- dovecot - dovecot.org

antispam plugin crash
by Thomas Witt 20 Aug '14

20 Aug '14

Hi, the dovecot antispam plugin crahses here on a nullpointer dereference in antispam_copy() at mailbox.c The crash tracks back to a call of antispam_classify_copy(asms->box_class, asmb->box_class) At that point in time asmb is set to 0x0, causing the process to crash with SIGSEGV. Regards, Thomas GDB output: Program received signal SIGSEGV, Segmentation fault. antispam_copy (ctx=0x14ed9f0, mail=0x14c6cc0) at mailbox.c:107 107 enum mailbox_copy_type copy_type = (gdb) bt #0 antispam_copy (ctx=0x14ed9f0, mail=0x14c6cc0) at mailbox.c:107 #1 0x00007fdadb627fcd in mailbox_copy () from /usr/lib64/dovecot/libdovecot-storage.so.0 #2 0x00007fdada6cdba0 in act_store_execute () from /usr/lib64/dovecot/libdovecot-sieve.so.0 #3 0x00007fdada6c59d9 in sieve_result_execute () from /usr/lib64/dovecot/libdovecot-sieve.so.0 #4 0x00007fdada6d4a2a in sieve_execute () from /usr/lib64/dovecot/libdovecot-sieve.so.0 #5 0x00007fdada9212cb in lda_sieve_deliver_mail () from /usr/lib64/dovecot/lib90_sieve_plugin.so #6 0x00007fdadb8cae1a in mail_deliver () from /usr/lib64/dovecot/libdovecot-lda.so.0 #7 0x0000000000405ebd in client_input_data_handle () #8 0x00007fdadb34e9bf in io_loop_call_io () from /usr/lib64/dovecot/libdovecot.so.0 #9 0x00007fdadb34f8b7 in io_loop_handler_run_internal () from /usr/lib64/dovecot/libdovecot.so.0 #10 0x00007fdadb34ea29 in io_loop_handler_run () from /usr/lib64/dovecot/libdovecot.so.0 #11 0x00007fdadb34eaa8 in io_loop_run () from /usr/lib64/dovecot/libdovecot.so.0 #12 0x00007fdadb2fe803 in master_service_run () from /usr/lib64/dovecot/libdovecot.so.0 #13 0x0000000000404995 in main () (gdb) p asms $3 = (struct antispam_mailbox *) 0x0

3 9

Panic: file istream.c: line 167 with 2.2.13 + istream.c/.h from dovecot-2-2-dffc6dd2be0d
by Toni Mattila 20 Aug '14

