plugins/Listescape >>> >>> seems to apply in your situation. >> >> Perhaps I'm just not understanding, but I've spent the last few days >> reading through those links, and I just don't understand. I'm really >> hoping my doveconf output will help. I've also understand this may be >> a long-standing thunderbird bug, but it worked up until a few months >> ago after a dovecot/fedora upgrade. >> >> # 2.2.18: /etc/dovecot/dovecot.conf >> # OS: Linux 4.0.7-300.fc22.x86_64 x86_64 Fedora release 22 (Twenty Two) >> default_client_limit = 2000 >> mail_gid = mail >> mail_location = mbox:~/mail:INBOX=/var/spool/mail/%u >> mail_privileged_group = mail >> mbox_write_locks = fcntl >> namespace inbox { >> inbox = yes >> location = >> mailbox Drafts { >> special_use = \Drafts >> } >> mailbox Junk { >> special_use = \Junk >> } >> mailbox Sent { >> special_use = \Sent >> } >> mailbox "Sent Messages" { >> special_use = \Sent >> } >> mailbox Trash { >> special_use = \Trash >> } >> prefix = >> } >> passdb { >> driver = pam >> } >> passdb { >> driver = pam >> } >> protocols = imap pop3 >> service auth { >> unix_listener /var/spool/postfix/private/auth { >> group = postfix >> mode = 0660 >> user = postfix >> } >> } >> service imap-login { >> inet_listener imap { >> port = 0 >> } >> inet_listener imaps { >> port = 993 >> } >> process_min_avail = 20 >> service_count = 0 >> } >> service pop3-login { >> inet_listener pop3 { >> port = 0 >> } >> inet_listener pop3s { >> address = 192.168.1.100 >> port = 995 >> } >> } >> ssl_ca = > ssl_cert = > ssl_key = > userdb { >> driver = passwd >> } >> userdb { >> driver = passwd >> } >> >> Please let me know what other information I can provide to help >> troubleshoot this. >> >> Thanks, >> Alex > > I think this link will prove useful. > > http://wiki2.dovecot.org/MboxChildFolders Yes, that was the solution. Add 'separator = "$"' to the 'namespace inbox {' section of dovecot.conf. Many suggestions were provided, including one that said it was a long-standing bug with Thunderbird. This happened after an upgrade, so I knew it wasn't that bug. I was surprised to find so much outdated dovecot documentation. Thanks very much to Edgar and the group for helping me resolve this. Thanks, Alex From tim at timg.ws Mon Aug 24 00:29:32 2015 From: tim at timg.ws (Tim Groeneveld) Date: Mon, 24 Aug 2015 10:29:32 +1000 Subject: PROXY protocol In-Reply-To: <55D86361.1010300@noa.gr> References: <54B01495.6040300@wheres5.com> <55D4946C.1080204@noa.gr> <55D4961D.2030304@rename-it.nl> <55D579EE.50605@noa.gr> <14f4a08e2a6.ddff6e08234627.1116309388216282884@timg.ws> <55D63407.10008@noa.gr> <55D70C15.2020204@noa.gr> <55D826F3.8070905@rename-it.nl> <55D86361.1010300@noa.gr> Message-ID: <14f5d1c30e3.bbdc4206221925.5646317634641273679@timg.ws> Hey Nikolaos, ---- On Sat, 22 Aug 2015 21:56:17 +1000 Nikolaos Milas wrote ---- > ...and everything seems to be working fine. Obviously, the proxy sends > requests to different ports. For example, requests received by the proxy > at port 993 are sent to dovecot's port 23003 etc. Have you come across any issues after enabling the haproxy on the inet_listeners? Would love to hear if everything is still going OK. I would love to throw a similar configuration into production. Regards, Tim From nmilas at noa.gr Mon Aug 24 11:27:18 2015 From: nmilas at noa.gr (Nikolaos Milas) Date: Mon, 24 Aug 2015 14:27:18 +0300 Subject: PROXY protocol In-Reply-To: <14f5d1c30e3.bbdc4206221925.5646317634641273679@timg.ws> References: <54B01495.6040300@wheres5.com> <55D4946C.1080204@noa.gr> <55D4961D.2030304@rename-it.nl> <55D579EE.50605@noa.gr> <14f4a08e2a6.ddff6e08234627.1116309388216282884@timg.ws> <55D63407.10008@noa.gr> <55D70C15.2020204@noa.gr> <55D826F3.8070905@rename-it.nl> <55D86361.1010300@noa.gr> <14f5d1c30e3.bbdc4206221925.5646317634641273679@timg.ws> Message-ID: <55DAFF96.4030704@noa.gr> On 24/8/2015 3:29 ??, Tim Groeneveld wrote: > Have you come across any issues after enabling the haproxy on the inet_listeners? > > Would love to hear if everything is still going OK. I would love to throw > a similar configuration into production. Hi Tim, The configuration I described works fine in our test environment: behavior and logging is as expected. However, I have not put it into production yet, so I haven't tested under normal load. We are still building our new production environment using haproxy servers, so we are not ready to go live. Experience from tests does not indicate anticipated performance or other issues in production. If you try it, please provide info about your experience. Best regards, Nick From paulo at matos-sorge.com Mon Aug 24 13:02:18 2015 From: paulo at matos-sorge.com (Paulo Matos) Date: Mon, 24 Aug 2015 14:02:18 +0100 Subject: Delivering email to Maildir PreINBOX In-Reply-To: References: Message-ID: Please disregard the duplicate. This was due to first submitting the message while unsubscribed which means that was held on queue and only accepted for posting now. Apologies. On 10/08/15 21:11, Paulo Matos wrote: > Hi all, > > I have a running dovecot with: > $ dovecot -n > # 2.2.18: /etc/dovecot/dovecot.conf > # OS: Linux 4.1.0-x86_64-linode59 x86_64 Fedora release 22 (Twenty Two) > ext4 > auth_debug = yes > auth_mechanisms = plain login digest-md5 cram-md5 > auth_verbose = yes > auth_verbose_passwords = yes > default_internal_user = root > imap_client_workarounds = delay-newmail tb-extra-mailbox-sep > mail_debug = yes > mail_location = maildir:/home/vmail/%d/%n/Maildir > maildir_very_dirty_syncs = yes > mbox_write_locks = fcntl > namespace { > inbox = yes > location = > prefix = INBOX. > separator = . > type = private > } > namespace inbox { > location = > mailbox Drafts { > special_use = \Drafts > } > mailbox Junk { > special_use = \Junk > } > mailbox Sent { > special_use = \Sent > } > mailbox "Sent Messages" { > special_use = \Sent > } > mailbox Trash { > special_use = \Trash > } > prefix = > } > passdb { > args = /etc/dovecot/dovecot-sql.conf.ext > driver = sql > } > postmaster_address = pmatos > protocols = imap > quota_full_tempfail = yes > service auth { > unix_listener /var/spool/postfix/private/auth { > group = postfix > mode = 0666 > user = postfix > } > unix_listener auth-master { > mode = 0600 > user = vmail > } > user = $default_internal_user > } > ssl = required > ssl_cert = ssl_key = userdb { > args = uid=5000 gid=5000 home=/home/vmail/%d/%n allow_all_users=yes > driver = static > } > protocol lda { > auth_socket_path = /var/run/dovecot/auth-master > deliver_log_format = msgid=%m: %$ > log_path = /home/vmail/dovecot-deliver.log > } > protocol imap { > mail_max_userip_connections = 100 > } > > I started to filter my email with imapfilter running on the same host as > dovecot. I would therefore like to deliver email to a folder called > PreINBOX so that imapfilter then sorts the email and ends up delivering > only the useful email to INBOX. > > How can I change the name of the inbox dovecot delivers to? > > Best, > From paulo at matos-sorge.com Mon Aug 24 13:09:53 2015 From: paulo at matos-sorge.com (Paulo Matos) Date: Mon, 24 Aug 2015 14:09:53 +0100 Subject: Duplicate folder names Message-ID: Hi, I have setup message filtering and for example, I might move a message which a given sender to folder special/. This folder is created and shows up twice in INBOX/special/ and special/. They are however exactly the same folder. If I subscribe to one I am automatically subscribed to the other and viceversa. Not too happy with this configuration. I have a feeling this is related to namespaces but I am not sure in which way to configure things in order to have only INBOX/special but not special/. How can I set this up. My current configuration is: $ dovecot -n # 2.2.18: /etc/dovecot/dovecot.conf # OS: Linux 4.1.0-x86_64-linode59 x86_64 Fedora release 22 (Twenty Two) ext4 auth_debug = yes auth_mechanisms = plain login digest-md5 cram-md5 auth_verbose = yes auth_verbose_passwords = yes default_internal_user = root imap_client_workarounds = delay-newmail tb-extra-mailbox-sep mail_debug = yes mail_location = maildir:/home/vmail/%d/%n/Maildir maildir_very_dirty_syncs = yes mbox_write_locks = fcntl namespace { inbox = yes location = prefix = INBOX. separator = . type = private } namespace inbox { location = mailbox Drafts { special_use = \Drafts } mailbox Junk { special_use = \Junk } mailbox Sent { special_use = \Sent } mailbox "Sent Messages" { special_use = \Sent } mailbox Trash { special_use = \Trash } prefix = } passdb { args = /etc/dovecot/dovecot-sql.conf.ext driver = sql } postmaster_address = pmatos protocols = imap quota_full_tempfail = yes service auth { unix_listener /var/spool/postfix/private/auth { group = postfix mode = 0666 user = postfix } unix_listener auth-master { mode = 0600 user = vmail } user = $default_internal_user } ssl = required ssl_cert = References: <55D6F433.3070501@lakano.org> <55D737AC.4090700@lakano.org> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, 21 Aug 2015, Guillaume Lakano - DOVECOT wrote: > Le 21/08/2015 16:21, Steffen Kaiser a ?crit : >> >> - -1 usually means exec() failed, that is: >> >> /usr/bin/sa-learn does not exist or the process thinks so at least. > > Hello Steffen, thanks for your reply. > > I've tried to use an external script sa-learn-pipe.sh : > >> #!/bin/bash >> echo /usr/bin/sa-learn $* /tmp/sendmail-msg-$$.txt echo "$$-start ($*)" >> >> /tmp/sa-learn-pipe.log >> echo $* > /tmp/sendmail-parms.txt cat<&0 >> /tmp/sendmail-msg-$$.txt >> /usr/bin/sa-learn $* /tmp/sendmail-msg-$$.txt && rm -f >> /tmp/sendmail-msg-$$.txt & >> echo "$$-end" >> /tmp/sa-learn-pipe.log >> exit 0 Mayhap, start with something smaller: ========== /usr/local/bin/sa-learn-wrapper.sh #!/bin/bash echo "$0 $@" >> /tmp/sa-learn-wrapper.log exit 0 ========= END chmod +x /usr/local/bin/sa-learn-wrapper.sh sudo -u "user" /usr/local/bin/sa-learn-wrapper.sh 1 2 3 > And to update my configuration like that: >> >> antispam_debug_target = syslog >> antispam_verbose_debug = 1 >> >> antispam_backend = pipe >> antispam_pipe_program_spam_arg = --spam >> antispam_pipe_program_notspam_arg = --ham >> antispam_pipe_program = /usr/local/bin/sa-learn-pipe.sh >> antispam_pipe_program_args = --username=%Lu > > Now, this seems better, no error, the email is correctly moved in /Junk/ and > I have no error > >> Aug 21 16:31:05 mail imap: antispam: mailbox_is_unsure(Junk): 0 >> Aug 21 16:31:05 mail imap: antispam: mailbox_is_trash(INBOX): 0 >> Aug 21 16:31:05 mail imap: antispam: mailbox_is_trash(Junk): 0 >> Aug 21 16:31:05 mail imap: antispam: mail copy: from trash: 0, to trash: 0 >> Aug 21 16:31:05 mail imap: antispam: mailbox_is_spam(INBOX): 0 >> Aug 21 16:31:05 mail imap: antispam: mailbox_is_spam(Junk): 0 >> Aug 21 16:31:05 mail imap: antispam: mailbox_is_unsure(INBOX): 0 >> Aug 21 16:31:05 mail imap: antispam: mail copy: src spam: 0, dst spam: 0, >> src unsure: 0 This means: your config does not trigger the script, because the Junk does not match any special folder. In your 1st message mailbox_is_spam(Junk) returned true (1) > BUT... the sa-learn-pipe.sh doesn't seems to be called ! No files > created in /tmp/ you seem to have changed something else, too. > I also tried to use the "mailtrain" backend but Dovecot doesn't accept it ( I > suppose I need to set "antispam_backend" to "mailtrain" but this doesn't > works ). there are two forks of the antispam plugin with incompatible settings. > > > >> >> Do you have SELinux active? http://wiki2.dovecot.org/WhyDoesItNotWork near >> the end. > > sestatus doesn't exists, package not installed ( policycoreutils ) > and no file /var/log/audit* > > So I suppose iRedMail doesn't install SELinux components. > > >> >> If this does not help, add a wrapper script in your config and check if the >> script is called at least >> > > It's not called at all :-/ > - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVdsaynz1H7kL/d9rAQKb0Af7BCK4+GNVscOOEtuMDWgeiMyhM5mycFsv vk4qi6GR5UpOK0ClVDM2xYzuTTN7YOQchNOz9qr2XqdlN4tG7nrsGUA0nUEtYbT7 9X6W6RE1xlRCAM9fh1XSBb6Gho2vSsNFXxqM3eQ9Nv5YohR62bovjbnMbSrBcZ43 5i0La3zkWb491qS7ZdJBLTfPYz03XTA4GCcxU6CsR6hz66V3I6u5O1yTjw2Jpj7D iV+b7GuleFePccwBe5ms5Q9mY4JHec/niXlgUpI58jdVMPiMPmWjejormwBtnRJr UnNKHO3TiQVc7mtOSpHhDfD9BDCAJeTtAMARTy00IfQiYij3gto6kw== =K1/o -----END PGP SIGNATURE----- From mcbdovecot at robuust.nl Mon Aug 24 13:28:15 2015 From: mcbdovecot at robuust.nl (Maarten Bezemer) Date: Mon, 24 Aug 2015 15:28:15 +0200 (CEST) Subject: Shortcomings of Maildir++ layout In-Reply-To: <55D9918A.7080003@danisch.de> References: <55D9918A.7080003@danisch.de> Message-ID: On Sun, 23 Aug 2015, Hadmut Danisch wrote: > astonished to see, what problems come with Maildir++ and how complicated > dovecot tries to deal with them. .. [fast fwd to "solution"] > INBOX -> ~/mail/inbox > abc -> ~/mail/abc > abc/def -> ~/mail/abc.dir/def or ~/mail/___abc___/def This would not work. For example, how would you handle a folder named "abc.dir"? In your scheme, that would end up as ~/mail/abc.dir which would still result in a name clash. How would you handle abc/def/ghi? Something like ~/mail/abc.dir/def.dir/ghi or? That would not make it very workable either. Sounds more like you just introduced a new self-created problem that would require yet another workaround.. ;-) But what was not clear from your mail, is what exactly you identified as "the problems" with standard Maildir++ setup? -- Maarten From tss at iki.fi Mon Aug 24 16:12:37 2015 From: tss at iki.fi (Timo Sirainen) Date: Mon, 24 Aug 2015 19:12:37 +0300 Subject: Shortcomings of Maildir++ layout In-Reply-To: <55D9918A.7080003@danisch.de> References: <55D9918A.7080003@danisch.de> Message-ID: <65756D7C-7DD8-4BA3-AD6C-5035AAEE234F@iki.fi> On 23 Aug 2015, at 12:25, Hadmut Danisch wrote: > > Hi, > > I am trying to deal with the problem, that under regular mbox format a > path can lead to either a file or a directory, but not both, while under > IMAP, a path is usually both, a message folder containing subfolders. > > I found dovecot's approach to this problem, the Maildir++ layout > described under http://wiki2.dovecot.org/MboxChildFolders , and was > astonished to see, what problems come with Maildir++ and how complicated > dovecot tries to deal with them. > > > I wonder why dovecot is not using a much easier approach and just uses a > configurable prefix and/or suffix like .dir or ____ for directory names, > e.g. It can do it. See: http://wiki2.dovecot.org/MailLocation http://wiki2.dovecot.org/MboxChildFolders From tss at iki.fi Mon Aug 24 16:18:41 2015 From: tss at iki.fi (Timo Sirainen) Date: Mon, 24 Aug 2015 19:18:41 +0300 Subject: segfault when using zlib module with glibc 2.22 In-Reply-To: <20150820003739.5e3e3897@marin.server-speed.net> References: <20150820003739.5e3e3897@marin.server-speed.net> Message-ID: <41DA8C27-1279-4757-AD78-BF348BE2445B@iki.fi> On 20 Aug 2015, at 01:37, Florian Pritz wrote: > > Hi, > > I'm seeing the following segfault when I run `doveadm search > mailbox inbox` with the zlib module enabled with dovecot 2.18 and glibc > 2.22 on Arch Linux. It does not happen with glibc 2.21. > > I've noticed that when I remove "zlib" from the mail_plugins line at > the beginning of the config file the segfault goes away. I've also > tested hg tip and the problem does not appear to be fixed yet. > > The segfault also happens with other commands like `doveadm purge`. > > Do you need any more information to reproduce and fix this? If it only happens with Arch Linux with an updated glibc version, I doubt the problem is with Dovecot's code. >> 0x00007ffff5eaf1e0 in pthread_mutex_unlock () from /usr/lib/libpthread.so.0 >> #0 0x00007ffff5eaf1e0 in pthread_mutex_unlock () from /usr/lib/libpthread.so.0 >> #1 0x00007ffff7def753 in _dl_close_worker () from /lib64/ld-linux-x86-64.so.2 >> #2 0x00007ffff7defede in _dl_close () from /lib64/ld-linux-x86-64.so.2 >> #3 0x00007ffff7dea124 in _dl_catch_error () from /lib64/ld-linux-x86-64.so.2 >> #4 0x00007ffff6d235e1 in ?? () from /usr/lib/libdl.so.2 >> #5 0x00007ffff6d230bf in dlclose () from /usr/lib/libdl.so.2 >> #6 0x00007ffff739326a in module_free (module=0x69da20) at module-dir.c:67 Looks very much like a non-Dovecot bug. You could verify by running with valgrind, ideally with glibc debuginfo packages installed. It'll likely point to glibc cod. From tss at iki.fi Mon Aug 24 16:21:40 2015 From: tss at iki.fi (Timo Sirainen) Date: Mon, 24 Aug 2015 19:21:40 +0300 Subject: Store ACL files within shared mailbox / rename shared folder problem In-Reply-To: References: Message-ID: <2468DE78-0EBC-4F2C-9185-697093596524@iki.fi> On 17 Aug 2015, at 19:11, Chris Laif wrote: > > Hi. > > If $otheruser shares a mailbox to $myuser, my private index is stored > in *my* Maildir, according to this template (from the dovecot docs): > > mail_location = maildir:%%h/Maildir:INDEXPVT=%h/Maildir/shared/%%u > > If $otheruser renames the shared mailbox, all my private mail flags > (Seen) are lost because my mailclient couldn't find the private index > anymore. > > Is there any way to store the private index within the shared mailbox? > If $otheruser renames the mailbox, the index will be 'migrated' as > well. Something like this: > > /data/mails/$otheruser/Maildir/.MagicSharedFolder/shared/$myuser.dovecot.index.pvt.log > > I see no way on how to do craft a INDEXPVT for this. If you use LAYOUT=index it'll use mailbox GUIDs instead of mailbox names in the directories. Renaming a mailbox won't affect the GUID. But you'd need to do a somewhat larger migration then. Also I'm not sure if LAYOUT=index has been tested with shared mailboxes or with Maildir. From tss at iki.fi Mon Aug 24 16:39:57 2015 From: tss at iki.fi (Timo