20 Aug '14

Hi, Getting following panics with traces: Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Error: Cached message size smaller than expected (1782 < 1872) Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Error: Maildir filename has wrong S value, renamed the file from /path/cencored/Maildir/cur/1201713258.3136.host.cencored.fi,S=1782:2,S to /path/cencored/Maildir/cur/1201713258.3136.host.cencored.fi,S=1872:2,S Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Error: Corrupted index cache file /path/cencored/Maildir/dovecot.index.cache: Broken physical size for mail UID 5758 Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Error: read(/path/cencored/Maildir/cur/1201713258.3136.host.cencored.fi,S=1782:2,S) failed: Invalid argument Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Error: read(/path/cencored/Maildir/cur/1201713258.3136.host.cencored.fi,S=1782:2,S) failed: Invalid argument (uid=5758) Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Error: Cached message size smaller than expected (1781 < 1869) Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Error: Maildir filename has wrong S value, renamed the file from /path/cencored/Maildir/cur/1201767850.29173.host.cencored.fi,S=1781:2,S to /path/cencored/Maildir/cur/1201767850.29173.host.cencored.fi,S=1869:2,S Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Error: Corrupted index cache file /path/cencored/Maildir/dovecot.index.cache: Broken physical size for mail UID 5767 Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Error: read(/path/cencored/Maildir/cur/1201767850.29173.host.cencored.fi,S=1781:2,S) failed: Invalid argument Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Error: read(/path/cencored/Maildir/cur/1201767850.29173.host.cencored.fi,S=1781:2,S) failed: Invalid argument (uid=5767) Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Error: Cached message size smaller than expected (9252 < 9346) Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Error: Maildir filename has wrong S value, renamed the file from /path/cencored/Maildir/cur/1205131753.12346.host.cencored.fi,S=9252:2,RS to /path/cencored/Maildir/cur/1205131753.12346.host.cencored.fi,S=9346:2,RS Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Error: Corrupted index cache file /path/cencored/Maildir/dovecot.index.cache: Broken physical size for mail UID 6296 Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Panic: file istream.c: line 167 (i_stream_read): assertion failed: (old_size <= _stream->pos - _stream->skip) Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Error: Raw backtrace: /usr/lib/dovecot/libdovecot.so.0 [0x575c1f] -> /usr/lib/dovecot/libdovecot.so.0 [0x575c9d] -> /usr/lib/dovecot/libdovecot.so.0 [0x575514] -> /usr/lib/dovecot/libdovecot.so.0(i_stream_read+0x13b) [0x57f99b] -> /usr/lib/dovecot/libdovecot.so.0(i_stream_read_data+0x2b) [0x57fbfb] -> /usr/lib/dovecot/libdovecot.so.0 [0x565b5e] -> /usr/lib/dovecot/libdovecot.so.0 [0x567469] -> /usr/lib/dovecot/libdovecot.so.0(message_parser_parse_next_block+0x43) [0x565cd3] -> /usr/lib/dovecot/libdovecot.so.0(message_parser_parse_body+0x2c) [0x565e4c] -> /usr/lib/dovecot/libdovecot-storage.so.0 [0x944dd2] -> /usr/lib/dovecot/libdovecot-storage.so.0(index_mail_get_parts+0x6a) [0x945c7a] -> /usr/lib/dovecot/libdovecot-storage.so.0(mail_get_parts+0x16) [0x91ad76] -> /usr/lib/dovecot/libdovecot-storage.so.0 [0x94aa7c] -> /usr/lib/dovecot/libdovecot-storage.so.0 [0x94ba89] -> /usr/lib/dovecot/libdovecot-storage.so.0(index_storage_search_next_nonblo Jul 6 08:11:52 dovecot: imap(user(a)domain.tld): Fatal: master: service(imap): child 31268 killed with signal 6 (core dumps disabled) These can of course be fixed with maildir-size-fix.pl and deleting dovecot index files, but there are thousands like these. Dovecot 1.2 didn't have any issues with these old maildirs transfered over from courierimap. Best Regards, Toni Mattila

2 9

Dovecot v2.2 FTS is not indexing "text/html" emails...
by Akash 20 Aug '14

20 Aug '14

Hi, I am not sure its intended or a fault in the newest Dovecot versions. I have been using Dovecot v1.2.15 on Debian squeeze and FTS is working as expected. When I search a quoted string "very good", I get 107 results including plain and HTML emails which have this phrase. In order to compare the benefits of lucene over squat, I recently started testing dovecot v2.2.13 on Debian Sid with the same maildir content. But now the same search "very good" yielded just 8 results. I thought it could be some problem with lucene so I tried switching to squat and got 107 results again. After this I deleted the old squat search index files created by v1.2.15 and re-indexed the mail-box by using doveadm index command. Now the same squat search is giving 8 results just as lucene. So I have realized that its not a problem with just lucene but FTS in newer dovecot isn't indexing those emails which have Content-type as text/html. Thus if a mail is like this: Content-Type: text/html <b>He is very good.</b> It isn't shown in search by the squat indexes created using dovecot v2.2.13. I have done further testing on some sample emails which confirmed this behavior. Why is this so? -Regards, Akash

2 2

2.2.13 + hg: Panic: file ioloop.c: line 39 (io_add_file): assertion failed: (fd >= 0)
by Toni Mattila 20 Aug '14