Sirainen) Date: Mon, 24 Aug 2015 19:39:57 +0300 Subject: ldap attribute modifiers. how to lowercase non ASCII In-Reply-To: <55B79C7D.8030504@securepoint.de> References: <55B79C7D.8030504@securepoint.de> Message-ID: <9E600F48-9B07-4F58-ABE7-F2E8AE8A74DC@iki.fi> On 28 Jul 2015, at 18:15, matthias lay wrote: > > hi all, > > I encountered some problem with lowercasing chars like an "?" > > with user_attrs like this: > > user_attrs = uid=home=/mail/%Lu, =myvar=%L{ldap:uid} > > the ? doesnt get lower cased in both cases There are various different algorithms to lowercase non-ASCII characters. Dovecot even has some code to do it if libicu is compiled in. Still, I'm not planning changing %L behavior anytime soon. Non-ASCII usernames don't work especially well now and I don't even want to try to support them well right now. That would require bigger design changes. From tss at iki.fi Mon Aug 24 23:27:48 2015 From: tss at iki.fi (Timo Sirainen) Date: Tue, 25 Aug 2015 02:27:48 +0300 Subject: IMAP hibernate feature committed Message-ID: http://hg.dovecot.org/dovecot-2.2/rev/64c73e6bd397 Today I finally committed the "imap-hibernate" feature that I first started developing about a year ago (and had been thinking about for several years before that). The main purpose here is to reduce the number of imap processes and the amount of memory they use by moving IDLEing connections into imap-hibernate processes where they are waiting for something to happen (changes to the mailbox or IMAP client to do something). Mailbox changes are noticed only if inotify/kqueue is enabled, so without them the hibernation feature won't work well. For now only the IMAP connections that are in IDLE can be hibernated. In theory non-IDLEing connections could also be hibernated, but the main problem here is that if EXPUNGEs can't be sent at the time the hibernation is over, the current code can't handle it. They would require recreating a view that contains such expunged messages, which could be possible with some extra work. Some IMAP extensions aren't supported yet: - NOTIFY: This would just need to import/export the NOTIFY state. Also hibernation could be done at any time if VANISHED replies could be sent to the selected mailbox. - SEARCH=CONTEXT: A bit more troublesome extension to support. I'm not sure how big the trouble would be. It's rarely used though. - COMPRESS: The [de]compression code would need to be moved to login process instead of being done in imap-compress plugin. This would be a larger change, although it's also useful for various other purposes. The hibernation can be enabled for now simply by setting imap_hibernate_timeout setting to non-zero, which controls how long to wait after client has issued IDLE (and nothing happens) until the process hibernates. Some future work might include: - Moving the hibernation all the way to the first Dovecot proxy layer to avoid the need for imap-hibernate process at all. (Also moving the COMPRESS handling all the way there.) - Dynamically adjusting when the client should be hibernated instead of just a fixed imap_hibernate_timeout setting. Especially try to avoid such a situation where the hibernation is almost always terminated soon after it started. Although that probably is unlikely, so maybe not worth the trouble to prevent.. - Hibernation could be used to control the load of the server. If it's running too slow or if there are already too many imap processes, some of the connections could be moved to hibernation stage. Or just kept in hibernation for a while longer even after the imap process would normally be recreated (e.g. wait for several minutes before reporting new emails, or several seconds before replying to DONE). - The EXPUNGE/VANISHED and FETCH FLAGS replies aren't sent asynchronously, but instead fully buffered to the output buffer. This works as long as the list of changes after hibernation isn't huge, but otherwise it's wasting a lot of memory. - Maybe support non-inotify/kqueue change notifications from lib-storage. This could be complicated though to do efficiently. Or a simple way would be to just send a UDP or UNIX datagram packet on all changes to all the imap-hibernate processes and see if anybody cares about those. I'm not sure how efficient that would be. From tlx at leuxner.net Tue Aug 25 07:45:34 2015 From: tlx at leuxner.net (Thomas Leuxner) Date: Tue, 25 Aug 2015 09:45:34 +0200 Subject: IMAP hibernate feature committed In-Reply-To: References: Message-ID: > http://hg.dovecot.org/dovecot-2.2/rev/64c73e6bd397 ==> /var/log/dovecot/dovecot.log <== Aug 25 09:42:07 nihlus dovecot: imap(tlx at leuxner.net): Error: net_connect_unix(/var/run/dovecot/imap-hibernate) failed: Permission denied Aug 25 09:42:07 nihlus dovecot: imap(tlx at leuxner.net): Error: Couldn't hibernate imap client: Couldn't export state: Virtual mailboxes have no GUIDs Aug 25 09:42:07 nihlus dovecot: imap(tlx at leuxner.net): Error: net_connect_unix(/var/run/dovecot/imap-hibernate) failed: Permission denied $ l /var/run/dovecot/ total 8 srw------- 1 root root 0 Aug 24 18:22 anvil srw------- 1 root root 0 Aug 24 18:22 anvil-auth-penalty srw------- 1 dovecot root 0 Aug 25 09:17 auth-client srw------- 1 dovecot root 0 Aug 25 09:17 auth-login srw------- 1 root root 0 Aug 25 09:17 auth-master -rw------- 1 root root 32 Aug 7 12:03 auth-token-secret.dat srw-rw-rw- 1 dovecot root 0 Aug 25 09:17 auth-userdb srw------- 1 doveauth root 0 Aug 25 09:17 auth-worker srw------- 1 root root 0 Aug 25 09:17 config srw------- 1 root root 0 Aug 25 09:17 dict srw------- 1 root root 0 Aug 25 09:17 director-admin srw------- 1 root root 0 Aug 25 09:17 director-userdb srw-rw-rw- 1 root root 0 Aug 25 09:17 dns-client srw------- 1 root root 0 Aug 25 09:17 doveadm-server lrwxrwxrwx 1 root root 25 Aug 24 18:22 dovecot.conf -> /etc/dovecot/dovecot.conf drwxr-xr-x 2 root root 40 Aug 7 12:03 empty srw------- 1 root root 0 Aug 25 09:17 imap-hibernate srw------- 1 root root 0 Aug 25 09:17 imap-master srw-rw-rw- 1 root root 0 Aug 25 09:17 imap-urlauth srw------- 1 dovecot root 0 Aug 25 09:17 imap-urlauth-worker srw-rw-rw- 1 root root 0 Aug 25 09:17 indexer srw------- 1 dovecot root 0 Aug 25 09:17 indexer-worker srw------- 1 root root 0 Aug 25 09:17 ipc srw-rw-rw- 1 root root 0 Aug 25 09:17 lmtp srw------- 1 root root 0 Aug 25 09:17 log-errors drwxr-x--- 2 root dovenull 120 Aug 25 09:17 login -rw------- 1 root root 6 Aug 24 18:22 master.pid srw------- 1 root root 0 Aug 25 09:17 replication-notify prw------- 1 root root 0 Aug 25 09:17 replication-notify-fifo srw------- 1 dovecot root 0 Aug 25 09:17 replicator srw-rw-rw- 1 root root 0 Aug 25 09:17 ssl-params srw------- 1 root root 0 Aug 25 09:17 stats prw------- 1 vmail root 0 Aug 25 09:43 stats-mail drwxr-x--- 2 root dovenull 80 Aug 25 09:17 token-login The socket is created as ?root? per default. Does it need to run under another user to mitigate the permissions issue? Regards Thomas -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 204 bytes Desc: Message signed with OpenPGP using GPGMail URL: From tlx at leuxner.net Tue Aug 25 08:01:36 2015 From: tlx at leuxner.net (Thomas Leuxner) Date: Tue, 25 Aug 2015 10:01:36 +0200 Subject: IMAP hibernate feature committed In-Reply-To: References: Message-ID: <20150825080136.GA47256@nihlus.leuxner.net> * Thomas Leuxner 2015.08.25 09:45: > > http://hg.dovecot.org/dovecot-2.2/rev/64c73e6bd397 > > ==> /var/log/dovecot/dovecot.log <== > Aug 25 09:42:07 nihlus dovecot: imap(tlx at leuxner.net): Error: net_connect_unix(/var/run/dovecot/imap-hibernate) failed: Permission denied > Aug 25 09:42:07 nihlus dovecot: imap(tlx at leuxner.net): Error: Couldn't hibernate imap client: Couldn't export state: Virtual mailboxes have no GUIDs > Aug 25 09:42:07 nihlus dovecot: imap(tlx at leuxner.net): Error: net_connect_unix(/var/run/dovecot/imap-hibernate) failed: Permission denied 'chmod 666' mitigates the permission issue on the socket. However it seems to have other issues then: ==> /var/log/dovecot/dovecot.log <== Aug 25 09:56:49 nihlus dovecot: imap-hibernate: Error: Failed to parse client input: Invalid peer_ip value: Aug 25 09:56:49 nihlus dovecot: imap(tlx at leuxner.net): Error: /var/run/dovecot/imap-hibernate returned failure: Failed to parse client input: Invalid peer_ip value: Aug 25 09:56:49 nihlus dovecot: imap: Error: Aug 25 09:56:49 nihlus dovecot: imap-hibernate: Error: Failed to parse client input: Invalid peer_ip value: Aug 25 09:56:49 nihlus dovecot: imap(tlx at leuxner.net): Error: /var/run/dovecot/imap-hibernate returned failure: Failed to parse client input: Invalid peer_ip value: Aug 25 09:56:49 nihlus dovecot: imap: Error: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: Digital signature URL: From skdovecot at smail.inf.fh-brs.de Tue Aug 25 09:22:25 2015 From: skdovecot at smail.inf.fh-brs.de (Steffen Kaiser) Date: Tue, 25 Aug 2015 11:22:25 +0200 (CEST) Subject: Duplicate folder names In-Reply-To: References: Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 24 Aug 2015, Paulo Matos wrote: > I have a feeling this is related to namespaces but I am not sure in which way > to configure things in order to have only INBOX/special but not special/. > > How can I set this up. My current configuration is: > $ dovecot -n > # 2.2.18: /etc/dovecot/dovecot.conf > # OS: Linux 4.1.0-x86_64-linode59 x86_64 Fedora release 22 (Twenty Two) ext4 > namespace { name: "" > inbox = yes > location = > prefix = INBOX. > separator = . > type = private > } > namespace inbox { name: "inbox" first shot: two namespaces with different names. Add "inbox" to the first namespace, too. - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVdwz0Xz1H7kL/d9rAQLDGQf/b6RjyLFPbVOeNuXA5FInvjoUc9Y7gUaY qrp8+w76x6uHokhYYRH+dANNCESXl3/uXFjSaK5G9An8KDo+IML2xfqWnBJghEzK P/GrpeDWwv/nWo6TWptPZahGH7La7bVNa1J05P13jnLLxKMpZ/MGOlFkbF7uH4pO EXLfGL7HsWELkiI/j03/V+Z3qzzxTAngSo77SggCfaAB6z1ndKMjPCZCEbXmKFmv xXJbs4IRDg8/ZWn1yPlD9qvBj/f4lfmNJC3ng7/WULGV8d9Lp9HL4abgNe93iKOg JuRUi8Lr6uelQ75XtVMjsUmCZe2Hs2R4UK98xqRa8tq/xKEU2+1GbA== =gXOW -----END PGP SIGNATURE----- From per at computer.org Tue Aug 25 09:40:36 2015 From: per at computer.org (Per Jessen) Date: Tue, 25 Aug 2015 11:40:36 +0200 Subject: sieve_after seems to ignored ? Message-ID: sieve and managesieve are both working fine, have been for years. Now I want to add a sieve_after script: /etc/dovecot/conf.d/90-sieve.conf : plugin { sieve = ~/.dovecot.sieve sieve_dir = ~/sieve sieve_before = sieve_after = ~/sieve/personal.sieve sieve_subaddress_sep = + } I have compiled ~/sieve/personal.sieve to ~/sieve/personal.svbin, but it doesn't seem to be read. Any hints? thanks Per Jessen -- Per Jessen, Z?rich (17.2?C) http://www.hostsuisse.com/ - dedicated server rental in Switzerland. From mkliewe at gmx.de Tue Aug 25 12:53:05 2015 From: mkliewe at gmx.de (Michael Kliewe) Date: Tue, 25 Aug 2015 14:53:05 +0200 Subject: sieve_after seems to ignored ? In-Reply-To: References: Message-ID: <55DC6531.5070707@gmx.de> Hi Per, On 25.08.2015 11:40, Per Jessen wrote: > sieve and managesieve are both working fine, have been for years. Now I > want to add a sieve_after script: > > /etc/dovecot/conf.d/90-sieve.conf : > > plugin { > sieve = ~/.dovecot.sieve > sieve_dir = ~/sieve > sieve_before = > sieve_after = ~/sieve/personal.sieve > sieve_subaddress_sep = + > } > > I have compiled ~/sieve/personal.sieve to ~/sieve/personal.svbin, but it > doesn't seem to be read. Any hints? sieve_after is only executed if no rule matched in other sieve-scripts before. Michael From per at computer.org Tue Aug 25 13:14:35 2015 From: per at computer.org (Per Jessen) Date: Tue, 25 Aug 2015 15:14:35 +0200 Subject: sieve_after seems to ignored ? References: <55DC6531.5070707@gmx.de> Message-ID: Michael Kliewe wrote: > Hi Per, > > On 25.08.2015 11:40, Per Jessen wrote: >> sieve and managesieve are both working fine, have been for years. Now >> I want to add a sieve_after script: >> >> /etc/dovecot/conf.d/90-sieve.conf : >> >> plugin { >> sieve = ~/.dovecot.sieve >> sieve_dir = ~/sieve >> sieve_before = >> sieve_after = ~/sieve/personal.sieve >> sieve_subaddress_sep = + >> } >> >> I have compiled ~/sieve/personal.sieve to ~/sieve/personal.svbin, but >> it >> doesn't seem to be read. Any hints? > > sieve_after is only executed if no rule matched in other sieve-scripts > before. > > Michael Thanks Michael - in this case I have no other scripts, not even a ~/.dovecot.sieve. Also, I thought sieve_after was run as long as the previous script doesn't say "keep"? http://wiki2.dovecot.org/Pigeonhole/Sieve/Configuration#multiscript -- Per Jessen, Z?rich (20.1?C) http://www.hostsuisse.com/ - dedicated server rental in Switzerland. From per at computer.org Tue Aug 25 13:18:32 2015 From: per at computer.org (Per Jessen) Date: Tue, 25 Aug 2015 15:18:32 +0200 Subject: sieve_after seems to ignored ? References: <55DC6531.5070707@gmx.de> Message-ID: Per Jessen wrote: > Michael Kliewe wrote: > >> Hi Per, >> >> On 25.08.2015 11:40, Per Jessen wrote: >>> sieve and managesieve are both working fine, have been for years. >>> Now I want to add a sieve_after script: >>> >>> /etc/dovecot/conf.d/90-sieve.conf : >>> >>> plugin { >>> sieve = ~/.dovecot.sieve >>> sieve_dir = ~/sieve >>> sieve_before = >>> sieve_after = ~/sieve/personal.sieve >>> sieve_subaddress_sep = + >>> } >>> >>> I have compiled ~/sieve/personal.sieve to ~/sieve/personal.svbin, >>> but it >>> doesn't seem to be read. Any hints? >> >> sieve_after is only executed if no rule matched in other >> sieve-scripts before. >> >> Michael > > Thanks Michael - in this case I have no other scripts, not even a > ~/.dovecot.sieve. Also, I thought sieve_after was run as long as the > previous script doesn't say "keep"? I have also tried it for a user that does have a ~/.dovecot.sieve - no rules are hit, but sieve_after still isn't executed. -- Per Jessen, Z?rich (20.1?C) http://www.dns24.ch/ - free dynamic DNS, made in Switzerland. From per at computer.org Tue Aug 25 13:49:07 2015 From: per at computer.org (Per Jessen) Date: Tue, 25 Aug 2015 15:49:07 +0200 Subject: sieve_after seems to be ignored ? References: Message-ID: Per Jessen wrote: > Per Jessen wrote: > >> Michael Kliewe wrote: >> >>> Hi Per, >>> >>> On 25.08.2015 11:40, Per Jessen wrote: >>>> sieve and managesieve are both working fine, have been for years. >>>> Now I want to add a sieve_after script: >>>> >>>> /etc/dovecot/conf.d/90-sieve.conf : >>>> >>>> plugin { >>>> sieve = ~/.dovecot.sieve >>>> sieve_dir = ~/sieve >>>> sieve_before = >>>> sieve_after = ~/sieve/personal.sieve >>>> sieve_subaddress_sep = + >>>> } >>>> >>>> I have compiled ~/sieve/personal.sieve to ~/sieve/personal.svbin, >>>> but it >>>> doesn't seem to be read. Any hints? >>> >>> sieve_after is only executed if no rule matched in other >>> sieve-scripts before. >>> >>> Michael >> >> Thanks Michael - in this case I have no other scripts, not even a >> ~/.dovecot.sieve. Also, I thought sieve_after was run as long as the >> previous script doesn't say "keep"? > > I have also tried it for a user that does have a ~/.dovecot.sieve - no > rules are hit, but sieve_after still isn't executed. As a test, I tried with an empty ~/.dovecot.sieve (contents moved to ~/sieve/personal.sieve), the sieve_after script still isn't picked up. Any suggestions on how to debug this? Thanks Per -- Per Jessen, Z?rich (20.4?C) http://www.dns24.ch/ - your free DNS host, made in Switzerland. From me at junc.eu Tue Aug 25 13:59:20 2015 From: me at junc.eu (Benny Pedersen) Date: Tue, 25 Aug 2015 15:59:20 +0200 Subject: =?UTF-8?Q?sieve=5Fafter=20seems=20to=20be=20ignored=20=3F?= In-Reply-To: References: Message-ID: Per Jessen skrev den 2015-08-25 15:49: >>>>> plugin { >>>>> sieve = ~/.dovecot.sieve >>>>> sieve_dir = ~/sieve >>>>> sieve_before = >>>>> sieve_after = ~/sieve/personal.sieve >>>>> sieve_subaddress_sep = + >>>>> } sieve is the personal, and sieve_after is the global, you have swapped names ? all that asside does both svbin exists ?, if not make sure dovecot can read them, run sievec on the source file imho global sieve files must not be in $HOME From per at computer.org Tue Aug 25 14:14:42 2015 From: per at computer.org (Per Jessen) Date: Tue, 25 Aug 2015 16:14:42 +0200 Subject: sieve_after seems to be ignored ? References: Message-ID: Benny Pedersen wrote: > Per Jessen skrev den 2015-08-25 15:49: > >>>>>> plugin { >>>>>> sieve = ~/.dovecot.sieve >>>>>> sieve_dir = ~/sieve >>>>>> sieve_before = >>>>>> sieve_after = ~/sieve/personal.sieve >>>>>> sieve_subaddress_sep = + >>>>>> } > > sieve is the personal, and sieve_after is the global, you have swapped > names ? > > all that asside does both svbin exists ?, if not make sure dovecot can > read them, run sievec on the source file Yes, it was compiled. However, I have found the problem - it was in fact a missing '}' in another config file. :-( > imho global sieve files must not be in $HOME I agree - I'm not really using a global, just a personal "after" script. I'm setting up SOGo with filters, but some advanced users have long sieve scripts that can't just be converted. So I'm moving their existing sieve scripts so ~/sieve/personal.sieve and I'm letting SOGo manage ~/.dovecot.sieve. -- Per Jessen, Z?rich (20.8?C) http://www.hostsuisse.com/ - dedicated server rental in Switzerland. From tss at iki.fi Tue Aug 25 15:28:35 2015 From: tss at iki.fi (Timo Sirainen) Date: Tue, 25 Aug 2015 18:28:35 +0300 Subject: IMAP hibernate feature committed In-Reply-To: <20150825080136.GA47256@nihlus.leuxner.net> References: <20150825080136.GA47256@nihlus.leuxner.net> Message-ID: On 25 Aug 2015, at 11:01, Thomas Leuxner wrote: > > * Thomas Leuxner 2015.08.25 09:45: > >>> http://hg.dovecot.org/dovecot-2.2/rev/64c73e6bd397 >> >> ==> /var/log/dovecot/dovecot.log <== >> Aug 25 09:42:07 nihlus dovecot: imap(tlx at leuxner.net): Error: net_connect_unix(/var/run/dovecot/imap-hibernate) failed: Permission denied >> Aug 25 09:42:07 nihlus dovecot: imap(tlx at leuxner.net): Error: Couldn't hibernate imap client: Couldn't export state: Virtual mailboxes have no GUIDs I guess virtual mailboxes should be fixed somehow. The lack of GUIDs for them has caused some other issues too .. So I guess they'll just need to be generated a GUID. >> Aug 25 09:42:07 nihlus dovecot: imap(tlx at leuxner.net): Error: net_connect_unix(/var/run/dovecot/imap-hibernate) failed: Permission denied > > 'chmod 666' mitigates the permission issue on the socket. However it seems to have other issues then: You can also change the unix_listener { user, group, mode } as needed for different services (imap, imap-hibernate). http://wiki2.dovecot.org/Services has some more info. > ==> /var/log/dovecot/dovecot.log <== > Aug 25 09:56:49 nihlus dovecot: imap-hibernate: Error: Failed to parse client input: Invalid peer_ip value: > Aug 25 09:56:49 nihlus dovecot: imap(tlx at leuxner.net): Error: /var/run/dovecot/imap-hibernate returned failure: Failed to parse client input: Invalid peer_ip value: > Aug 25 09:56:49 nihlus dovecot: imap: Error: Looks like it was broken for TLS connections. Fixed now. From nmilas at noa.gr Tue Aug 25 17:31:22 2015 From: nmilas at noa.gr (Nikolaos Milas) Date: Tue, 25 Aug 2015 20:31:22 +0300 Subject: PROXY protocol In-Reply-To: <55D86361.1010300@noa.gr> References: <54B01495.6040300@wheres5.com> <55D4946C.1080204@noa.gr> <55D4961D.2030304@rename-it.nl> <55D579EE.50605@noa.gr> <14f4a08e2a6.ddff6e08234627.1116309388216282884@timg.ws> <55D63407.10008@noa.gr> <55D70C15.2020204@noa.gr> <55D826F3.8070905@rename-it.nl> <55D86361.1010300@noa.gr> Message-ID: <55DCA66A.5040700@noa.gr> On 22/8/2015 2:56 ??, Nikolaos Milas wrote: > Can we "stamp" log entries originating from different listeners with a > configurable label? Hello, I haven't received any reply on it. Is it possible or not? Thanks, Nick From tlx at leuxner.net Tue Aug 25 17:55:29 2015 From: tlx at leuxner.net (Thomas Leuxner) Date: Tue, 25 Aug 2015 19:55:29 +0200 Subject: IMAP hibernate feature committed In-Reply-To: References: <20150825080136.GA47256@nihlus.leuxner.net> Message-ID: <20150825175529.GA13298@nihlus.leuxner.net> * Timo Sirainen 2015.08.25 17:28: > >> ==> /var/log/dovecot/dovecot.log <== > >> Aug 25 09:42:07 nihlus dovecot: imap(tlx at leuxner.net): Error: net_connect_unix(/var/run/dovecot/imap-hibernate) failed: Permission denied > >> Aug 25 09:42:07 nihlus dovecot: imap(tlx at leuxner.net): Error: Couldn't hibernate imap client: Couldn't export state: Virtual mailboxes have no GUIDs Those are completely gone with the latest two commits. I was expecting the permission error to fire up. It seems a bit too quite. If it works would it spawn a hibernate-process? It looks so from the service section, but I don't see any "hibernate" processes active. > > 'chmod 666' mitigates the permission issue on the socket. However it seems to have other issues then: > > You can also change the unix_listener { user, group, mode } as needed for different services (imap, imap-hibernate). http://wiki2.dovecot.org/Services has some more info. $ doveconf -a | grep -A 20 'service imap-hibernate' service imap-hibernate { [?] unix_listener imap-hibernate { group = mode = 0600 user = } user = $default_internal_user The question is what user it should be - or what user it should match in case several users come into play. With the standard setting $default_internal_user as above it does not work out of the box (at least with my config). Regards Thomas -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: Digital signature URL: From tss at iki.fi Tue Aug 25 20:13:32 2015 From: tss at iki.fi (Timo Sirainen) Date: Tue, 25 Aug 2015 23:13:32 +0300 Subject: PROXY protocol In-Reply-To: <55DCA66A.5040700@noa.gr> References: <54B01495.6040300@wheres5.com> <55D4946C.1080204@noa.gr> <55D4961D.2030304@rename-it.nl> <55D579EE.50605@noa.gr> <14f4a08e2a6.ddff6e08234627.1116309388216282884@timg.ws> <55D63407.10008@noa.gr> <55D70C15.2020204@noa.gr> <55D826F3.8070905@rename-it.nl> <55D86361.1010300@noa.gr> <55DCA66A.5040700@noa.gr> Message-ID: <38E58A4C-F63A-4CBB-A20A-00B138E9DC48@iki.fi> On 25 Aug 2015, at 20:31, Nikolaos Milas wrote: > > On 22/8/2015 2:56 ??, Nikolaos Milas wrote: > >> Can we "stamp" log entries originating from different listeners with a configurable label? > > Hello, > > I haven't received any reply on it. Is it possible or not? Added: http://hg.dovecot.org/dovecot-2.2/rev/a919414e68d2 From tss at iki.fi Tue Aug 25 20:21:11 2015 From: tss at iki.fi (Timo Sirainen) Date: Tue, 25 Aug 2015 23:21:11 +0300 Subject: IMAP hibernate feature committed In-Reply-To: <20150825175529.GA13298@nihlus.leuxner.net> References: <20150825080136.GA47256@nihlus.leuxner.net> <20150825175529.GA13298@nihlus.leuxner.net> Message-ID: > On 25 Aug 2015, at 20:55, Thomas Leuxner wrote: > > * Timo Sirainen 2015.08.25 17:28: > >>>> ==> /var/log/dovecot/dovecot.log <== >>>> Aug 25 09:42:07 nihlus dovecot: imap(tlx at leuxner.net): Error: net_connect_unix(/var/run/dovecot/imap-hibernate) failed: Permission denied >>>> Aug 25 09:42:07 nihlus dovecot: imap(tlx at leuxner.net): Error: Couldn't hibernate imap client: Couldn't export state: Virtual mailboxes have no GUIDs > > Those are completely gone with the latest two commits. I was expecting the permission error to fire up. It seems a bit too quite. If it works would it spawn a hibernate-process? It looks so from the service section, but I don't see any "hibernate" processes active. It no longer logs an error if the selected mailbox is virtual. It simply doesn't start up the hibernate process. If you set mail_debug=yes it'll log why it won't start the hibernation. Also just committed a change that logs the mailbox name. >>> 'chmod 666' mitigates the permission issue on the socket. However it seems to have other issues then: >> >> You can also change the unix_listener { user, group, mode } as needed for different services (imap, imap-hibernate). http://wiki2.dovecot.org/Services has some more info. > > $ doveconf -a | grep -A 20 'service imap-hibernate' > service imap-hibernate { > [?] > unix_listener imap-hibernate { > group = > mode = 0600 > user = > } > user = $default_internal_user > > The question is what user it should be - or what user it should match in case several users come into play. With the standard setting $default_internal_user as above it does not work out of the box (at least with my config). There's no good default setting here. It depends on your userdb settings and/or mail_uid setting. So for example if your imap processes are running as vmail user, you should set service imap-hibernate { unix_listener imap-hibernate { user = vmail } }. Then again if you are using system users (or otherwise multiple UIDs) it gets more difficult to implement this securely (mode=0666 works always, but security isn't too good). This same problem exists for various other parts of Dovecot, for example indexer-worker and dict services. From mark.usrey at solaray-sunglasses.com Tue Aug 25 21:44:52 2015 From: mark.usrey at solaray-sunglasses.com (Mark Usrey) Date: Tue, 25 Aug 2015 16:44:52 -0500 Subject: Shared Mailboxes Not Visible Message-ID: <55DCE1D4.70808@solaray-sunglasses.com> OK, gurus, I've spent the last 2 days crawling through all of the posts and web postings regarding shared folders, and I simply cannot get it to work. I'm sure that there is something simple that I've overlooked, but cannot for the life of me see it :) Here's my current config: # 2.2.9: /etc/dovecot/dovecot.conf # OS: Linux 3.16.0-46-generic x86_64 Ubuntu 14.04.3 LTS xfs auth_mechanisms = plain login debug_log_path = /var/log/debug.log listen = * mail_gid = mail mail_location = mdbox:/mail/data/private/%n:INDEX=/mail/index/private/%n mail_plugins = acl notify mail_log stats mail_privileged_group = mail namespace { list = children location = mdbox:/mail/data/shared/%%n:INDEXPVT=/mail/index/private/%n/shared_indices/%%n prefix = Shared/%%n/ separator = / subscriptions = no type = shared } namespace inbox { inbox = yes location = mailbox Drafts { special_use = \Drafts } mailbox Junk { special_use = \Junk } mailbox Sent { special_use = \Sent } mailbox "Sent Messages" { special_use = \Sent } mailbox Trash { special_use = \Trash } prefix = separator = / type = private } passdb { args = /etc/dovecot/dovecot-ldap.conf.ext driver = ldap } plugin { acl = vfile acl_shared_dict = file:/mail/dict/shared-mailboxes autocreate = Trash autocreate2 = Spam autocreate3 = Drafts autocreate4 = Junk autosubscribe = Trash autosubscribe2 = Spam autosubscribe3 = Drafts autosubscribe4 = Junk mail_log_events = delete undelete expunge copy mailbox_delete mailbox_rename flag_change append mail_log_fields = uid box msgid size subject stats_refresh = 30 secs stats_track_cmds = yes } protocols = " imap lmtp" service auth { unix_listener /var/spool/postfix/private/auth { user = postfix } } service dict { unix_listener dict { group = mail mode = 0600 user = root } } service imap-login { process_min_avail = 5 service_count = 0 } service imap { process_limit = 4096 } service lmtp { unix_listener /var/spool/postfix/private/dovecot-lmtp { group = postfix mode = 0600 user = postfix } } service stats { fifo_listener stats-mail { group = mail mode = 0660 user = root } } ssl_ca = From pch at myzel.net Tue Aug 25 22:22:53 2015 From: pch at myzel.net (Peter Chiochetti) Date: Wed, 26 Aug 2015 00:22:53 +0200 Subject: mail_log proposal: recipient Message-ID: <55DCEABD.5060305@myzel.net> As Timo seems to be on a committing spree; I'd like to propose that in the mail_log_fields the recipient would be available. Like probably many dovecot users, my MUAs save to Sent; so when grepping the log to learn etc. that would just be useful Kind regards -- peter From nmilas at noa.gr Wed Aug 26 09:53:38 2015 From: nmilas at noa.gr (Nikolaos Milas) Date: Wed, 26 Aug 2015 12:53:38 +0300 Subject: PROXY protocol In-Reply-To: <38E58A4C-F63A-4CBB-A20A-00B138E9DC48@iki.fi> References: <54B01495.6040300@wheres5.com> <55D4946C.1080204@noa.gr> <55D4961D.2030304@rename-it.nl> <55D579EE.50605@noa.gr> <14f4a08e2a6.ddff6e08234627.1116309388216282884@timg.ws> <55D63407.10008@noa.gr> <55D70C15.2020204@noa.gr> <55D826F3.8070905@rename-it.nl> <55D86361.1010300@noa.gr> <55DCA66A.5040700@noa.gr> <38E58A4C-F63A-4CBB-A20A-00B138E9DC48@iki.fi> Message-ID: <55DD8CA2.4000905@noa.gr> On 25/8/2015 11:13 ??, Timo Sirainen wrote: >> >> Can we "stamp" log entries originating from different listeners with a configurable label? > Added: http://hg.dovecot.org/dovecot-2.2/rev/a919414e68d2 > Thank you Timo! I am sure this feature will be helpful to all us poor sysadmins.... :-) Cheers, Nick From tlx at leuxner.net Wed Aug 26 10:33:38 2015 From: tlx at leuxner.net (Thomas Leuxner) Date: Wed, 26 Aug 2015 12:33:38 +0200 Subject: IMAP hibernate feature committed In-Reply-To: References: <20150825080136.GA47256@nihlus.leuxner.net> <20150825175529.GA13298@nihlus.leuxner.net> Message-ID: <20150826103338.GA8337@nihlus.leuxner.net> * Timo Sirainen 2015.08.25 22:21: > There's no good default setting here. It depends on your userdb settings and/or mail_uid setting. So for example if your imap processes are running as vmail user, you should set service imap-hibernate { unix_listener imap-hibernate { user = vmail } }. Then again if you are using system users (or otherwise multiple UIDs) it gets more difficult to implement this securely (mode=0666 works always, but security isn't too good). This same problem exists for various other parts of Dovecot, for example indexer-worker and dict services. I have it working (I guess) with these user settings (virtual users using 'vmail'): service imap-hibernate { unix_listener imap-hibernate { user = vmail } } I had to assign the imap-master socket the user the imap-hibernate process is using to avoid messages like this: Aug 25 23:16:02 nihlus dovecot: imap-hibernate(tlx at leuxner.net): Error: net_connect_unix(/var/run/dovecot/imap-master) failed: Permission denied Aug 25 23:16:02 nihlus dovecot: imap-hibernate(tlx at leuxner.net): Failed to connect to master socket in=126 out=944 hdr=0 body=0 del=0 exp=0 trash=0 service imap { unix_listener imap-master { user = dovecot } } With this I see messages like this in the logs: Aug 26 09:48:06 nihlus dovecot: imap-hibernate(tlx at leuxner.net): Connection closed in=189 out=4252 hdr=0 body=0 del=0 exp=0 trash=0 Aug 26 12:20:29 nihlus dovecot: imap-hibernate(tlx at leuxner.net): Connection closed in=109 out=4714 hdr=0 body=0 del=0 exp=0 trash=0 I'm a bit puzzled as to when hibernate actually kicks in because most of the time I see normal imap processes running without them being hibernated: $ ps aux | grep dovecot/imap dovenull 6791 0.0 0.0 18196 4772 ? S 06:39 0:00 dovecot/imap-login dovenull 7107 0.0 0.0 18196 4736 ? S 08:00 0:00 dovecot/imap-login dovenull 7112 0.0 0.0 18332 4492 ? S 08:00 0:00 dovecot/imap-login dovenull 7333 0.0 0.0 18332 4772 ? S 08:45 0:00 dovecot/imap-login dovenull 7675 0.0 0.0 18196 4628 ? S 10:13 0:00 dovecot/imap-login dovenull 7677 0.0 0.0 18332 4532 ? S 10:14 0:00 dovecot/imap-login dovenull 7821 0.0 0.0 18196 4532 ? S 10:44 0:00 dovecot/imap-login dovenull 8156 0.0 0.0 18196 4756 ? S 12:01 0:00 dovecot/imap-login vmail 8157 0.0 0.0 45624 9608 ? S 12:01 0:00 dovecot/imap dovenull 8158 0.0 0.0 18332 4628 ? S 12:01 0:00 dovecot/imap-login vmail 8159 0.0 0.0 44772 9256 ? S 12:01 0:00 dovecot/imap dovenull 8160 0.0 0.0 18196 4652 ? S 12:01 0:00 dovecot/imap-login vmail 8161 0.0 0.0 46072 9760 ? S 12:01 0:00 dovecot/imap dovenull 8162 0.0 0.0 18196 4548 ? S 12:01 0:00 dovecot/imap-login dovenull 8279 0.0 0.0 18332 4736 ? S 12:22 0:00 dovecot/imap-login vmail 8280 0.0 0.0 40712 5164 ? S 12:22 0:00 dovecot/imap dovenull 8341 0.0 0.0 18196 4740 ? S 12:25 0:00 dovecot/imap-login vmail 8344 0.0 0.0 46312 10568 ? S 12:25 0:00 dovecot/imap -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: Digital signature URL: From alec at alec.pl Wed Aug 26 10:54:08 2015 From: alec at alec.pl (A.L.E.C) Date: Wed, 26 Aug 2015 12:54:08 +0200 Subject: mail_log proposal: recipient In-Reply-To: <55DCEABD.5060305@myzel.net> References: <55DCEABD.5060305@myzel.net> Message-ID: <55DD9AD0.80403@alec.pl> On 08/26/2015 12:22 AM, Peter Chiochetti wrote: > As Timo seems to be on a committing spree; I'd like to propose that in > the mail_log_fields the recipient would be available. > > Like probably many dovecot users, my MUAs save to Sent; so when grepping > the log to learn etc. that would just be useful You mean content of To: Cc: and Bcc: headers or what? This don't need to match recipients used in SMTP session, so likely not so useful. -- Aleksander 'A.L.E.C' Machniak Kolab Groupware Developer [http://kolab.org] Roundcube Webmail Developer [http://roundcube.net] --------------------------------------------------- PGP: 19359DC1 @@ GG: 2275252 @@ WWW: http://alec.pl From paulo at matos-sorge.com Wed Aug 26 20:57:57 2015 From: paulo at matos-sorge.com (Paulo Matos) Date: Wed, 26 Aug 2015 21:57:57 +0100 Subject: Duplicate folder names In-Reply-To: References: Message-ID: On 25/08/15 10:22, Steffen Kaiser wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On Mon, 24 Aug 2015, Paulo Matos wrote: > >> I have a feeling this is related to namespaces but I am not sure in >> which way to configure things in order to have only INBOX/special but >> not special/. >> >> How can I set this up. My current configuration is: >> $ dovecot -n >> # 2.2.18: /etc/dovecot/dovecot.conf >> # OS: Linux 4.1.0-x86_64-linode59 x86_64 Fedora release 22 (Twenty >> Two) ext4 >> namespace { > > first shot: > two namespaces with different names. Add "inbox" to the first namespace, > too. > That worked great. Thanks, -- Paulo Matos From pch at myzel.net Wed Aug 26 22:48:02 2015 From: pch at myzel.net (Peter Chiochetti) Date: Thu, 27 Aug 2015 00:48:02 +0200 Subject: mail_log proposal: recipient In-Reply-To: <55DD9AD0.80403@alec.pl> References: <55DCEABD.5060305@myzel.net> <55DD9AD0.80403@alec.pl> Message-ID: <55DE4222.80404@myzel.net> Am 2015-08-26 um 12:54 schrieb A.L.E.C: > On 08/26/2015 12:22 AM, Peter Chiochetti wrote: >> As Timo seems to be on a committing spree; I'd like to propose that in >> the mail_log_fields the recipient would be available. >> >> Like probably many dovecot users, my MUAs save to Sent; so when grepping >> the log to learn etc. that would just be useful > > You mean content of To: Cc: and Bcc: headers or what? This don't need to > match recipients used in SMTP session, so likely not so useful. Thank you, I only later realized, that there may be more than one recipient of a message; Probably the first "To" header, fully or up to the first comma, if any, would be appropriate. (The log isnt csv.) -- peter From paulo at matos-sorge.com Thu Aug 27 09:07:43 2015 From: paulo at matos-sorge.com (Paulo Matos) Date: Thu, 27 Aug 2015 10:07:43 +0100 Subject: Mailbox can't be created Message-ID: Hi, I have setup postfix master.cf to use dovecot-lda to deliver (using -m) all messages to PreInbox: dovecot unix - n n - - pipe flags=DRhu user=vmail:vmail argv=/usr/libexec/dovecot/deliver -f ${sender} -a ${recipient} -d ${user}@${nexthop} -m PreINBOX And enabled lda_mailbox_autocreate. The message I have now it: Aug 27 10:02:29 lda(): Info: msgid=: save failed to open mailbox PreINBOX: Mailbox can't be created Aug 27 10:02:29 lda(): Info: msgid=: saved mail to INBOX Why wouldn't dovecot