20 Aug '14

Hi, Found this in logfiles: Jul 25 14:12:38 dovecot: pop3(cen(a)so.red): Error: UIDL: File name lookup failed: Message was expunged Jul 25 14:12:38 dovecot: pop3(cen(a)so.red): Disconnected: POP3 UIDLs couldn't be listed top=187/356348, retr=0/0, del=0/1207, size=789346199 Jul 25 14:12:38 dovecot: pop3(cen(a)so.red): Panic: file ioloop.c: line 39 (io_add_file): assertion failed: (fd >= 0) Jul 25 14:12:38 dovecot: pop3(cen(a)so.red): Error: Raw backtrace: /usr/lib/dovecot/libdovecot.so.0 [0xbdfc1f] -> /usr/lib/dovecot/libdovecot.so.0 [0xbdfc9d] -> /usr/lib/dovecot/libdovecot.so.0 [0xbdf514] -> /usr/lib/dovecot/libdovecot.so.0 [0xbf162d] -> /usr/lib/dovecot/libdovecot.so.0(io_add_istream+0x3f) [0xbf272f] -> dovecot/pop3 [0x804d149] -> /usr/lib/dovecot/libdovecot.so.0 [0xbff17a] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_call_io+0x48) [0xbf1d88] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_handler_run_internal+0xff) [0xbf375f] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_handler_run+0x1e) [0xbf1e2e] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_run+0x48) [0xbf1eb8] -> /usr/lib/dovecot/libdovecot.so.0(master_service_run+0x2d) [0xb9842d] -> dovecot/pop3(main+0x2a9) [0x804b3b9] -> /lib/i686/nosegneg/libc.so.6(__libc_start_main+0xdc) [0x6f7dec] -> dovecot/pop3 [0x804aeb1] Jul 25 14:12:38 dovecot: pop3(cen(a)so.red): Fatal: master: service(pop3): child 27510 killed with signal 6 (core dumped) (gdb) bt #0 0x00b1a402 in __kernel_vsyscall () #1 0x0070af30 in raise () from /lib/i686/nosegneg/libc.so.6 #2 0x0070c911 in abort () from /lib/i686/nosegneg/libc.so.6 #3 0x00bdfc14 in default_fatal_finish (type=<value optimized out>, status=0) at failures.c:193 #4 0x00bdfc9d in i_internal_fatal_handler (ctx=0xbf82cd34, format=0xc0c784 "file %s: line %d (%s): assertion failed: (%s)", args=0xbf82cd54 "\202\232�") at failures.c:657 #5 0x00bdf514 in i_panic (format=0xc0c784 "file %s: line %d (%s): assertion failed: (%s)") at failures.c:267 #6 0x00bf162d in io_add_file (fd=-1, condition=IO_READ, source_linenum=816, callback=0x804d150 <client_input>, context=0x83ba628) at ioloop.c:40 #7 0x00bf272f in io_add_istream (input=0x83ba960, source_linenum=816, callback=0x804d150 <client_input>, context=0x83ba628) at ioloop.c:86 #8 0x0804d149 in client_output (client=0x83ba628) at pop3-client.c:815 #9 0x00bff17a in stream_send_io (fstream=0x83baa48) at ostream-file.c:468 #10 0x00bf1d88 in io_loop_call_io (io=0x83baaf8) at ioloop.c:441 #11 0x00bf375f in io_loop_handler_run_internal (ioloop=0x83a5498) at ioloop-epoll.c:220 #12 0x00bf1e2e in io_loop_handler_run (ioloop=0x83a5498) at ioloop.c:488 #13 0x00bf1eb8 in io_loop_run (ioloop=0x83a5498) at ioloop.c:465 #14 0x00b9842d in master_service_run (service=0x83a53c0, callback=0x804b4e0 <client_connected>) at master-service.c:566 #15 0x0804b3b9 in main (argc=Cannot access memory at address 0x0 ) at main.c:277 (gdb) Thanks, Toni

2 1

[Patch] mail_log plugin supports remote, local ip addresses
by Vitaliy T 20 Aug '14

20 Aug '14

Hello, In attached file you will find patch for mail_log plugin to output local, remote ip addresses. It may be usefull for somebody. To enable support for dovecot 2.2.x versions change the line 'include "net.h"' instead of 'include "network.h"'. Thank you. P.S. Sorry, for previous empty message. -- With Best Regards, Vitaliy T.