itself be able to create mailboxes? I am using virtual domains and users on a mysql database through postfix and mailboxes saved in MailDir format. Configuration of dovecot is: # 2.2.18: /etc/dovecot/dovecot.conf # OS: Linux 4.1.5-x86_64-linode61 x86_64 Fedora release 22 (Twenty Two) ext4 auth_debug = yes auth_mechanisms = plain login digest-md5 cram-md5 auth_verbose = yes auth_verbose_passwords = yes default_internal_user = root imap_client_workarounds = delay-newmail tb-extra-mailbox-sep lda_mailbox_autocreate = yes mail_debug = yes mail_location = maildir:/home/vmail/%d/%n/Maildir maildir_very_dirty_syncs = yes mbox_write_locks = fcntl namespace inbox { inbox = yes location = mailbox Drafts { special_use = \Drafts } mailbox Junk { special_use = \Junk } mailbox Sent { special_use = \Sent } mailbox "Sent Messages" { special_use = \Sent } mailbox Trash { special_use = \Trash } prefix = INBOX. separator = . type = private } passdb { args = /etc/dovecot/dovecot-sql.conf.ext driver = sql } postmaster_address = pmatos protocols = imap quota_full_tempfail = yes service auth { unix_listener /var/spool/postfix/private/auth { group = postfix mode = 0666 user = postfix } unix_listener auth-master { mode = 0600 user = vmail } user = $default_internal_user } ssl = required ssl_cert = References: Message-ID: Please check permissions as this sounds like a permissions issue m On 27/08/2015 12:08 PM, "Paulo Matos" wrote: > Hi, > > I have setup postfix master.cf to use dovecot-lda to deliver (using -m) > all messages to PreInbox: > dovecot unix - n n - - pipe > flags=DRhu user=vmail:vmail argv=/usr/libexec/dovecot/deliver -f > ${sender} -a ${recipient} -d ${user}@${nexthop} -m PreINBOX > > And enabled lda_mailbox_autocreate. > > The message I have now it: > Aug 27 10:02:29 lda(): Info: > msgid=: save > failed to open mailbox PreINBOX: Mailbox can't be created > Aug 27 10:02:29 lda(): Info: > msgid=: saved > mail to INBOX > > Why wouldn't dovecot itself be able to create mailboxes? > > I am using virtual domains and users on a mysql database through postfix > and mailboxes saved in MailDir format. > > Configuration of dovecot is: > # 2.2.18: /etc/dovecot/dovecot.conf > # OS: Linux 4.1.5-x86_64-linode61 x86_64 Fedora release 22 (Twenty Two) > ext4 > auth_debug = yes > auth_mechanisms = plain login digest-md5 cram-md5 > auth_verbose = yes > auth_verbose_passwords = yes > default_internal_user = root > imap_client_workarounds = delay-newmail tb-extra-mailbox-sep > lda_mailbox_autocreate = yes > mail_debug = yes > mail_location = maildir:/home/vmail/%d/%n/Maildir > maildir_very_dirty_syncs = yes > mbox_write_locks = fcntl > namespace inbox { > inbox = yes > location = > mailbox Drafts { > special_use = \Drafts > } > mailbox Junk { > special_use = \Junk > } > mailbox Sent { > special_use = \Sent > } > mailbox "Sent Messages" { > special_use = \Sent > } > mailbox Trash { > special_use = \Trash > } > prefix = INBOX. > separator = . > type = private > } > passdb { > args = /etc/dovecot/dovecot-sql.conf.ext > driver = sql > } > postmaster_address = pmatos > protocols = imap > quota_full_tempfail = yes > service auth { > unix_listener /var/spool/postfix/private/auth { > group = postfix > mode = 0666 > user = postfix > } > unix_listener auth-master { > mode = 0600 > user = vmail > } > user = $default_internal_user > } > ssl = required > ssl_cert = ssl_key = userdb { > args = uid=5000 gid=5000 home=/home/vmail/%d/%n allow_all_users=yes > driver = static > } > protocol lda { > auth_socket_path = /var/run/dovecot/auth-master > deliver_log_format = msgid=%m: %$ > log_path = /home/vmail/dovecot-deliver.log > } > protocol imap { > mail_max_userip_connections = 100 > } > > Regards, > > -- > Paulo Matos > From teemu.huovila at dovecot.fi Thu Aug 27 11:37:59 2015 From: teemu.huovila at dovecot.fi (Teemu Huovila) Date: Thu, 27 Aug 2015 14:37:59 +0300 Subject: question on autch cache parameters In-Reply-To: <55C331E1.8060702@securepoint.de> References: <55C21E22.10808@securepoint.de> <55C22CC4.5030006@securepoint.de> <55C331E1.8060702@securepoint.de> Message-ID: <55DEF697.30800@dovecot.fi> Hello Thank you for your report. We really appreciate it, especially when you can pinpoint a commit. However, I am unable to reproduce this. Could you post your doveconf -n please? Im especially interested in your passdb and userdb configurations and auth-cache settings. br, Teemu Huovila On 08/06/2015 01:07 PM, matthias lay wrote: > hi timo, > > I checked out the commit causing this. > > its this one: > > http://hg.dovecot.org/dovecot-2.2/diff/5e445c659f89/src/auth/auth-request.c#l1.32 > > > if I move this block back as it was. everything is fine > > > diff -r a46620d6e0ff -r 5e445c659f89 src/auth/auth-request.c > --- a/src/auth/auth-request.c Tue May 05 13:35:52 2015 +0300 > +++ b/src/auth/auth-request.c Tue May 05 14:16:31 2015 +0300 > @@ -618,30 +627,28 @@ > auth_request_want_skip_passdb(request, next_passdb)) > next_passdb = next_passdb->next; > > + if (*result == PASSDB_RESULT_OK) { > + /* this passdb lookup succeeded, preserve its extra fields */ > + auth_fields_snapshot(request->extra_fields); > + request->snapshot_have_userdb_prefetch_set = > + request->userdb_prefetch_set; > + if (request->userdb_reply != NULL) > + auth_fields_snapshot(request->userdb_reply); > + } else { > + /* this passdb lookup failed, remove any extra fields it set */ > + auth_fields_rollback(request->extra_fields); > + if (request->userdb_reply != NULL) { > + auth_fields_rollback(request->userdb_reply); > + request->userdb_prefetch_set = > + request->snapshot_have_userdb_prefetch_set; > + } > + } > + > if (passdb_continue && next_passdb != NULL) { > /* try next passdb. */ > request->passdb = next_passdb; > request->passdb_password = NULL; > > - if (*result == PASSDB_RESULT_OK) { > - /* this passdb lookup succeeded, preserve its extra > - fields */ > - auth_fields_snapshot(request->extra_fields); > - request->snapshot_have_userdb_prefetch_set = > - request->userdb_prefetch_set; > - if (request->userdb_reply != NULL) > - auth_fields_snapshot(request->userdb_reply); > - } else { > - /* this passdb lookup failed, remove any extra fields > - it set */ > - auth_fields_rollback(request->extra_fields); > - if (request->userdb_reply != NULL) { > - auth_fields_rollback(request->userdb_reply); > - request->userdb_prefetch_set = > - request->snapshot_have_userdb_prefetch_set; > - } > - } > - > if (*result == PASSDB_RESULT_USER_UNKNOWN) { > /* remember that we did at least one successful > passdb lookup */ > > > > > > > > > > > > > > > On 08/05/2015 05:33 PM, matthias lay wrote: >> just tested against dovecot 2.2.15 >> >> everythings works fine. so might be a bug introduced between 2.2.16 and >> 2.2.18 >> >> >> >> >> >> On 08/05/2015 04:30 PM, matthias lay wrote: >>> Hi list, >>> >>> I have a question on auth caching in 2.2.18. >>> >>> I am using acl_groups for a master user, appended in a static userdb file >>> >>> # snip ############################### >>> master at uma:{SHA}XXXX=::::::userdb_acl_groups=umareadmaster >>> allow_nets=127.0.0.1 >>> # snap ############################### >>> >>> and use this group in a global ACL file. >>> I discovered this only works on first NOT-cached login >>> >>> >>> >>> environment in imap-postlogin script on first login: >>> >>> >>> AUTH_TOKEN=e96b5a32ceb2cafc4460c210ad2e92e3d7ab388c >>> MASTER_USER=master at uma >>> SPUSER=private/pdf >>> LOCAL_IP=127.0.0.1 >>> USER=pdf >>> AUTH_USER=master at uma >>> PWD=/var/run/dovecot >>> USERDB_KEYS=ACL_GROUPS HOME SPUSER MASTER_USER AUTH_TOKEN AUTH_USER >>> SHLVL=1 >>> HOME=/var/data/vmail/private/pdf >>> ACL_GROUPS=umareadmaster >>> IP=127.0.0.1 >>> _=/usr/bin/env >>> >>> >>> on the second cached login it looks like this >>> >>> >>> AUTH_TOKEN=12703b11932f233520f6d4b33559c33aeb1cfc7f >>> MASTER_USER=master at uma >>> SPUSER=private/pdf >>> LOCAL_IP=127.0.0.1 >>> USER=pdf >>> AUTH_USER=master at uma >>> PWD=/var/run/dovecot >>> USERDB_KEYS=HOME SPUSER MASTER_USER AUTH_TOKEN AUTH_USER >>> SHLVL=1 >>> HOME=/var/data/vmail/private/pdf >>> IP=127.0.0.1 >>> _=/usr/bin/env >>> >>> so the ACL_GROUPS is gone. >>> >>> is this intended to be like that. >>> so groups not included in cache and I have to find another approach? >>> >>> anybody else encountered similar problems with some auth Variables and >>> caching? >>> >>> >>> Greetz Matze >>> >> From paulo at matos-sorge.com Thu Aug 27 11:46:19 2015 From: paulo at matos-sorge.com (Paulo Matos) Date: Thu, 27 Aug 2015 12:46:19 +0100 Subject: Mailbox can't be created In-Reply-To: References: Message-ID: On 27/08/15 10:42, Muzaffer Tolga Ozses wrote: > Please check permissions as this sounds like a permissions issue m I am confused then. /home/vmail/Maildir is 700 for vmail:vmail. If dovecot-lda runs as vmail then there's no problem, if it runs as root it shouldn't face problems either. How can I determine which user dovecot-lda runs under? Thanks, -- Paulo Matos From teemu.huovila at dovecot.fi Thu Aug 27 11:58:57 2015 From: teemu.huovila at dovecot.fi (Teemu Huovila) Date: Thu, 27 Aug 2015 14:58:57 +0300 Subject: IMAP hibernate feature committed In-Reply-To: <20150826103338.GA8337@nihlus.leuxner.net> References: <20150825080136.GA47256@nihlus.leuxner.net> <20150825175529.GA13298@nihlus.leuxner.net> <20150826103338.GA8337@nihlus.leuxner.net> Message-ID: <55DEFB81.4050805@dovecot.fi> On 08/26/2015 01:33 PM, Thomas Leuxner wrote: > * Timo Sirainen 2015.08.25 22:21: > >> There's no good default setting here. It depends on your userdb settings and/or mail_uid setting. So for example if your imap processes are running as vmail user, you should set service imap-hibernate { unix_listener imap-hibernate { user = vmail } }. Then again if you are using system users (or otherwise multiple UIDs) it gets more difficult to implement this securely (mode=0666 works always, but security isn't too good). This same problem exists for various other parts of Dovecot, for example indexer-worker and dict services. > > I have it working (I guess) with these user settings (virtual users using 'vmail'): > > service imap-hibernate { > unix_listener imap-hibernate { > user = vmail > } > } > > I had to assign the imap-master socket the user the imap-hibernate process is using to avoid messages like this: > > Aug 25 23:16:02 nihlus dovecot: imap-hibernate(tlx at leuxner.net): Error: net_connect_unix(/var/run/dovecot/imap-master) failed: Permission denied > Aug 25 23:16:02 nihlus dovecot: imap-hibernate(tlx at leuxner.net): Failed to connect to master socket in=126 out=944 hdr=0 body=0 del=0 exp=0 trash=0 > > service imap { > unix_listener imap-master { > user = dovecot > } > } > > With this I see messages like this in the logs: > > Aug 26 09:48:06 nihlus dovecot: imap-hibernate(tlx at leuxner.net): Connection closed in=189 out=4252 hdr=0 body=0 del=0 exp=0 trash=0 > Aug 26 12:20:29 nihlus dovecot: imap-hibernate(tlx at leuxner.net): Connection closed in=109 out=4714 hdr=0 body=0 del=0 exp=0 trash=0 > > I'm a bit puzzled as to when hibernate actually kicks in because most of the time I see normal imap processes running without them being hibernated: Did you specify a value other than zero for 'imap_hibernate_timeout'? br, Teemu > $ ps aux | grep dovecot/imap > dovenull 6791 0.0 0.0 18196 4772 ? S 06:39 0:00 dovecot/imap-login > dovenull 7107 0.0 0.0 18196 4736 ? S 08:00 0:00 dovecot/imap-login > dovenull 7112 0.0 0.0 18332 4492 ? S 08:00 0:00 dovecot/imap-login > dovenull 7333 0.0 0.0 18332 4772 ? S 08:45 0:00 dovecot/imap-login > dovenull 7675 0.0 0.0 18196 4628 ? S 10:13 0:00 dovecot/imap-login > dovenull 7677 0.0 0.0 18332 4532 ? S 10:14 0:00 dovecot/imap-login > dovenull 7821 0.0 0.0 18196 4532 ? S 10:44 0:00 dovecot/imap-login > dovenull 8156 0.0 0.0 18196 4756 ? S 12:01 0:00 dovecot/imap-login > vmail 8157 0.0 0.0 45624 9608 ? S 12:01 0:00 dovecot/imap > dovenull 8158 0.0 0.0 18332 4628 ? S 12:01 0:00 dovecot/imap-login > vmail 8159 0.0 0.0 44772 9256 ? S 12:01 0:00 dovecot/imap > dovenull 8160 0.0 0.0 18196 4652 ? S 12:01 0:00 dovecot/imap-login > vmail 8161 0.0 0.0 46072 9760 ? S 12:01 0:00 dovecot/imap > dovenull 8162 0.0 0.0 18196 4548 ? S 12:01 0:00 dovecot/imap-login > dovenull 8279 0.0 0.0 18332 4736 ? S 12:22 0:00 dovecot/imap-login > vmail 8280 0.0 0.0 40712 5164 ? S 12:22 0:00 dovecot/imap > dovenull 8341 0.0 0.0 18196 4740 ? S 12:25 0:00 dovecot/imap-login > vmail 8344 0.0 0.0 46312 10568 ? S 12:25 0:00 dovecot/imap > From tolga at ozses.net Thu Aug 27 12:00:08 2015 From: tolga at ozses.net (Muzaffer Tolga Ozses) Date: Thu, 27 Aug 2015 15:00:08 +0300 Subject: Mailbox can't be created In-Reply-To: References: Message-ID: ps aux | grep lda On 27/08/2015 2:58 PM, "Paulo Matos" wrote: > On 27/08/15 10:42, Muzaffer Tolga Ozses wrote: > >> Please check permissions as this sounds like a permissions issue m >> > > I am confused then. /home/vmail/Maildir is 700 for vmail:vmail. If > dovecot-lda runs as vmail then there's no problem, if it runs as root it > shouldn't face problems either. > > How can I determine which user dovecot-lda runs under? > > Thanks, > -- > Paulo Matos > From tolga at ozses.net Thu Aug 27 12:03:47 2015 From: tolga at ozses.net (Muzaffer Tolga Ozses) Date: Thu, 27 Aug 2015 15:03:47 +0300 Subject: Mailbox can't be created In-Reply-To: References: Message-ID: Having said that, /home/vmail should be owned by vmail, however dovecot-lda should have write access so 700 is a bad idea. On 27/08/2015 3:00 PM, "Muzaffer Tolga Ozses" wrote: > ps aux | grep lda > On 27/08/2015 2:58 PM, "Paulo Matos" wrote: > >> On 27/08/15 10:42, Muzaffer Tolga Ozses wrote: >> >>> Please check permissions as this sounds like a permissions issue m >>> >> >> I am confused then. /home/vmail/Maildir is 700 for vmail:vmail. If >> dovecot-lda runs as vmail then there's no problem, if it runs as root it >> shouldn't face problems either. >> >> How can I determine which user dovecot-lda runs under? >> >> Thanks, >> -- >> Paulo Matos >> > From paulo at matos-sorge.com Thu Aug 27 11:49:04 2015 From: paulo at matos-sorge.com (Paulo Matos) Date: Thu, 27 Aug 2015 12:49:04 +0100 Subject: Mailbox can't be created In-Reply-To: References: Message-ID: On 27/08/15 12:46, Paulo Matos wrote: > On 27/08/15 10:42, Muzaffer Tolga Ozses wrote: >> Please check permissions as this sounds like a permissions issue m > > I am confused then. /home/vmail/Maildir is 700 for vmail:vmail. If > dovecot-lda runs as vmail then there's no problem, if it runs as root it > shouldn't face problems either. > > How can I determine which user dovecot-lda runs under? > Doh, it says on the line I quoted earlier from master.cf: dovecot unix - n n - - pipe flags=DRhu user=vmail:vmail argv=/usr/libexec/dovecot/deliver -f ${sender} -a ${recipient} -d ${user}@${nexthop} -m PreINBOX It runs under vmail:vmail. There should be no problem creating a new mailbox. Is there a way to understand what's really going on with dovecot? > Thanks, From tolga at ozses.net Thu Aug 27 12:07:15 2015 From: tolga at ozses.net (Muzaffer Tolga Ozses) Date: Thu, 27 Aug 2015 15:07:15 +0300 Subject: Mailbox can't be created In-Reply-To: References: Message-ID: Make it 755 and be done with it. On 27/08/2015 3:05 PM, "Paulo Matos" wrote: > On 27/08/15 12:46, Paulo Matos wrote: > >> On 27/08/15 10:42, Muzaffer Tolga Ozses wrote: >> >>> Please check permissions as this sounds like a permissions issue m >>> >> >> I am confused then. /home/vmail/Maildir is 700 for vmail:vmail. If >> dovecot-lda runs as vmail then there's no problem, if it runs as root it >> shouldn't face problems either. >> >> How can I determine which user dovecot-lda runs under? >> >> > Doh, it says on the line I quoted earlier from master.cf: > dovecot unix - n n - - pipe > flags=DRhu user=vmail:vmail argv=/usr/libexec/dovecot/deliver -f > ${sender} -a ${recipient} -d ${user}@${nexthop} -m PreINBOX > > It runs under vmail:vmail. There should be no problem creating a new > mailbox. > > Is there a way to understand what's really going on with dovecot? > > Thanks, >> > From a.helwig at heinlein-support.de Thu Aug 27 12:09:28 2015 From: a.helwig at heinlein-support.de (Andre Helwig) Date: Thu, 27 Aug 2015 14:09:28 +0200 Subject: mailbox deletion is not replicated correctly In-Reply-To: <55A4C9B8.7080307@heinlein-support.de> References: <552FE1C1.4050804@heinlein-support.de> <4FB37D9A-B0DE-46C8-AC86-461CC5AC24E1@iki.fi> <5530DB62.9010502@heinlein-support.de> <55A4C9B8.7080307@heinlein-support.de> Message-ID: <55DEFDF8.10203@heinlein-support.de> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 07/14/2015 10:35 AM, Andre Helwig wrote: > On 04/17/2015 12:07 PM, Dennis Kuhn wrote: >> On 16.04.2015 20:32, Timo Sirainen wrote: >>> On 16 Apr 2015, at 19:22, Dennis Kuhn >>> wrote: >>>> i can't delete several folders at once. I have a folder >>>> structure like folder1/folder2/folder3 (we use slash as >>>> hierarchy separator). When i delete folder1 server1 executes >>>> the command correctly >>>> >>>> >>>> Apr 16 13:44:28 server1 dovecot: >>>> imap(d.kuhn at heinlein-support.de): Mailbox deleted: >>>> INBOX/Trash/folder1/folder2/folder3 Apr 16 13:44:28 server1 >>>> dovecot: imap(d.kuhn at heinlein-support.de): Mailbox deleted: >>>> INBOX/Trash/folder1/folder2 Apr 16 13:44:28 server1 dovecot: >>>> imap(d.kuhn at heinlein-support.de): Mailbox deleted: >>>> INBOX/Trash/folder1 >>>> >>>> >>>> but the replica server2 reverses the order: >>>> >>>> Apr 16 13:44:29 server2 dovecot: >>>> dsync-server(d.kuhn at heinlein-support.de): Mailbox deleted: >>>> INBOX/Trash/folder1 Apr 16 13:44:29 server2 dovecot: >>>> dsync-server(d.kuhn at heinlein-support.de): Mailbox deleted: >>>> INBOX/Trash/folder1/folder2 