2 1

auth_username_translation and LTMP problem
by Arkadiusz Miśkiewicz 20 Aug '14

20 Aug '14

auth_username_translation seems to be appliet on RCPT TO address of LTMP transport. Why dovecot is doing that? And better question - is there a way to disable auth_username_translation for LTMP but leave enabled for the rest (imap, pop3 etc) ? Background: I'm doing auth_username_translation = @= to allow logins like aaa(a)bbb.pl to be internally translated to aaa=bbb.pl. That works fine. Now my exim delivers mail to dovecot using LTMP and it does translation or it own, so it does: RCPT TO: <aaa=bbb.pl(a)mymbox.pl> Unfortunately looks like dovecot it doing translation one more time and looking in user database for aaa=bbb.pl=mymbox.pl where such user doesn't exist. Only aaa=bbb.pl user exists. Thanks, -- Arkadiusz Miśkiewicz, arekm / maven.pl

2 2

Mail Archiving/Lucene Indexes/mbox
by Larry Rosenman 20 Aug '14

20 Aug '14

I have a script (available on demand) that moves all my mail for a given month to a separate NAMESPACE (#ARCHIVE). When I do that, the lucene-indexes folder in the PRIMARY namespace seems to get cleaned out, and NOTHING gets put in the lucene-indexes folder (although it gets created) in the #ARCHIVE namespace. Is this to spec? the archiving is done via doveadm move commands. Thanks! -- Larry Rosenman http://www.lerctr.org/~ler Phone: +1 214-642-9640 (c) E-Mail: larryrtx(a)gmail.com US Mail: 108 Turvey Cove, Hutto, TX 78634-5688

2 3

Segfault with indexer-worker using Tika/Solr
by G H 20 Aug '14

20 Aug '14

Using Dovecot 2.2.13 with Tika 1.5 on Centos 5 64bit Tika is throwing an exception for some reason, which seems to be causing Dovecot's indexer to segfault. I'm using mdbox/gzipped compressed mail and private indexes for the shared folder. I'm importing mail via formail/dovecot-lda: formail -s /usr/libexec/dovecot/dovecot-lda -d admin -m Public/TestSubFolder < /tmp/Inbox.mbox dovecot -n and the GDB backtrace are attached. # find /var/mail/ -maxdepth 2 /var/mail/_public /var/mail/_public/dovecot.mailbox.log /var/mail/_public/dovecot-uidvalidity /var/mail/_public/dovecot-uidvalidity.53ecade1 /var/mail/_public/storage /var/mail/_public/subscriptions /var/mail/_public/mailboxes /var/mail/admin /var/mail/admin/dovecot-uidvalidity.53ecade2 /var/mail/admin/dovecot.mailbox.log /var/mail/admin/public /var/mail/admin/dovecot-uidvalidity /var/mail/admin/storage /var/mail/admin/subscriptions /var/mail/admin/mailboxes

2 1

BUG: Solr FTS engine appends domain name to index records
by G H 20 Aug '14

20 Aug '14

The Solr FTS engine appends domain names to the "user" and "id" fields inside Solr when indexing. However, when an IMAP search is performed, the domain name is not appended, resulting in 0 results, 100% of the time. You can reproduce this by setting up dovecot to have user names like "admin", "james", etc (as opposed to admin(a)example.com) Then import some mail (ie: "dovecot-lda -d admin < /tmp/mymail.txt"). Now attempt to search for that mail and it will not be returned as a result. You can see in the Solr GUI (http://xxxx:8983/solr/) that record is indexed as: { "uid": 1, "box": "7dff0b0d4f84df53f25300003d823887", "user": "admin(a)example.com", "id": "1/7dff0b0d4f84df53f25300003d823887/admin(a)example.com", "_version_": 1475511406809841700 } However, the IMAP search is sent to Solr performed with user:"admin" instead of user:"admin(a)example.com". Either the Solr FTS plugin needs to not append the domain name or the search does need to do so.