Apr 16 13:44:29 server2 dovecot: >>>> dsync-server(d.kuhn at heinlein-support.de): Mailbox deleted: >>>> INBOX/Trash/folder1/folder2/folder3 >>>> >>>> >>>> This creates the folder INBOX/Trash/folder1/folder2 on >>>> server2 again which replicates back to server1. >>> With what Dovecot version? What mailbox format? >>> >> The mailbox format is mdbox. Server1 is 2.2.15, server2 2.2.16. > Hi Timo, is there any update on this issue? > > cheers Andre > Just want to bring it to your attention again. - -- Heinlein Support GmbH Schwedter Str. 8/9b, 10119 Berlin http://www.heinlein-support.de Tel: 030 / 405051-0 Fax: 030 / 405051-19 Zwangsangaben lt. ?35a GmbHG: HRB 93818 B / Amtsgericht Berlin-Charlottenburg, Gesch?ftsf?hrer: Peer Heinlein -- Sitz: Berlin -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAEBAgAGBQJV3v34AAoJEAoTNwRDnEhRx+0H/01YT39mvkgwgDGAUQzVtaKJ 0lZZJzkMwua6qOBRWGsiK8jH5PYvOWdRjbRxVE5y1H8n8/CLNCJ3nkpvuDu/vMf3 8A4i/DqHy+Dvof+3AagECGLnqdvCi9zcqTgE3Vu9dJyEUzK7A1y1sxLB8CN6GBo2 LBXsrHUWi+eBLDkQdR94LaA6Ssa8TMnSr6ZcpbRmx+bR54uzJ6jwUcmO6db2osqC ESl42Vov0Yps0tp0h7sz+qRmSnIDpAatJys/qjKvZmr+WawC4g5vGcT7Bv33WevC JHt10YOOcftIGZE2zZGo9UyCOHjmnWiluk3toIOdBr4FAVNZhrQj8pplol8m+jU= =ZiRI -----END PGP SIGNATURE----- From paulo at matos-sorge.com Thu Aug 27 11:54:02 2015 From: paulo at matos-sorge.com (Paulo Matos) Date: Thu, 27 Aug 2015 12:54:02 +0100 Subject: Mailbox can't be created In-Reply-To: References: Message-ID: On 27/08/15 13:00, Muzaffer Tolga Ozses wrote: > ps aux | grep lda Strange... postfix 1315 0.0 0.1 121220 7472 ? S 12:59 0:00 pipe -n dovecot -t unix flags=DRhu user=vmail vmail argv=/usr/libexec/dovecot/deliver -f ${sender} -a ${recipient} -d ${user}@${nexthop} -m PreINBOX I notice that the arg to pipe is 'user=vmail vmail' instead of 'user=vmail:vmail'. This is strange. I certanly have a colon on the master.cf file. According to 'man pipe' the correct syntax is user=: so I can only assume this is where it's going wrong somehow. > On 27/08/2015 2:58 PM, "Paulo Matos" wrote: > >> On 27/08/15 10:42, Muzaffer Tolga Ozses wrote: >> >>> Please check permissions as this sounds like a permissions issue m >>> >> >> I am confused then. /home/vmail/Maildir is 700 for vmail:vmail. If >> dovecot-lda runs as vmail then there's no problem, if it runs as root it >> shouldn't face problems either. >> >> How can I determine which user dovecot-lda runs under? >> >> Thanks, >> -- >> Paulo Matos >> > From paulo at matos-sorge.com Thu Aug 27 12:14:41 2015 From: paulo at matos-sorge.com (Paulo Matos) Date: Thu, 27 Aug 2015 13:14:41 +0100 Subject: Mailbox can't be created In-Reply-To: References: Message-ID: <68506054d1644416956542747d938919@matos-sorge.com> dovecot-lda is being executed as vmail so it shouldn't be a problem in theory. On 27/08/2015 13:03, Muzaffer Tolga Ozses wrote: > Having said that, /home/vmail should be owned by vmail, however dovecot-lda should have write access so 700 is a bad idea. > On 27/08/2015 3:00 PM, "Muzaffer Tolga Ozses" wrote: > > ps aux | grep lda > On 27/08/2015 2:58 PM, "Paulo Matos" wrote: > On 27/08/15 10:42, Muzaffer Tolga Ozses wrote: > Please check permissions as this sounds like a permissions issue m > I am confused then. /home/vmail/Maildir is 700 for vmail:vmail. If dovecot-lda runs as vmail then there's no problem, if it runs as root it shouldn't face problems either. > > How can I determine which user dovecot-lda runs under? > > Thanks, > -- > Paulo Matos -- Paulo Matos From matthias.lay at securepoint.de Thu Aug 27 12:30:11 2015 From: matthias.lay at securepoint.de (matthias lay) Date: Thu, 27 Aug 2015 14:30:11 +0200 Subject: question on autch cache parameters In-Reply-To: <55DEF697.30800@dovecot.fi> References: <55C21E22.10808@securepoint.de> <55C22CC4.5030006@securepoint.de> <55C331E1.8060702@securepoint.de> <55DEF697.30800@dovecot.fi> Message-ID: <20150827143011.7b76de3c@eugen.spdev.local> hi teemu, thx for your reply. the user is a masteruser that hast a static passwd file. this is where the ACL_GROUPS is applied ############ cat /etc/dovecot/passwd.masteruser master at uma:{SHA}ojN+jsbELZbRJeRb0qj9+MMjPUs=::::::userdb_acl_groups=umareadmaster allow_nets=127.0.0.1 ########## * the standard lookup method for users is ldap. only masterusers are in static user/passdbs * auth cache is enabled I cant post my whole conf but will paste the parts you requested. if its not enough for you to reproduce, I will setup a clean instance and reproduce it there. ###################################################################################################################### # 2.2.16: /etc/dovecot/dovecot.conf # Pigeonhole version 0.4.7 # OS: Linux 3.15.10-dist i686 auth_cache_negative_ttl = 30 mins auth_cache_size = 10 k auth_master_user_separator = * #### snip default namespace: (maildir gets overwritten by ldap lookup on most users) namespace { hidden = no inbox = no list = children location = maildir:/var/data/vmail/public/%%Lu/Maildir:LAYOUT=fs:INBOX=/var/data/vmail/public/%%Lu/Maildir/INBOX prefix = public/%%u/ separator = / subscriptions = no type = shared } userdb { args = uid=vmail gid=vmail home=/var/data/vmail/public/%Ln driver = static } .... protocol imap { imap_client_workarounds = tb-extra-mailbox-sep mail_plugins = acl notify mailbox_alias imap_acl ssl = yes ssl_cert = wrote: > Hello > > Thank you for your report. We really appreciate it, especially when > you can pinpoint a commit. > > However, I am unable to reproduce this. Could you post your doveconf > -n please? Im especially interested in your passdb and userdb > configurations and auth-cache settings. > > br, > Teemu Huovila > > > On 08/06/2015 01:07 PM, matthias lay wrote: > > hi timo, > > > > I checked out the commit causing this. > > > > its this one: > > > > http://hg.dovecot.org/dovecot-2.2/diff/5e445c659f89/src/auth/auth-request.c#l1.32 > > > > > > if I move this block back as it was. everything is fine > > > > > > diff -r a46620d6e0ff -r 5e445c659f89 src/auth/auth-request.c > > --- a/src/auth/auth-request.c Tue May 05 13:35:52 2015 +0300 > > +++ b/src/auth/auth-request.c Tue May 05 14:16:31 2015 +0300 > > @@ -618,30 +627,28 @@ > > auth_request_want_skip_passdb(request, next_passdb)) > > next_passdb = next_passdb->next; > > > > + if (*result == PASSDB_RESULT_OK) { > > + /* this passdb lookup succeeded, preserve its > > extra fields */ > > + auth_fields_snapshot(request->extra_fields); > > + request->snapshot_have_userdb_prefetch_set = > > + request->userdb_prefetch_set; > > + if (request->userdb_reply != NULL) > > + > > auth_fields_snapshot(request->userdb_reply); > > + } else { > > + /* this passdb lookup failed, remove any extra > > fields it set */ > > + auth_fields_rollback(request->extra_fields); > > + if (request->userdb_reply != NULL) { > > + > > auth_fields_rollback(request->userdb_reply); > > + request->userdb_prefetch_set = > > + > > request->snapshot_have_userdb_prefetch_set; > > + } > > + } > > + > > if (passdb_continue && next_passdb != NULL) { > > /* try next passdb. */ > > request->passdb = next_passdb; > > request->passdb_password = NULL; > > > > - if (*result == PASSDB_RESULT_OK) { > > - /* this passdb lookup succeeded, preserve > > its extra > > - fields */ > > - > > auth_fields_snapshot(request->extra_fields); > > - request->snapshot_have_userdb_prefetch_set > > = > > - request->userdb_prefetch_set; > > - if (request->userdb_reply != NULL) > > - > > auth_fields_snapshot(request->userdb_reply); > > - } else { > > - /* this passdb lookup failed, remove any > > extra fields > > - it set */ > > - > > auth_fields_rollback(request->extra_fields); > > - if (request->userdb_reply != NULL) { > > - > > auth_fields_rollback(request->userdb_reply); > > - request->userdb_prefetch_set = > > - > > request->snapshot_have_userdb_prefetch_set; > > - } > > - } > > - > > if (*result == PASSDB_RESULT_USER_UNKNOWN) { > > /* remember that we did at least one > > successful passdb lookup */ > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > On 08/05/2015 05:33 PM, matthias lay wrote: > >> just tested against dovecot 2.2.15 > >> > >> everythings works fine. so might be a bug introduced between > >> 2.2.16 and 2.2.18 > >> > >> > >> > >> > >> > >> On 08/05/2015 04:30 PM, matthias lay wrote: > >>> Hi list, > >>> > >>> I have a question on auth caching in 2.2.18. > >>> > >>> I am using acl_groups for a master user, appended in a static > >>> userdb file > >>> > >>> # snip ############################### > >>> master at uma:{SHA}XXXX=::::::userdb_acl_groups=umareadmaster > >>> allow_nets=127.0.0.1 > >>> # snap ############################### > >>> > >>> and use this group in a global ACL file. > >>> I discovered this only works on first NOT-cached login > >>> > >>> > >>> > >>> environment in imap-postlogin script on first login: > >>> > >>> > >>> AUTH_TOKEN=e96b5a32ceb2cafc4460c210ad2e92e3d7ab388c > >>> MASTER_USER=master at uma > >>> SPUSER=private/pdf > >>> LOCAL_IP=127.0.0.1 > >>> USER=pdf > >>> AUTH_USER=master at uma > >>> PWD=/var/run/dovecot > >>> USERDB_KEYS=ACL_GROUPS HOME SPUSER MASTER_USER AUTH_TOKEN > >>> AUTH_USER SHLVL=1 > >>> HOME=/var/data/vmail/private/pdf > >>> ACL_GROUPS=umareadmaster > >>> IP=127.0.0.1 > >>> _=/usr/bin/env > >>> > >>> > >>> on the second cached login it looks like this > >>> > >>> > >>> AUTH_TOKEN=12703b11932f233520f6d4b33559c33aeb1cfc7f > >>> MASTER_USER=master at uma > >>> SPUSER=private/pdf > >>> LOCAL_IP=127.0.0.1 > >>> USER=pdf > >>> AUTH_USER=master at uma > >>> PWD=/var/run/dovecot > >>> USERDB_KEYS=HOME SPUSER MASTER_USER AUTH_TOKEN AUTH_USER > >>> SHLVL=1 > >>> HOME=/var/data/vmail/private/pdf > >>> IP=127.0.0.1 > >>> _=/usr/bin/env > >>> > >>> so the ACL_GROUPS is gone. > >>> > >>> is this intended to be like that. > >>> so groups not included in cache and I have to find another > >>> approach? > >>> > >>> anybody else encountered similar problems with some auth > >>> Variables and caching? > >>> > >>> > >>> Greetz Matze > >>> > >> From eugene at raptor.kiev.ua Thu Aug 27 13:08:49 2015 From: eugene at raptor.kiev.ua (Eugene Paskevich) Date: Thu, 27 Aug 2015 16:08:49 +0300 Subject: Panic in service(log) Message-ID: <55DF0BE1.8010808@raptor.kiev.ua> Hello, I'm using the Debian builds of dovecot-2.2 by S[r]us. Today I started getting the following panic: log: Panic: file priorityq.c: line 122 (priorityq_remove_idx): assertion failed: (idx < count) log: Error: Raw backtrace: /usr/lib/dovecot/libdovecot.so.0(+0x7f1ae) [0x7fd368dc31ae] -> /usr/lib/dovecot/libdovecot.so.0(+0x7f269) [0x7fd368dc3269] -> /usr/lib/dovecot/libdovecot.so.0(i_fatal+0) [0x7fd368d6c81e] -> /usr/lib/dovecot/libdovecot.so.0(+0xa0de8) [0x7fd368de4de8] -> /usr/lib/dovecot/libdovecot.so.0(priorityq_remove+0xb) [0x7fd368de4eeb] -> /usr/lib/dovecot/libdovecot.so.0(+0x9179c) [0x7fd368dd579c] -> /usr/lib/dovecot/libdovecot.so.0(master_status_update+0x93) [0x7fd368d71263] -> /usr/lib/dovecot/libdovecot.so.0(+0x2ecf3) [0x7fd368d72cf3] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_call_io+0x4c) [0x7fd368dd6b2c] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_handler_run_internal+0xc5) [0x7fd368dd7e35] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_handler_run+0x25) [0x7fd368dd6bb5] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_run+0x38) [0x7fd368dd6d58] -> /usr/lib/dovecot/libdovecot.so.0(master_service_run+0x13) [0x7fd368d72213] -> dovecot/log(main+0x11c) [0x555e9f81771c] -> /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf5) [0x7fd3689bcb45] -> dovecot/log(+0x277d) [0x555e9f81777d] master: Error: service(log): child 28845 killed with signal 6 (core dumps disabled) Could anyone please advise? -- Eugene Paskevich | *==)----------- | Plug me into eugene at raptor.kiev.ua | -----------(==* | The Matrix From paulo at matos-sorge.com Thu Aug 27 13:03:35 2015 From: paulo at matos-sorge.com (Paulo Matos) Date: Thu, 27 Aug 2015 14:03:35 +0100 Subject: Mailbox can't be created In-Reply-To: References: Message-ID: On 27/08/15 13:07, Muzaffer Tolga Ozses wrote: > Make it 755 and be done with it. Unfortunately that doesn't work either. I still get that dovecot can't create mailbox. :-/ From pgspm at yahoo.no Thu Aug 27 14:26:48 2015 From: pgspm at yahoo.no (Petter Gunnerud) Date: Thu, 27 Aug 2015 14:26:48 +0000 (UTC) Subject: Merging maildirs Message-ID: <2004270624.1949627.1440685608316.JavaMail.yahoo@mail.yahoo.com> Hi List How do I merge files from one maildir into another maildir? Long version: I've been using dovecot/postfix for a 12 user environment for several years, and today was the first time I ran into an issue I could not solve from reading the documentation. The server is gentoo running on esxi. A few times a year I've made a snapshot of the server and copied the files to my home esxi server as a full system backup. 5 days ago the internet connection at the office broke, and I decided to power on the last copy of the server at my home and change public dns so that users could access their new mails via webmail from this spare server. (I didn't allow imap through firewall.) Now the network at the office is back. DNS is updated so that mails are delivered to the office server again. But all the mails sent / received during those 5 days is only at my house. How can I merge those into the office server?I'd like to do this as a server administrator. I do not want to have each user forward the mails to them self from the temporarily webmail. I tried to just copy the mail files on my own account - like I did once when I used courier-imap, but that did not work with dovecot. I hope someone has an idea of how to solve this. From troeder at univention.de Thu Aug 27 14:56:04 2015 From: troeder at univention.de (=?UTF-8?B?RGFuaWVsIFRyw7ZkZXI=?=) Date: Thu, 27 Aug 2015 16:56:04 +0200 Subject: Merging maildirs In-Reply-To: <2004270624.1949627.1440685608316.JavaMail.yahoo@mail.yahoo.com> References: <2004270624.1949627.1440685608316.JavaMail.yahoo@mail.yahoo.com> Message-ID: <55DF2504.6090907@univention.de> On 08/27/2015 16:26, Petter Gunnerud wrote: > Hi List How do I merge files from one maildir into another maildir? > > Long version: I've been using dovecot/postfix for a 12 user > environment for several years, and today was the first time I ran > into an issue I could not solve from reading the documentation. > > The server is gentoo running on esxi. A few times a year I've made a > snapshot of the server and copied the files to my home esxi server as > a full system backup. > > 5 days ago the internet connection at the office broke, and I decided > to power on the last copy of the server at my home and change public > dns so that users could access their new mails via webmail from this > spare server. (I didn't allow imap through firewall.) Now the network > at the office is back. DNS is updated so that mails are delivered to > the office server again. But all the mails sent / received during > those 5 days is only at my house. How can I merge those into the > office server?I'd like to do this as a server administrator. I do not > want to have each user forward the mails to them self from the > temporarily webmail. I tried to just copy the mail files on my own > account - like I did once when I used courier-imap, but that did not > work with dovecot. I hope someone has an idea of how to solve this. Just using rsync should work - I've done it before. You'll have to check the file system permissions. The cleaner way would be to use imapsync or Dovecots dsync. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From miloslav.hula at gmail.com Thu Aug 27 16:09:13 2015 From: miloslav.hula at gmail.com (=?UTF-8?Q?Miloslav_H=c5=afla?=) Date: Thu, 27 Aug 2015 18:09:13 +0200 Subject: Allow delivery to existing accounts only with LDAP and static Message-ID: <55DF3629.7090005@gmail.com> Hi, we are using LDAP binding as a passdb, and static with allow_all_users=yes as an userdb. Works fine, but problem is, Maildirs are created for non-existent accounts too. We would like to prevent it. The LDAP binding does not supporta user lookups. Is the correct way to use checkpassword as a passdb before LDAP, check for account existency here and: result_success=continue result_failure=return-fail ? Thank you, regards, Miloslav From eugene at raptor.kiev.ua Thu Aug 27 16:17:36 2015 From: eugene at raptor.kiev.ua (Eugene Paskevich) Date: Thu, 27 Aug 2015 19:17:36 +0300 Subject: Panic in service(log) In-Reply-To: <55DF0BE1.8010808@raptor.kiev.ua> References: <55DF0BE1.8010808@raptor.kiev.ua> Message-ID: <182351440692256@web21m.yandex.ru> 27.08.2015, 16:09, "Eugene Paskevich" : > Hello, > > I'm using the Debian builds of dovecot-2.2 by S[r]us. > Today I started getting the following panic: > > log: Panic: file priorityq.c: line 122 (priorityq_remove_idx): assertion > failed: (idx < count) Just in case, here is the full backtrace. (gdb) thread apply all bt full Thread 1 (LWP 28412): #0 0x00007fe01d5df107 in raise () from /lib/x86_64-linux-gnu/libc.so.6 No symbol table info available. #1 0x00007fe01d5e04e8 in abort () from /lib/x86_64-linux-gnu/libc.so.6 No symbol table info available. #2 0x00007fe01d9d21a6 in default_fatal_finish (type=, status=status at entry=0) at failures.c:202 backtrace = 0x559b3daf8198 "/usr/lib/dovecot/libdovecot.so.0(+0x7f1ae) [0x7fe01d9d21ae] -> /usr/lib/dovecot/libdovecot.so.0(+0x7f29c) [0x7fe01d9d229c] -> /usr/lib/dovecot/libdovecot.so.0(i_fatal+0) [0x7fe01d97b81e] -> /usr/lib/d"... #3 0x00007fe01d9d229c in i_internal_fatal_handler (ctx=0x7fffdc318f70, format=, args=) at failures.c:671 status = 0 #4 0x00007fe01d97b81e in i_panic (format=format at entry=0x7fe01d9ff898 "file %s: line %d (%s): assertion failed: (%s)") at failures.c:276 ctx = {type = LOG_TYPE_PANIC, exit_status = 0, timestamp = 0x0, timestamp_usecs = 0} args = {{gp_offset = 40, fp_offset = 48, overflow_arg_area = 0x7fffdc319070, reg_save_area = 0x7fffdc318fb0}} #5 0x00007fe01d9f3de8 in priorityq_remove_idx (pq=0x559b3db007e0, idx=4294967295) at priorityq.c:122 pq = 0x559b3db007e0 idx = 4294967295 count = 1 #6 0x00007fe01d9f3eeb in priorityq_remove (pq=, item=item at entry=0x559b3db221f0) at priorityq.c:138 No locals. #7 0x00007fe01d9e479c in timeout_reset_timeval (timeout=0x559b3db221f0, tv_now=tv_now at entry=0x0) at ioloop.c:329 __FUNCTION__ = "timeout_reset_timeval" #8 0x00007fe01d9e4b4d in timeout_reset (timeout=) at ioloop.c:336 __FUNCTION__ = "timeout_reset" #9 0x00007fe01dd062a6 in index_mailbox_want_full_sync (box=box at entry=0x559b3db1c410, flags=flags at entry=(unknown: 0)) at index-sync.c:49 ibox = 0x559b3db1c888 #10 0x00007fe01dcbbb60 in maildir_storage_sync_init (box=0x559b3db1c410, flags=(unknown: 0)) at maildir-sync.c:1101 mbox = 0x559b3db1c410 lost_files = false force_resync = false ret = 0 __FUNCTION__ = "maildir_storage_sync_init" #11 0x00007fe01dc90bbb in mailbox_sync_init (box=box at entry=0x559b3db1c410, flags=flags at entry=(unknown: 0)) at mail-storage.c:1651 _data_stack_cur_id = 4 ctx = #12 0x0000559b3d622a01 in imap_sync_init (client=0x559b3db19010, box=, imap_flags=imap_flags at entry=(unknown: 0), flags=flags at entry=(unknown: 0)) at imap-sync.c:230 __FUNCTION__ = "imap_sync_init" #13 0x0000559b3d60f05e in idle_sync_now (box=, ctx=0x559b3db19d08) at cmd-idle.c:145 No locals. #14 0x0000559b3d60f35d in cmd_idle (cmd=0x559b3db19bf0) at cmd-idle.c:308 client = 0x559b3db19010 #15 0x0000559b3d61875c in command_exec (cmd=cmd at entry=0x559b3db19bf0) at imap-commands.c:168 hook = 0x559b3db01120 diff = finished = #16 0x0000559b3d616cf2 in client_command_input (cmd=cmd at entry=0x559b3db19bf0) at imap-client.c:821 client = 0x559b3db19010 command = __FUNCTION__ = "client_command_input" #17 0x0000559b3d616d80 in client_command_input (cmd=0x559b3db19bf0) at imap-client.c:881 client = 0x559b3db19010 command = __FUNCTION__ = "client_command_input" #18 0x0000559b3d6170ed in client_handle_next_command (remove_io_r=, client=0x559b3db19010) at imap-client.c:919 No locals. #19 client_handle_input (client=0x559b3db19010) at imap-client.c:931 _data_stack_cur_id = 3 remove_io = false handled_commands = false client = 0x559b3db19010 #20 0x0000559b3d6174b5 in client_input (client=0x559b3db19010) at imap-client.c:973 cmd = 0xa51de output = 0x559b3db19a30 bytes = 8 __FUNCTION__ = "client_input" #21 0x00007fe01d9e5b2c in io_loop_call_io (io=0x559b3db19b00) at ioloop.c:540 ioloop = 0x559b3db00770 t_id = 2 __FUNCTION__ = "io_loop_call_io" #22 0x00007fe01d9e6e35 in io_loop_handler_run_internal (ioloop=ioloop at entry=0x559b3db00770) at ioloop-epoll.c:220 ctx = 0x559b3db01510 io = tv = {tv_sec = 1799, tv_usec = 999712} events_count = msecs = ret = 1 i = 0 j = call = __FUNCTION__ = "io_loop_handler_run_internal" #23 0x00007fe01d9e5bb5 in io_loop_handler_run (ioloop=ioloop at entry=0x559b3db00770) at ioloop.c:588 No locals. #24 0x00007fe01d9e5d58 in io_loop_run (ioloop=0x559b3db00770) at ioloop.c:564 __FUNCTION__ = "io_loop_run" #25 0x00007fe01d981213 in master_service_run (service=0x559b3db00610, callback=) at master-service.c:642 No locals. #26 0x0000559b3d60a8f7 in main (argc=1, argv=0x559b3db00390) at main.c:443 set_roots = {0x559b3d833740 , 0x0} login_set = {auth_socket_path = 0x559b3daf8048 "\001", postlogin_socket_path = 0x0, postlogin_timeout_secs = 60, callback = 0x559b3d624150 , failure_callback = 0x559b3d623ac0 , request_auth_token = 1} service_flags = storage_service_flags = username = 0x0 c = (gdb) -- Eugene Paskevich | *==)----------- | Plug me into eugene at raptor.kiev.ua | -----------(==* | The Matrix From tlx at leuxner.net Thu Aug 27 16:39:14 2015 From: tlx at leuxner.net (Thomas Leuxner) Date: Thu, 27 Aug 2015 18:39:14 +0200 Subject: IMAP hibernate feature committed In-Reply-To: <55DEFB81.4050805@dovecot.fi> References: <20150825080136.GA47256@nihlus.leuxner.net> <20150825175529.GA13298@nihlus.leuxner.net> <20150826103338.GA8337@nihlus.leuxner.net> <55DEFB81.4050805@dovecot.fi> Message-ID: <20150827163914.GA9085@nihlus.leuxner.net> * Teemu Huovila 2015.08.27 13:58: > Did you specify a value other than zero for 'imap_hibernate_timeout'? Yes I did: $ doveconf imap_hibernate_timeout imap_hibernate_timeout = 1 mins I sometimes see one imap-hibernate process (only one), but several imap processes active which should be idling... -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: Digital signature URL: From stephan at rename-it.nl Thu Aug 27 17:09:50 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Thu, 27 Aug 2015 19:09:50 +0200 Subject: Panic in service(log) In-Reply-To: <182351440692256@web21m.yandex.ru> References: <55DF0BE1.8010808@raptor.kiev.ua> <182351440692256@web21m.yandex.ru> Message-ID: <55DF445E.7010808@rename-it.nl> Op 8/27/2015 om 6:17 PM schreef Eugene Paskevich: > 27.08.2015, 16:09, "Eugene Paskevich" : >> Hello, >> >> I'm using the Debian builds of dovecot-2.2 by S[r]us. >> Today I started getting the following panic: >> >> log: Panic: file priorityq.c: line 122 (priorityq_remove_idx): assertion >> failed: (idx < count) > Just in case, here is the full backtrace. OK, that confirms my suspicion. Should be fixed now. Thanks. Regards, Stephan. > (gdb) thread apply all bt full > > Thread 1 (LWP 28412): > #0 0x00007fe01d5df107 in raise () from /lib/x86_64-linux-gnu/libc.so.6 > No symbol table info available. > #1 0x00007fe01d5e04e8 in abort () from /lib/x86_64-linux-gnu/libc.so.6 > No symbol table info available. > #2 0x00007fe01d9d21a6 in default_fatal_finish (type=, status=status at entry=0) at failures.c:202 > backtrace = 0x559b3daf8198 "/usr/lib/dovecot/libdovecot.so.0(+0x7f1ae) [0x7fe01d9d21ae] -> /usr/lib/dovecot/libdovecot.so.0(+0x7f29c) [0x7fe01d9d229c] -> /usr/lib/dovecot/libdovecot.so.0(i_fatal+0) [0x7fe01d97b81e] -> /usr/lib/d"... > #3 0x00007fe01d9d229c in i_internal_fatal_handler (ctx=0x7fffdc318f70, format=, args=) at failures.c:671 > status = 0 > #4 0x00007fe01d97b81e in i_panic (format=format at entry=0x7fe01d9ff898 "file %s: line %d (%s): assertion failed: (%s)") at failures.c:276 > ctx = {type = LOG_TYPE_PANIC, exit_status = 0, timestamp = 0x0, timestamp_usecs = 0} > args = {{gp_offset = 40, fp_offset = 48, overflow_arg_area = 0x7fffdc319070, reg_save_area = 0x7fffdc318fb0}} > #5 0x00007fe01d9f3de8 in priorityq_remove_idx (pq=0x559b3db007e0, idx=4294967295) at priorityq.c:122 > pq = 0x559b3db007e0 > idx = 4294967295 > count = 1 > #6 0x00007fe01d9f3eeb in priorityq_remove (pq=, item=item at entry=0x559b3db221f0) at priorityq.c:138 > No locals. > #7 0x00007fe01d9e479c in timeout_reset_timeval (timeout=0x559b3db221f0, tv_now=tv_now at entry=0x0) at ioloop.c:329 > __FUNCTION__ = "timeout_reset_timeval" > #8 0x00007fe01d9e4b4d in timeout_reset (timeout=) at ioloop.c:336 > __FUNCTION__ = "timeout_reset" > #9 0x00007fe01dd062a6 in index_mailbox_want_full_sync (box=box at entry=0x559b3db1c410, flags=flags at entry=(unknown: 0)) at index-sync.c:49 > ibox = 0x559b3db1c888 > #10 0x00007fe01dcbbb60 in maildir_storage_sync_init (box=0x559b3db1c410, flags=(unknown: 0)) at maildir-sync.c:1101 > mbox = 0x559b3db1c410 > lost_files = false > force_resync = false > ret = 0 > __FUNCTION__ = "maildir_storage_sync_init" > #11 0x00007fe01dc90bbb in mailbox_sync_init (box=box at entry=0x559b3db1c410, flags=flags at entry=(unknown: 0)) at mail-storage.c:1651 > _data_stack_cur_id = 4 > ctx = > #12 0x0000559b3d622a01 in imap_sync_init (client=0x559b3db19010, box=, imap_flags=imap_flags at entry=(unknown: 0), flags=flags at entry=(unknown: 0)) at imap-sync.c:230 > __FUNCTION__ = "imap_sync_init" > #13 0x0000559b3d60f05e in idle_sync_now (box=, ctx=0x559b3db19d08) at cmd-idle.c:145 > No locals. > #14 0x0000559b3d60f35d in cmd_idle (cmd=0x559b3db19bf0) at cmd-idle.c:308 > client = 0x559b3db19010 > #15 0x0000559b3d61875c in command_exec (cmd=cmd at entry=0x559b3db19bf0) at imap-commands.c:168 > hook = 0x559b3db01120 > diff = > finished = > #16 0x0000559b3d616cf2 in client_command_input (cmd=cmd at entry=0x559b3db19bf0) at imap-client.c:821 > client = 0x559b3db19010 > command = > __FUNCTION__ = "client_command_input" > #17 0x0000559b3d616d80 in client_command_input (cmd=0x559b3db19bf0) at imap-client.c:881 > client = 0x559b3db19010 > command = > __FUNCTION__ = "client_command_input" > #18 0x0000559b3d6170ed in client_handle_next_command (remove_io_r=, client=0x559b3db19010) at imap-client.c:919 > No locals. > #19 client_handle_input (client=0x559b3db19010) at imap-client.c:931 > _data_stack_cur_id = 3 > remove_io = false > handled_commands = false > client = 0x559b3db19010 > #20 0x0000559b3d6174b5 in client_input (client=0x559b3db19010) at imap-client.c:973 > cmd = 0xa51de > output = 0x559b3db19a30 > bytes = 8 > __FUNCTION__ = "client_input" > #21 0x00007fe01d9e5b2c in io_loop_call_io (io=0x559b3db19b00) at ioloop.c:540 > ioloop = 0x559b3db00770 > t_id = 2 > __FUNCTION__ = "io_loop_call_io" > #22 0x00007fe01d9e6e35 in io_loop_handler_run_internal (ioloop=ioloop at entry=0x559b3db00770) at ioloop-epoll.c:220 > ctx = 0x559b3db01510 > io = > tv = {tv_sec = 1799, tv_usec = 999712} > events_count = > msecs = > ret = 1 > i = 0 > j = > call = > __FUNCTION__ = "io_loop_handler_run_internal" > #23 0x00007fe01d9e5bb5 in io_loop_handler_run (ioloop=ioloop at entry=0x559b3db00770) at ioloop.c:588 > No locals. > #24 0x00007fe01d9e5d58 in io_loop_run (ioloop=0x559b3db00770) at ioloop.c:564 > __FUNCTION__ = "io_loop_run" > #25 0x00007fe01d981213 in master_service_run (service=0x559b3db00610, callback=) at master-service.c:642 > No locals. > #26 0x0000559b3d60a8f7 in main (argc=1, argv=0x559b3db00390) at main.c:443 > set_roots = {0x559b3d833740 , 0x0} > login_set = {auth_socket_path = 0x559b3daf8048 "\001", postlogin_socket_path = 0x0, postlogin_timeout_secs = 60, callback = 0x559b3d624150 , failure_callback = 0x559b3d623ac0 , > request_auth_token = 1} > service_flags = > storage_service_flags = > username = 0x0 > c = > (gdb) > From ad+lists at uni-x.org Thu Aug 27 17:55:36 2015 From: ad+lists at uni-x.org (Alexander Dalloz) Date: Thu, 27 Aug 2015 19:55:36 +0200 Subject: Mailbox can't be created In-Reply-To: References: Message-ID: <55DF4F18.2030300@uni-x.org> Am 27.08.2015 um 13:46 schrieb Paulo Matos: > On 27/08/15 10:42, Muzaffer Tolga Ozses wrote: >> Please check permissions as this sounds like a permissions issue m > > I am confused then. /home/vmail/Maildir is 700 for vmail:vmail. If > dovecot-lda runs as vmail then there's no problem, if it runs as root it > shouldn't face problems either. That statement is not correct nowadays as UNIX permissions aren't all any longer. There are tools like AppArmor and in your case with Fedora there is SELinux preventing even actions by root user. Make sure the SELinux context fits. "/home/vmail" and the subfolder "Maildir" has to be SELinux labeled correctly. ausearch -m avc > How can I determine which user dovecot-lda runs under? > > Thanks, Alexander From skdovecot at smail.inf.fh-brs.de Fri Aug 28 07:54:02 2015 From: skdovecot at smail.inf.fh-brs.de (Steffen Kaiser) Date: Fri, 28 Aug 2015 09:54:02 +0200 (CEST) Subject: Merging maildirs In-Reply-To: <2004270624.1949627.1440685608316.JavaMail.yahoo@mail.yahoo.com> References: <2004270624.1949627.1440685608316.JavaMail.yahoo@mail.yahoo.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thu, 27 Aug 2015, Petter Gunnerud wrote: > How do I merge files from one maildir into another maildir? > 5 days ago the internet connection at the office broke, and I decided to > power on the last copy of the server at my home and change public dns so > that users could access their new mails via webmail from this spare > server. (I didn't allow imap through firewall.) > Now the network at the office is back. DNS is updated so that mails are > delivered to the office server again. But all the mails sent / received > during those 5 days is only at my house. How can I merge those into the > office server?I'd like to do this as a server administrator. I do not > want to have each user forward the mails to them self from the > temporarily webmail. You have now: two copies of old messages, 1 set with new ones and another one with newer ones. Both copies have a different set of messages, that habe been moved, copied, flagged, tagged, and expunged. > I tried to just copy the mail files on my own account - like I did once when I used courier-imap, but that did not work with dovecot. You say, you use Maildir. If you copy the files from {new,cur} of the backup server into the corresponding directories of the production server (no dovecot* files!), the mtime of the {new,cur} directories of the production server should update und Dovecot is to detect the changes. It will then re-index the new messages. However, you might run into issues, because message tagging or flagging is reflected in the filename following the ','. E.g: 1426783429.M295788P8020.mailsrv,S=5497,W=5593:2,Sa 'Sa' are tags and flags. The rest is the 'stem'. I did "merging" in this way: 1) copy the backup files to the production server, 2) remove all files from this set, which are not located in cur/ or new/ directories, 3) check, which files in both sets have the same content, ignoring filename and mtime and such, those are removed from the backup set, 4) move all remaining files retaining the path into the production Maildir. Step 3 is a bit tricky, because users may copy one message into another mailbox, thus making duplicates. You can compare mailbox by mailbox or Maildir-wide. Both ways have pros and cons. - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVeATmnz1H7kL/d9rAQJz7wf+JlG0lNb9454aUuVFtnGLx+bs4122fZgI j2o5YewEWUze+FJEdwJXRa+CD2OjdUrcnSJnAPdD2f3M6WMuVNiO1+3oG149TfNJ TZu1KhyopsU49snFBMjhjU5/1w9GJTnfPd7rFKBa841lYwbQmw9sdnBnkkdcJWG0 oZGgfRwcsKpasVwiqwvjF62lwoWImRsFvsCDIdv81lGBBmRSULs4CwG0wl/qi94B WCAznbDCB4v6dBD2bvsrX+62MHtv+V4d34beDZAEXEv697Ge4Cs1cCTC+gQjHcyH e8sNMpEaHxgpOMjYM8DwDB6kNQen9gbG9cEaBl8rxRGTx3yjjaRSYA== =MP88 -----END PGP SIGNATURE----- From skdovecot at smail.inf.fh-brs.de Fri Aug 28 07:56:04 2015 From: skdovecot at smail.inf.fh-brs.de (Steffen Kaiser) Date: Fri, 28 Aug 2015 09:56:04 +0200 (CEST) Subject: Allow delivery to existing accounts only with LDAP and static In-Reply-To: <55DF3629.7090005@gmail.com> References: <55DF3629.7090005@gmail.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thu, 27 Aug 2015, Miloslav H?la wrote: > we are using LDAP binding as a passdb, and static with allow_all_users=yes as > an userdb. > > Works fine, but problem is, Maildirs are created for non-existent accounts > too. We would like to prevent it. > > The LDAP binding does not supporta user lookups. Is the correct way to use > checkpassword as a passdb before LDAP, check for account existency here and: "the correct way" is to reject messages to non-existant users by the MTA. Which one do you use? - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVeAUFHz1H7kL/d9rAQKdQQgAsewxfUyP2p67bpQNCZQSOJz+kVmkMVl5 LBFu+fjOoPh9KH2qRsiK0ldrAj2sMljZURInedrakw9AgKQbj8HNiPkPoPylhJh7 nWfkTdMGTJc82hfN1MGDlNG0LlTErsts92UVxrJ//LzTezqwYPBXRImnYg604zLx 06ePXrgUKoQ5xXdxJUMfUzWDfC2k/JeNu3dqECGFkhuZQfcJT+Y9T9zfeJYqjJGS 6QEBJknqYihBp9Dpr3a2eJckiEaC1qqYIiYE4/6+QIDmJkWqyveRtuD/oNGmUy1P D964VNyY/jExYRCbWZjLk7Lg/TUmkAbSWIhcydctmBQZZGeRpmIHZw== =Of1j -----END PGP SIGNATURE----- From miloslav.hula at gmail.com Fri Aug 28 08:45:06 2015 From: miloslav.hula at gmail.com (=?UTF-8?Q?Miloslav_H=c5=afla?=) Date: Fri, 28 Aug 2015 10:45:06 +0200 Subject: Allow delivery to existing accounts only with LDAP and static In-Reply-To: References: <55DF3629.7090005@gmail.com> Message-ID: <55E01F92.4020801@gmail.com> Dne 28.8.2015 v 9:56 Steffen Kaiser napsal(a): >> we are using LDAP binding as a passdb, and static with >> allow_all_users=yes as an userdb. >> >> Works fine, but problem is, Maildirs are created for non-existent >> accounts too. We would like to prevent it. >> >> The LDAP binding does not supporta user lookups. Is the correct way to >> use checkpassword as a passdb before LDAP, check for account existency >> here and: > > "the correct way" is to reject messages to non-existant users by the MTA. > > Which one do you use? We are using Postfix. Thanks in advance. -- Miloslav From skdovecot at smail.inf.fh-brs.de Fri Aug 28 09:07:36 2015 From: skdovecot at smail.inf.fh-brs.de (Steffen Kaiser) Date: Fri, 28 Aug 2015 11:07:36 +0200 (CEST) Subject: Allow delivery to existing accounts only with LDAP and static In-Reply-To: <55E01F92.4020801@gmail.com> References: <55DF3629.7090005@gmail.com> <55E01F92.4020801@gmail.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, 28 Aug 2015, Miloslav H?la wrote: > Dne 28.8.2015 v 9:56 Steffen Kaiser napsal(a): >>> we are using LDAP binding as a passdb, and static with >>> allow_all_users=yes as an userdb. >>> >>> Works fine, but problem is, Maildirs are created for non-existent >>> accounts too. We would like to prevent it. >>> >>> The LDAP binding does not supporta user lookups. Is the correct way to >>> use checkpassword as a passdb before LDAP, check for account existency >>> here and: >> >> "the correct way" is to reject messages to non-existant users by the MTA. >> >> Which one do you use? > > We are using Postfix. Then this link is probably helpful: http://www.postfix.org/LDAP_README.html - -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEVAwUBVeAk2Hz1H7kL/d9rAQIQfAgAxS3oFXkU8wbU24x1oh+eNh9p7PsvABe2 FFwXS1VTKmaND19t7tLcz53h6NWtIXcAyuoLmA1vKXqLtQV5Y9nR4W9tpbDmultG toD6L8e22ooeT3m5ancQVYUDMJ/kEYwryRUP14HLwhXfrF1uQ5FcZGQ+nBl9rckR 8q2n9q5ZKcxVisWsECaoQU2XrNnXQ9AwPvUrX0dxz5YwUAFtNPynTDyoh5TgSLU0 Qr0hc9HazJVIVH2GBVwo0xIGMNbCVMJv4ISXc09aEN4gBXHTa3pNWUzMW+bQElEk kEfEKNQWdc3g+tv6LZKcBv1CjG3eoSpMsd/MryMXcGCjOl90cyXjNA== =deBY -----END PGP SIGNATURE----- From miloslav.hula at gmail.com Fri Aug 28 09:16:37 2015 From: miloslav.hula at gmail.com (=?UTF-8?Q?Miloslav_H=c5=afla?=) Date: Fri, 28 Aug 2015 11:16:37 +0200 Subject: Allow delivery to existing accounts only with LDAP and static In-Reply-To: References: <55DF3629.7090005@gmail.com> <55E01F92.4020801@gmail.com> Message-ID: <55E026F5.9090206@gmail.com> Dne 28.8.2015 v 11:07 Steffen Kaiser napsal(a): > On Fri, 28 Aug 2015, Miloslav H?la wrote: >> Dne 28.8.2015 v 9:56 Steffen Kaiser napsal(a): >>>> we are using LDAP binding as a passdb, and static with >>>> allow_all_users=yes as an userdb. >>>> >>>> Works fine, but problem is, Maildirs are created for non-existent >>>> accounts too. We would like to prevent it. >>>> >>>> The LDAP binding does not supporta user lookups. Is the correct way to >>>> use checkpassword as a passdb before LDAP, check for account existency >>>> here and: >>> >>> "the correct way" is to reject messages to non-existant users by the >>> MTA. >>> >>> Which one do you use? >> >> We are using Postfix. > > Then this link is probably helpful: > > http://www.postfix.org/LDAP_README.html Thank you Steffen, at first, I didn't realized that MTA should reject it. We can use LDAP only for auth binds for now, but thanks to pointing me out. Best regards, Miloslav From axel.luttgens at skynet.be Fri Aug 28 16:52:39 2015 From: axel.luttgens at skynet.be (Axel Luttgens) Date: Fri, 28 Aug 2015 18:52:39 +0200 Subject: IMAP hibernate feature committed In-Reply-To: References: Message-ID: <3A0ACAF0-10D4-42C4-B2ED-0BB71B880306@skynet.be> > Le 25 ao?t 2015 ? 01:27, Timo Sirainen a ?crit : > > http://hg.dovecot.org/dovecot-2.2/rev/64c73e6bd397 > > Today I finally committed the "imap-hibernate" feature that I first started developing about a year ago (and had been thinking about for several years before that). The main purpose here is to reduce the number of imap processes and the amount of memory they use by moving IDLEing connections into imap-hibernate processes where they are waiting for something to happen (changes to the mailbox or IMAP client to do something). Mailbox changes are noticed only if inotify/kqueue is enabled, so without them the hibernation feature won't work well. > > [?] Wow! I remember you told us about such a development some times ago; many, many thanks for having persevered. Unfortunately, I won?t be able to test that feature in a near future. In the meantime, I?ll be enthusiastically reading any report our list?s fellows could share. :-) Axel From teemu.huovila at dovecot.fi Fri Aug 28 17:22:22 2015 From: teemu.huovila at dovecot.fi (Teemu Huovila) Date: Fri, 28 Aug 2015 20:22:22 +0300 Subject: IMAP hibernate feature committed In-Reply-To: <20150827163914.GA9085@nihlus.leuxner.net> References: <20150825080136.GA47256@nihlus.leuxner.net> <20150825175529.GA13298@nihlus.leuxner.net> <20150826103338.GA8337@nihlus.leuxner.net> <55DEFB81.4050805@dovecot.fi> <20150827163914.GA9085@nihlus.leuxner.net> Message-ID: <55E098CE.9040208@dovecot.fi> On 08/27/2015 07:39 PM, Thomas Leuxner wrote: > * Teemu Huovila 2015.08.27 13:58: > >> Did you specify a value other than zero for 'imap_hibernate_timeout'? > > Yes I did: > > $ doveconf imap_hibernate_timeout > imap_hibernate_timeout = 1 mins > > I sometimes see one imap-hibernate process (only one), but several imap processes active which should be idling... Does "should be" mean you know or suspect the clients have issued the IMAP IDLE command more than one minute ago? If yes and you dont see any errors in Dovecot logs, I do not know why that is. Teemu From someone.somewhere123123 at gmail.com Fri Aug 28 18:44:08 2015 From: someone.somewhere123123 at gmail.com (efs efefsfse) Date: Fri, 28 Aug 2015 20:44:08 +0200 Subject: Dovecot proxy and password scheme updating Message-ID: Hello, I run a dovecot proxy (which is doing authentification itself) and a dovecot backend on separate boxes. I want to change the actual password scheme. Since postlogin scripting is not supported by proxies and %w variable is not forwaded to dovecot backend is there any way that I can update the password scheme with my current setup ? Thanks in advance. From anmeyer at mailbox.org Sat Aug 29 08:10:10 2015 From: anmeyer at mailbox.org (Andreas Meyer) Date: Sat, 29 Aug 2015 10:10:10 +0200 Subject: virtual users with postfix Message-ID: <20150829101010.1ec50d17@workstation.bitcorner.intern> Hello! My passwd file looks like this: info at nimmini.de:{CRAM-MD5}e33cbd144b372a9c919228923464c806b57e35df2d6d1484ceaabe7e3530cd58:5000:5000::/var/spool/vhosts/nimmini.de/info::userdb_mail=maildir:~/ When I login for the first time, I see this in the log: Aug 29 09:52:57 imap(info at nimmini.de): Error: mkdir_parents(/var/lib/dovecot/db) failed: Permission denied Aug 29 09:52:57 imap(info at nimmini.de): Error: acl: dict commit failed Although the mailbox "info" is created with this content: drwx------ 2 5000 5000 4096 29. Aug 09:52 cur -rw------- 1 5000 5000 0 29. Aug 09:52 dovecot-acl-list -rw------- 1 5000 5000 272 29. Aug 09:52 dovecot.index.log -rw------- 1 5000 5000 72 29. Aug 09:52 dovecot.mailbox.log -rw------- 1 5000 5000 51 29. Aug 09:52 dovecot-uidlist -rw------- 1 5000 5000 8 29. Aug 09:52 dovecot-uidvalidity -r--r--r-- 1 5000 5000 0 29. Aug 09:52 dovecot-uidvalidity.55e164dc drwx------ 5 5000 5000 4096 29. Aug 09:52 .Drafts drwx------ 2 5000 5000 4096 29. Aug 09:52 new drwx------ 5 5000 5000 4096 29. Aug 09:52 .Sent -rw------- 1 5000 5000 18 29. Aug 09:52 subscriptions drwx------ 2 5000 5000 4096 29. Aug 09:52 tmp drwx------ 5 5000 5000 4096 29. Aug 09:52 .Trash /var/lib looks like this: ... drwxr-xr-x 2 root root 4096 29. Aug 00:28 dovecot ... Should the folder dovecot be owned by the user dovecot? What causes the error above? Greetings Andreas From dravion.smith at gmx.net Sat Aug 29 08:41:09 2015 From: dravion.smith at gmx.net (dravion.smith at gmx.net) Date: Sat, 29 Aug 2015 10:41:09 +0200 Subject: virtual users with postfix In-Reply-To: <20150829101010.1ec50d17@workstation.bitcorner.intern> References: <20150829101010.1ec50d17@workstation.bitcorner.intern> Message-ID: <55E17025.9060904@gmx.net> Hi Andreas, Am 29.08.2015 um 10:10 schrieb Andreas Meyer: > Aug 29 09:52:57 imap(info at nimmini.de): Error: mkdir_parents(/var/lib/dovecot/db) failed: Permission denied > Aug 29 09:52:57 imap(info at nimmini.de): Error: acl: dict commit failed > try chmod /var/lib/dovecot 755 Greetings, Dravion From anmeyer at mailbox.org Sat Aug 29 08:44:00 2015 From: anmeyer at mailbox.org (Andreas Meyer) Date: Sat, 29 Aug 2015 10:44:00 +0200 Subject: virtual users with postfix In-Reply-To: <55E17025.9060904@gmx.net> References: <20150829101010.1ec50d17@workstation.bitcorner.intern> <55E17025.9060904@gmx.net> Message-ID: <20150829104400.7b7c0e68@workstation.bitcorner.intern> Hello! "dravion.smith at gmx.net" schrieb am 29.08.15 um 10:41:09 Uhr: > Am 29.08.2015 um 10:10 schrieb Andreas Meyer: > > Aug 29 09:52:57 imap(info at nimmini.de): Error: mkdir_parents(/var/lib/dovecot/db) failed: Permission denied > > Aug 29 09:52:57 imap(info at nimmini.de): Error: acl: dict commit failed > > > try chmod /var/lib/dovecot 755 this is what it already has. Andreas From dravion.smith at gmx.net Sat Aug 29 08:53:24 2015 From: dravion.smith at gmx.net (dravion.smith at gmx.net) Date: Sat, 29 Aug 2015 10:53:24 +0200 Subject: virtual users with postfix In-Reply-To: <20150829104400.7b7c0e68@workstation.bitcorner.intern> References: <20150829101010.1ec50d17@workstation.bitcorner.intern> <55E17025.9060904@gmx.net> <20150829104400.7b7c0e68@workstation.bitcorner.intern> Message-ID: <55E17304.9040304@gmx.net> Am 29.08.2015 um 10:44 schrieb Andreas Meyer: > Hello! > > "dravion.smith at gmx.net" schrieb am 29.08.15 um 10:41:09 Uhr: > >> Am 29.08.2015 um 10:10 schrieb Andreas Meyer: >>> Aug 29 09:52:57 imap(info at nimmini.de): Error: mkdir_parents(/var/lib/dovecot/db) failed: Permission denied >>> Aug 29 09:52:57 imap(info at nimmini.de): Error: acl: dict commit failed >>> >> try chmod /var/lib/dovecot 755 > this is what it already has. > > Andreas# Hmm, did you use recursion? Maybe something like chmod -R 775 /var/lib/dovecot or chmod -R 777 /var/lib/dovecot service dovecot restart Will do the trick. Other users in the maillist archive reported this solved thadt error. But maybe youre right and dovecot needs read/write/execute permissions as well chown -R /var/lib/dovecot service dovecot restart And then relogin with Thunderbird or ou favorite IMAP client. Hope this helps. Greetings, Dravion From ml+dovecot at valo.at Sat Aug 29 09:00:23 2015 From: ml+dovecot at valo.at (Christian Kivalo) Date: Sat, 29 Aug 2015 11:00:23 +0200 Subject: virtual users with postfix In-Reply-To: <20150829104400.7b7c0e68@workstation.bitcorner.intern> References: <20150829101010.1ec50d17@workstation.bitcorner.intern> <55E17025.9060904@gmx.net> <20150829104400.7b7c0e68@workstation.bitcorner.intern> Message-ID: <5DB1CD94-5B66-4421-8FCB-F0FE477BF80C@valo.at> Am 29. August 2015 10:44:00 MESZ, schrieb Andreas Meyer : >Hello! > >"dravion.smith at gmx.net" schrieb am 29.08.15 um >10:41:09 Uhr: > >> Am 29.08.2015 um 10:10 schrieb Andreas Meyer: >> > Aug 29 09:52:57 imap(info at nimmini.de): Error: >mkdir_parents(/var/lib/dovecot/db) failed: Permission denied >> > Aug 29 09:52:57 imap(info at nimmini.de): Error: acl: dict commit >failed >> > >> try chmod /var/lib/dovecot 755 > >this is what it already has. Please share your doveconf -n output. Wild guess: changing the owner of /var/lib/dovecot to 5000:5000 should probably help. > > Andreas Regsrds - c From anmeyer at mailbox.org Sat Aug 29 09:17:12 2015 From: anmeyer at mailbox.org (Andreas Meyer) Date: Sat, 29 Aug 2015 11:17:12 +0200 Subject: virtual users with postfix In-Reply-To: <55E17304.9040304@gmx.net> References: <20150829101010.1ec50d17@workstation.bitcorner.intern> <55E17025.9060904@gmx.net> <20150829104400.7b7c0e68@workstation.bitcorner.intern> <55E17304.9040304@gmx.net> Message-ID: <20150829111712.1f94b260@workstation.bitcorner.intern> "dravion.smith at gmx.net" schrieb am 29.08.15 um 10:53:24 Uhr: > >> Am 29.08.2015 um 10:10 schrieb Andreas Meyer: > >>> Aug 29 09:52:57 imap(info at nimmini.de): Error: mkdir_parents(/var/lib/dovecot/db) failed: Permission denied > >>> Aug 29 09:52:57 imap(info at nimmini.de): Error: acl: dict commit failed > >>> > >> try chmod /var/lib/dovecot 755 > > this is what it already has. > > > > Andreas# > Hmm, did you use recursion? > Maybe something like > chmod -R 775 /var/lib/dovecot > or > chmod -R 777 /var/lib/dovecot > service dovecot restart > > Will do the trick. Other users in the maillist archive reported this > solved thadt error. I compared with a bit older installation and there I see: drwxr-x--- 3 5000 5000 4096 18. Mai 2014 dovecot # ll insgesamt 20 drwxr-xr-x 2 5000 5000 4096 8. Mai 2014 db -rw-r--r-- 1 root root 62 18. Mai 2014 instances -rw-r--r-- 1 root root 78 15. Mai 2014 mounts -rw------- 1 root root 1789 22. Dez 2013 replicator.db -rw-r--r-- 1 root root 230 15. Mai 2014 ssl-parameters.dat I' ll try with set it to 5000:5000 Andreas From anmeyer at mailbox.org Sat Aug 29 09:26:47 2015 From: anmeyer at mailbox.org (Andreas Meyer) Date: Sat, 29 Aug 2015 11:26:47 +0200 Subject: virtual users with postfix In-Reply-To: <5DB1CD94-5B66-4421-8FCB-F0FE477BF80C@valo.at> References: <20150829101010.1ec50d17@workstation.bitcorner.intern> <55E17025.9060904@gmx.net> <20150829104400.7b7c0e68@workstation.bitcorner.intern> <5DB1CD94-5B66-4421-8FCB-F0FE477BF80C@valo.at> Message-ID: <20150829112647.1b6c9b3e@workstation.bitcorner.intern> Christian Kivalo schrieb am 29.08.15 um 11:00:23 Uhr: > >> Am 29.08.2015 um 10:10 schrieb Andreas Meyer: > >> > Aug 29 09:52:57 imap(info at nimmini.de): Error: > >mkdir_parents(/var/lib/dovecot/db) failed: Permission denied > >> > Aug 29 09:52:57 imap(info at nimmini.de): Error: acl: dict commit > >failed > >> > > >> try chmod /var/lib/dovecot 755 > > > >this is what it already has. > > Please share your doveconf -n output. > > Wild guess: changing the owner of /var/lib/dovecot to 5000:5000 should probably help. Yes, that did the trick! Andreas From tss at iki.fi Sat Aug 29 10:20:38 2015 From: tss at iki.fi (Timo Sirainen) Date: Sat, 29 Aug 2015 13:20:38 +0300 Subject: IMAP hibernate feature committed In-Reply-To: <20150826103338.GA8337@nihlus.leuxner.net> References: <20150825080136.GA47256@nihlus.leuxner.net> <20150825175529.GA13298@nihlus.leuxner.net> <20150826103338.GA8337@nihlus.leuxner.net> Message-ID: <0682AF4D-2D04-4100-B542-5F2AE3C4B327@iki.fi> On 26 Aug 2015, at 13:33, Thomas Leuxner wrote: > > With this I see messages like this in the logs: > > Aug 26 09:48:06 nihlus dovecot: imap-hibernate(tlx at leuxner.net): Connection closed in=189 out=4252 hdr=0 body=0 del=0 exp=0 trash=0 > Aug 26 12:20:29 nihlus dovecot: imap-hibernate(tlx at leuxner.net): Connection closed in=109 out=4714 hdr=0 body=0 del=0 exp=0 trash=0 So hibernation is at least sometimes used. Here the IMAP client got disconnected while IDLEing. > I'm a bit puzzled as to when hibernate actually kicks in because most of the time I see normal imap processes running without them being hibernated: > > $ ps aux | grep dovecot/imap > vmail 8157 0.0 0.0 45624 9608 ? S 12:01 0:00 dovecot/imap > dovenull 8158 0.0 0.0 18332 4628 ? S 12:01 0:00 dovecot/imap-login > vmail 8159 0.0 0.0 44772 9256 ? S 12:01 0:00 dovecot/imap > dovenull 8160 0.0 0.0 18196 4652 ? S 12:01 0:00 dovecot/imap-login > vmail 8161 0.0 0.0 46072 9760 ? S 12:01 0:00 dovecot/imap > dovenull 8162 0.0 0.0 18196 4548 ? S 12:01 0:00 dovecot/imap-login > dovenull 8279 0.0 0.0 18332 4736 ? S 12:22 0:00 dovecot/imap-login > vmail 8280 0.0 0.0 40712 5164 ? S 12:22 0:00 dovecot/imap > dovenull 8341 0.0 0.0 18196 4740 ? S 12:25 0:00 dovecot/imap-login > vmail 8344 0.0 0.0 46312 10568 ? S 12:25 0:00 dovecot/imap Set verbose_proctitle=yes and you'll see what the imap processes are doing. It'll show IDLE in the ps output if the client is IDLEing. Although it doesn't say how long the client has been doing it. From tss at iki.fi Sat Aug 29 10:31:22 2015 From: tss at iki.fi (Timo Sirainen) Date: Sat, 29 Aug 2015 13:31:22 +0300 Subject: Mailbox can't be created In-Reply-To: References: Message-ID: <15F589F3-06B8-4191-9A9A-4BE18133F938@iki.fi> On 27 Aug 2015, at 12:07, Paulo Matos wrote: > > Aug 27 10:02:29 lda(): Info: msgid=: save failed to open mailbox PreINBOX: Mailbox can't be created I have no idea how you could get that error message. Looking at the code I don't see any way it could happen.. This error message exists only if the mailbox is tried to be created for a "fail" storage, which you can't really do. I'd maybe try recompiling Dovecot or something. From tss at iki.fi Sat Aug 29 10:36:44 2015 From: tss at iki.fi (Timo Sirainen) Date: Sat, 29 Aug 2015 13:36:44 +0300 Subject: IMAP hibernate feature committed In-Reply-To: <0682AF4D-2D04-4100-B542-5F2AE3C4B327@iki.fi> References: <20150825080136.GA47256@nihlus.leuxner.net> <20150825175529.GA13298@nihlus.leuxner.net> <20150826103338.GA8337@nihlus.leuxner.net> <0682AF4D-2D04-4100-B542-5F2AE3C4B327@iki.fi> Message-ID: <292F9C6A-0EA9-45C0-BDB8-4143EC0613E1@iki.fi> On 29 Aug 2015, at 13:20, Timo Sirainen wrote: > >> I'm a bit puzzled as to when hibernate actually kicks in because most of the time I see normal imap processes running without them being hibernated: Also if you set mail_debug=yes Dovecot will log "Couldn't hibernate imap client:" lines if it attempted to start hibernation but it couldn't be done for whatever reason. From me at junc.eu Sat Aug 29 10:44:15 2015 From: me at junc.eu (Benny Pedersen) Date: Sat, 29 Aug 2015 12:44:15 +0200 Subject: virtual users with postfix In-Reply-To: <55E17304.9040304@gmx.net> References: <20150829101010.1ec50d17@workstation.bitcorner.intern> <55E17025.9060904@gmx.net> <20150829104400.7b7c0e68@workstation.bitcorner.intern> <55E17304.9040304@gmx.net> Message-ID: dravion.smith at gmx.net skrev den 2015-08-29 10:53: > chmod -R 777 /var/lib/dovecot i cant figure out why this one is a no go... From dravion.smith at gmx.net Sat Aug 29 11:14:19 2015 From: dravion.smith at gmx.net (dravion.smith at gmx.net) Date: Sat, 29 Aug 2015 13:14:19 +0200 Subject: virtual users with postfix In-Reply-To: References: <20150829101010.1ec50d17@workstation.bitcorner.intern> <55E17025.9060904@gmx.net> <20150829104400.7b7c0e68@workstation.bitcorner.intern> <55E17304.9040304@gmx.net> Message-ID: <55E1940B.3@gmx.net> Am 29.08.2015 um 12:44 schrieb Benny Pedersen: > dravion.smith at gmx.net skrev den 2015-08-29 10:53: > >> chmod -R 777 /var/lib/dovecot > > i cant figure out why this one is a no go... blame on me chmod -R 777 /var/lib/dovecot From emailgrant at gmail.com Sat Aug 29 16:23:22 2015 From: emailgrant at gmail.com (Grant) Date: Sat, 29 Aug 2015 09:23:22 -0700 Subject: Post-migration questions Message-ID: I just migrated from courier-imap to dovecot and dovecot is a lot faster. My only client is webmail on the same server as dovecot. Max ~10 users. Should I still run imapproxy like I did with courier? Does dovecot periodically empty the trash folder like courier does? I reduced the number of processes courier was running to save RAM and improve performance. Are there any performance tweaks I should consider for dovecot? - Grant From me at junc.eu Sat Aug 29 16:36:55 2015 From: me at junc.eu (Benny Pedersen) Date: Sat, 29 Aug 2015 18:36:55 +0200 Subject: virtual users with postfix In-Reply-To: <55E1940B.3@gmx.net> References: <20150829101010.1ec50d17@workstation.bitcorner.intern> <55E17025.9060904@gmx.net> <20150829104400.7b7c0e68@workstation.bitcorner.intern> <55E17304.9040304@gmx.net> <55E1940B.3@gmx.net> Message-ID: <115ef0f6cf559bc4c79b58492669b93e@junc.eu> dravion.smith at gmx.net skrev den 2015-08-29 13:14: > blame on me i will > chmod -R 777 /var/lib/dovecot Dovecot is an open source IMAP and POP3 server for Linux/UNIX-like systems, written primarily with security in mind. Developed by Timo ... did Timo do a job in wain ? From ml+dovecot at valo.at Sat Aug 29 17:09:02 2015 From: ml+dovecot at valo.at (Christian Kivalo) Date: Sat, 29 Aug 2015 19:09:02 +0200 Subject: Post-migration questions In-Reply-To: References: Message-ID: <990d16b9497d322fa7e5d9842f74c9c5@valo.at> Hi, On 2015-08-29 18:23, Grant wrote: > I just migrated from courier-imap to dovecot and dovecot is a lot > faster. My only client is webmail on the same server as dovecot. Max > ~10 users. > > Should I still run imapproxy like I did with courier? I'd say this depends on the webmail you're using. I used to go without imapproxy while i used Hastymail2. After switching to Roundcube at first the webmail was extremly slow, imapproxy helped for my use case (personal server, 8 users). I think roundcube is opening and closing the imap connection for every action but i still haven't looked into this. > Does dovecot periodically empty the trash folder like courier does? > > I reduced the number of processes courier was running to save RAM and > improve performance. Are there any performance tweaks I should > consider for dovecot? Check the wiki http://wiki2.dovecot.org for tuning tipps. > - Grant hth - christian From tlx at leuxner.net Sat Aug 29 19:08:01 2015 From: tlx at leuxner.net (Thomas Leuxner) Date: Sat, 29 Aug 2015 21:08:01 +0200 Subject: IMAP hibernate feature committed In-Reply-To: <0682AF4D-2D04-4100-B542-5F2AE3C4B327@iki.fi> References: <20150825080136.GA47256@nihlus.leuxner.net> <20150825175529.GA13298@nihlus.leuxner.net> <20150826103338.GA8337@nihlus.leuxner.net> <0682AF4D-2D04-4100-B542-5F2AE3C4B327@iki.fi> Message-ID: <20150829190801.GA2031@nihlus.leuxner.net> * Timo Sirainen 2015.08.29 12:20: > Set verbose_proctitle=yes and you'll see what the imap processes are doing. It'll show IDLE in the ps output if the client is IDLEing. Although it doesn't say how long the client has been doing it. dovenull 1318 0.0 0.0 18340 4772 ? S 17:50 0:00 dovecot/imap-login [1 connections (1 TLS)] dovenull 1333 0.0 0.0 18340 4576 ? S 17:50 0:00 dovecot/imap-login [1 connections (1 TLS)] dovenull 1335 0.0 0.0 18200 4652 ? S 17:50 0:00 dovecot/imap-login dovenull 1336 0.0 0.0 18200 4676 ? S 17:50 0:00 dovecot/imap-login dovenull 1342 0.0 0.0 18340 4772 ? S 17:51 0:00 dovecot/imap-login [1 connections (1 TLS)] dovenull 1704 0.0 0.0 18200 4572 ? S 19:15 0:00 dovecot/imap-login dovenull 1709 0.0 0.0 18200 4652 ? S 19:15 0:00 dovecot/imap-login dovenull 1711 0.0 0.0 18200 4636 ? S 19:15 0:00 dovecot/imap-login dovenull 1994 0.0 0.0 18200 4740 ? S 20:56 0:00 dovecot/imap-login dovenull 1997 0.0 0.0 18200 4568 ? S 20:56 0:00 dovecot/imap-login vmail 1998 0.0 0.0 45448 10048 ? S 20:56 0:00 dovecot/imap [tlx at leuxner.net 1.2.3.4 IDLE] dovenull 1999 0.0 0.0 18340 4732 ? S 20:56 0:00 dovecot/imap-login [1 connections (1 TLS)] dovenull 2001 0.0 0.0 18200 4796 ? S 20:56 0:00 dovecot/imap-login vmail 2002 0.0 0.0 52880 13500 ? S 20:56 0:00 dovecot/imap [tlx at leuxner.net 1.2.3.4 IDLE] dovecot 2008 0.0 0.0 9380 936 ? S 20:57 0:00 dovecot/imap-hibernate [2 connections] After disabling "check for new messages" polling in the client it looks a lot better, multiple connections have been hibernated. Thanks. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: Digital signature URL: From emailgrant at gmail.com Sat Aug 29 20:23:41 2015 From: emailgrant at gmail.com (Grant) Date: Sat, 29 Aug 2015 13:23:41 -0700 Subject: Post-migration questions In-Reply-To: <990d16b9497d322fa7e5d9842f74c9c5@valo.at> References: <990d16b9497d322fa7e5d9842f74c9c5@valo.at> Message-ID: >> I just migrated from courier-imap to dovecot and dovecot is a lot >> faster. My only client is webmail on the same server as dovecot. Max >> ~10 users. >> >> Should I still run imapproxy like I did with courier? > > I'd say this depends on the webmail you're using. I used to go without > imapproxy while i used Hastymail2. After switching to Roundcube at first the > webmail was extremly slow, imapproxy helped for my use case (personal > server, 8 users). I think roundcube is opening and closing the imap > connection for every action but i still haven't looked into this. Wow, thank you, now it's even faster. I'm using Roundcube too. >> Does dovecot periodically empty the trash folder like courier does? Is this the right choice here: http://wiki2.dovecot.org/Plugins/Expire - Grant From tlx at leuxner.net Sun Aug 30 10:43:13 2015 From: tlx at leuxner.net (Thomas Leuxner) Date: Sun, 30 Aug 2015 12:43:13 +0200 Subject: HG port=0 changes still problematic Message-ID: <20150830104312.GA63195@nihlus.leuxner.net> Aug 30 12:35:04 nihlus.leuxner.net systemd[1]: Started Dovecot IMAP/POP3 email server. Aug 30 12:35:04 nihlus.leuxner.net dovecot[60404]: doveconf: Fatal: Error in configuration file /etc/dovecot/conf.d/10-master.conf line 24: Invalid port number 0 Aug 30 12:35:04 nihlus.leuxner.net systemd[1]: dovecot.service: main process exited, code=exit System with HG e1c7fcd62813 fails on inet_listener imaps { port = 0 #port = 993 #ssl = yes } Regards Thomas -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: Digital signature URL: From stephan at rename-it.nl Mon Aug 31 00:13:09 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Mon, 31 Aug 2015 02:13:09 +0200 Subject: HG port=0 changes still problematic In-Reply-To: <20150830104312.GA63195@nihlus.leuxner.net> References: <20150830104312.GA63195@nihlus.leuxner.net> Message-ID: <55E39C15.6010807@rename-it.nl> Op 8/30/2015 om 12:43 PM schreef Thomas Leuxner: > Aug 30 12:35:04 nihlus.leuxner.net systemd[1]: Started Dovecot IMAP/POP3 email server. > Aug 30 12:35:04 nihlus.leuxner.net dovecot[60404]: doveconf: Fatal: Error in configuration file /etc/dovecot/conf.d/10-master.conf line 24: Invalid port number 0 > Aug 30 12:35:04 nihlus.leuxner.net systemd[1]: dovecot.service: main process exited, code=exit > > System with HG e1c7fcd62813 fails on > > inet_listener imaps { > port = 0 > #port = 993 > #ssl = yes > } Yes, yes, yes.. you may smack me now :) http://hg.rename-it.nl/dovecot-2.2-patches/file/c51a10802ce6/in_port_t_fix_2.patch Regards, Stephan. From jakob at j-mb.de Mon Aug 31 12:51:17 2015 From: jakob at j-mb.de (=?UTF-8?Q?Jakob-Matthias_B=c3=b6ttger?=) Date: Mon, 31 Aug 2015 14:51:17 +0200 Subject: Problem with the Path argument in pigeonhole 0.4.3 at ubuntu 14.04 Message-ID: <55E44DC5.8040204@j-mb.de> Hello list members I want dovecot managesieve to store its config unter ~/.dovecot.sieve additionally dovecot should read as sieve_after the file ~/.sieve It's working fine BUT dovecot managesieve is always symlinking the sieve file at activation to ~/.sieve and not to ~/.dovecot.sieve as i configured it. Any idears about this strange behaviour? dovecot is configured as follows # 2.2.9: /etc/dovecot/dovecot.conf # OS: Linux 3.13.0-53-generic x86_64 Ubuntu 14.04.3 LTS auth_mechanisms = plain login dict { acl = mysql:/etc/dovecot/dovecot-dict-sql.conf.ext expire = mysql:/etc/dovecot/dovecot-dict-expire.conf.ext } disable_plaintext_auth = no lda_mailbox_autocreate = yes lda_mailbox_autosubscribe = yes listen = *,[::] log_timestamp = "%Y-%m-%d %H:%M:%S " mail_plugins = " expire quota acl" mail_privileged_group = vmail maildir_very_dirty_syncs = yes managesieve_notify_capability = mailto managesieve_sieve_capability = fileinto reject envelope encoded-character vacation subaddress comparator-i;ascii-numeric relational regex imap4flags copy include variables body enotify environment mailbox date ihave namespace { inbox = yes location = prefix = separator = / type = private } namespace { list = children location = maildir:%%h/Maildir:INDEXPVT=~/Maildir/shared/%%u prefix = shared/%%u/ separator = / subscriptions = yes type = shared } passdb { args = /etc/dovecot/dovecot-sql.conf driver = sql } plugin { acl = vfile acl_shared_dict = proxy::acl expire = Trash expire2 = Trash/* expire3 = Spam expire4 = Junk expire_dict = proxy::expire quota = dict:user::file:/var/vmail/%d/%n/.quotausage sieve = file:~/sieve;active=~/.dovecot.sieve sieve_after = ~/.sieve } postmaster_address = admin at domain.tld protocols = imap lmtp sieve service auth { unix_listener /var/spool/postfix/private/auth { group = postfix mode = 0660 user = postfix } unix_listener auth-userdb { group = vmail mode = 0600 user = vmail } user = root } service dict { unix_listener dict { group = vmail mode = 0660 user = vmail } } service imap-login { client_limit = 1000 process_limit = 500 } service lmtp { inet_listener lmtp { address = 127.0.0.1 port = 24 } } service managesieve-login { inet_listener sieve { port = 4190 } } ssl = required ssl_cert = References: <55C21E22.10808@securepoint.de> <55C22CC4.5030006@securepoint.de> <55C331E1.8060702@securepoint.de> <55DEF697.30800@dovecot.fi> Message-ID: <20150831150201.7451060a@eugen.spdev.local> hi again, On Thu, 27 Aug 2015 14:37:59 +0300 Teemu Huovila wrote: > > However, I am unable to reproduce this. Could you post your doveconf > -n please? Im especially interested in your passdb and userdb > configurations and auth-cache settings. > just reproduced the bug with a fresh clean 2.2.18 install ldap userdb an 2 masterusers with the ACL_GROUP attribut in passwd file env output in imap-postlogin first login: AUTH_TOKEN=4adba75022f765fc3215ac5243337fd99adfdbf5 MASTER_USER=master2 SPUSER=private/johnd LOCAL_IP=127.0.0.1 USER=johnd AUTH_USER=master2 PWD=/run/dovecot USERDB_KEYS=ACL_GROUPS HOME SPUSER MASTER_USER AUTH_TOKEN AUTH_USER SHLVL=1 HOME=/home/vmail/private/johnd ACL_GROUPS=umareadmaster IP=127.0.0.1 _=/usr/bin/env logout and next login: AUTH_TOKEN=83d7ede27b4fbc4de2abad58e84e65ac1073e4ec MASTER_USER=master2 SPUSER=private/johnd LOCAL_IP=127.0.0.1 USER=johnd AUTH_USER=master2 PWD=/run/dovecot USERDB_KEYS=HOME SPUSER MASTER_USER AUTH_TOKEN AUTH_USER SHLVL=1 HOME=/home/vmail/private/johnd IP=127.0.0.1 _=/usr/bin/env ############################## % doveconf -n: # 2.2.18: /etc/dovecot/dovecot.conf # OS: Linux 3.12.44-gentoo x86_64 Gentoo Base System release 2.2 auth_cache_negative_ttl = 30 mins auth_cache_size = 10 k auth_master_user_separator = * auth_use_winbind = yes auth_username_chars = auth_verbose = yes log_path = /var/log/dovecot.log mail_gid = vmail mail_home = /home/vmail/private/%u mail_location = maildir:~/Maildir:LAYOUT=fs:INBOX=~/Maildir/INBOX mail_uid = vmail namespace { inbox = yes location = mailbox Sent { auto = subscribe special_use = \Sent } prefix = separator = / subscriptions = yes type = private } namespace { hidden = no inbox = no list = children location = maildir:/home/vmail/public/%%Lu/Maildir:LAYOUT=fs:INBOX=/home/vmail/public/%%Lu/Maildir/INBOX prefix = public/%%u/ separator = / subscriptions = no type = shared } passdb { args = /etc/dovecot/master-users1 driver = passwd-file master = yes } passdb { args = /etc/dovecot/master-users2 driver = passwd-file master = yes } service auth { unix_listener auth-client { group = mode = 0600 user = $default_internal_user } unix_listener auth-login { group = mode = 0600 user = $default_internal_user } unix_listener auth-master { group = mode = 0600 user = $default_internal_user } unix_listener auth-userdb { group = vmail mode = 0660 user = $default_internal_user } unix_listener login/login { group = mode = 0666 user = $default_internal_user } user = $default_internal_user } service imap-login { inet_listener imap { port = 143 } } service imap-postlogin { executable = script-login /usr/libexec/dovecot/imap-postlogin user = vmail } service imap { executable = imap imap-postlogin } ssl_cert = auth_master_user_separator = * # Example master user passdb using passwd-file. You can use any passdb though. passdb { driver = passwd-file master = yes args = /etc/dovecot/master-users1 # Unless you're using PAM, you probably still want the destination user to # be looked up from passdb that it really exists. pass=yes does that. #pass = yes } passdb { driver = passwd-file master = yes args = /etc/dovecot/master-users2 # Unless you're using PAM, you probably still want the destination user to # be looked up from passdb that it really exists. pass=yes does that. #pass = yes } ############################################### % cat /etc/dovecot/master-users1 master1:{SHA}xxxxxxx=::::::userdb_acl_groups=umareadmaster allow_nets=127.0.0.1 master2 is the same. Greetz From me at junc.eu Mon Aug 31 13:46:25 2015 From: me at junc.eu (Benny Pedersen) Date: Mon, 31 Aug 2015 15:46:25 +0200 Subject: HG =?UTF-8?Q?port=3D=30=20changes=20still=20problematic?= In-Reply-To: <55E39C15.6010807@rename-it.nl> References: <20150830104312.GA63195@nihlus.leuxner.net> <55E39C15.6010807@rename-it.nl> Message-ID: <280d64125bef32ffc7fc333a0ee0ff2b@junc.eu> Stephan Bosch skrev den 2015-08-31 02:13: >> System with HG e1c7fcd62813 fails on >> >> inet_listener imaps { >> port = 0 >> #port = 993 >> #ssl = yes >> } > > Yes, yes, yes.. you may smack me now :) > > http://hg.rename-it.nl/dovecot-2.2-patches/file/c51a10802ce6/in_port_t_fix_2.patch but port 0 is valid, and your error msgs says invalid, hmm maybe it works, but it should not say invalid when its supported as disable that service hack :=) the above should say "imaps is disabled, cant use" From tlx at leuxner.net Mon Aug 31 14:10:48 2015 From: tlx at leuxner.net (Thomas Leuxner) Date: Mon, 31 Aug 2015 16:10:48 +0200 Subject: HG port=0 changes still problematic In-Reply-To: <55E39C15.6010807@rename-it.nl> References: <20150830104312.GA63195@nihlus.leuxner.net> <55E39C15.6010807@rename-it.nl> Message-ID: <20150831141034.GA58774@nihlus.leuxner.net> * Stephan Bosch 2015.08.31 02:13: > Yes, yes, yes.. you may smack me now :) > > http://hg.rename-it.nl/dovecot-2.2-patches/file/c51a10802ce6/in_port_t_fix_2.patch I wouldn't go that far :) Latest HG works. Thanks -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: Digital signature URL: From stephan at rename-it.nl Mon Aug 31 16:17:09 2015 From: stephan at rename-it.nl (Stephan Bosch) Date: Mon, 31 Aug 2015 18:17:09 +0200 Subject: Problem with the Path argument in pigeonhole 0.4.3 at ubuntu 14.04 In-Reply-To: <55E44DC5.8040204@j-mb.de> References: <55E44DC5.8040204@j-mb.de> Message-ID: <55E47E05.3060002@rename-it.nl> Op 31-8-2015 om 14:51 schreef Jakob-Matthias B?ttger: > Hello list members > > I want dovecot managesieve to store its config unter ~/.dovecot.sieve > additionally dovecot should read as sieve_after the file ~/.sieve > > It's working fine BUT dovecot managesieve is always symlinking the > sieve file at activation to ~/.sieve and not to ~/.dovecot.sieve as i > configured it. > Any idears about this strange behaviour? > > dovecot is configured as follows > You can enable mail_debug=yes, which will log in detail what paths it is using and why. Regards, Stephan. > # 2.2.9: /etc/dovecot/dovecot.conf > # OS: Linux 3.13.0-53-generic x86_64 Ubuntu 14.04.3 LTS > auth_mechanisms = plain login > dict { > acl = mysql:/etc/dovecot/dovecot-dict-sql.conf.ext > expire = mysql:/etc/dovecot/dovecot-dict-expire.conf.ext > } > disable_plaintext_auth = no > lda_mailbox_autocreate = yes > lda_mailbox_autosubscribe = yes > listen = *,[::] > log_timestamp = "%Y-%m-%d %H:%M:%S " > mail_plugins = " expire quota acl" > mail_privileged_group = vmail > maildir_very_dirty_syncs = yes > managesieve_notify_capability = mailto > managesieve_sieve_capability = fileinto reject envelope > encoded-character vacation subaddress comparator-i;ascii-numeric > relational regex imap4flags copy include variables body enotify > environment mailbox date ihave > namespace { > inbox = yes > location = > prefix = > separator = / > type = private > } > namespace { > list = children > location = maildir:%%h/Maildir:INDEXPVT=~/Maildir/shared/%%u > prefix = shared/%%u/ > separator = / > subscriptions = yes > type = shared > } > passdb { > args = /etc/dovecot/dovecot-sql.conf > driver = sql > } > plugin { > acl = vfile > acl_shared_dict = proxy::acl > expire = Trash > expire2 = Trash/* > expire3 = Spam > expire4 = Junk > expire_dict = proxy::expire > quota = dict:user::file:/var/vmail/%d/%n/.quotausage > sieve = file:~/sieve;active=~/.dovecot.sieve > sieve_after = ~/.sieve > } > postmaster_address = admin at domain.tld > protocols = imap lmtp sieve > service auth { > unix_listener /var/spool/postfix/private/auth { > group = postfix > mode = 0660 > user = postfix > } > unix_listener auth-userdb { > group = vmail > mode = 0600 > user = vmail > } > user = root > } > service dict { > unix_listener dict { > group = vmail > mode = 0660 > user = vmail > } > } > service imap-login { > client_limit = 1000 > process_limit = 500 > } > service lmtp { > inet_listener lmtp { > address = 127.0.0.1 > port = 24 > } > } > service managesieve-login { > inet_listener sieve { > port = 4190 > } > } > ssl = required > ssl_cert = ssl_key = ssl_protocols = !SSLv2 !SSLv3 > userdb { > args = /etc/dovecot/dovecot-sql.conf > driver = sql > } > protocol imap { > mail_max_userip_connections = 100 > mail_plugins = quota imap_quota expire acl imap_acl > } > protocol pop3 { > mail_plugins = quota > pop3_uidl_format = %08Xu%08Xv > } > protocol lda { > mail_plugins = sieve quota > } > protocol lmtp { > mail_plugins = sieve quota > }