2 2

Trouble getting public folders working
by Bjarne Mark Kondrup 20 Aug '14

20 Aug '14

Hi i have trouble getting my public folders working properly. I have a few issues atm with them. First off i'm using version 2.0.9 as it came with CentOS and would greatly prefer to keep this version. Here is my dovecot.conf file # dovecot -n # 2.0.9: /etc/dovecot/dovecot.conf # OS: Linux 2.6.32-431.23.3.el6.centos.plus.x86_64 x86_64 CentOS release 6.5 (Final) ext4 auth_debug = yes auth_verbose = yes disable_plaintext_auth = no first_valid_uid = 101 last_valid_uid = 101 login_greeting = Dovecot ready. CEGO webmail server mail_debug = yes mail_gid = vmail mail_location = maildir:/home/vmail/domains/%d/%n mail_uid = vmail 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 mbox_write_locks = fcntl namespace { inbox = yes location = prefix = separator = / type = private } namespace { location = maildir:/home/vmail/shared:LAYOUT=fs:INDEX=/home/vmail/shared/ prefix = shared/ separator = / subscriptions = no type = public } passdb { args = /etc/dovecot/dovecot-ldap.conf driver = ldap } plugin { sieve = /home/vmail/domains/%d/%n/dovecot.sieve sieve_dir = /home/vmail/domains/%d/%n/sieve_scripts sieve_global_dir = /etc/dovecot/sieve/global/ } protocols = imap lmtp sieve service auth { unix_listener auth-client { group = postfix mode = 0660 user = postfix } unix_listener auth-master { group = vmail mode = 0660 user = vmail } user = root } service lmtp { unix_listener /var/spool/postfix/private/dovecot-lmtp { group = postfix mode = 0600 user = postfix } } service managesieve-login { inet_listener sieve { port = 4190 } } ssl = no ssl_cert = </etc/pki/dovecot/certs/dovecot.pem ssl_key = </etc/pki/dovecot/private/dovecot.pem userdb { driver = passwd } userdb { args = /etc/dovecot/dovecot-ldap.conf driver = ldap } valid_chroot_dirs = /home/vmail/domains verbose_ssl = yes protocol lmtp { log_path = /var/log/dovecot-deliver.log mail_plugins = quota sieve sieve acl postmaster_address = postmaster(a)cegotest.dk } protocol sieve { info_log_path = /var/log/dovecot-sieve.log log_path = /var/log/dovecot-sieve-errors.log managesieve_implementation_string = dovecot managesieve_max_line_length = 65536 } And my dovecot-ldap file hosts = localhost dn = cn=dovecot,dc=cegotest,dc=dk dnpass = password auth_bind = yes ldap_version = 3 base = o=hosting,dc=cegotest,dc=dk deref = never scope = subtree user_filter = (&(objectClass=JammMailAccount)(mail=%u)(accountActive=TRUE)(delete=FALSE)) pass_attrs = mail,userPassword pass_filter = (&(objectClass=JammMailAccount)(mail=%u)(accountActive=TRUE)(delete=FALSE)) default_pass_scheme = CRYPT Here is my current issues. 1. I cannot get shared folders to work with pr. user settings. What i want is a public folder that all my users can see but each user marks mail as read individually. Atm no matter how i set it up if user1 marks a mail as read, then user2 will also see that mail as read :( 2. ACL to control the permissions. Right now all users can move mail and remove mail from the shared folders. I want people (except a few specific users) to be read only on this public folders. 3. Need to know if i'm creating the folders correctly. Right now i'm under my shared folder directory creating shared_folder and 3 subfolders beneath it (cur, new and tmp). After this it shows up in eg. roundcube and i can subscribe to it. I have read all the documentation and tried all the solutions i can. Mostly moving the index and/or config files to seperate dirs and from my understanding that should give each user their own index file marking what mails have been read ect. But this doesn't happen. Hope you guys are able to help me out a little here. -Bjarne

2 1

Huge difference between the lucene index size created by v2.1 and v2.2
by Akash 20 Aug '14

20 Aug '14

Hi everyone, While examining dovecot versions v2.1 and v2.2 for their lucene search performances, I have noticed a huge difference in the index sizes created by them. Both versions were compiled on same system, against same libclucene, same configure options and were used with same dovecot.conf during run-time. I deleted the lucene-indexes folder and dovecot* files in the Maildir prior to indexing with both versions. The tests were performed on a untouched mail archive folder containing 300000 mails without any dovecot* files in it: root@server:/home/admin/mails/.Archive# ls -l total 78640 drwx------ 2 2500 2500 14098432 Aug 17 22:16 cur drwx------ 2 2500 2500 12435456 Jul 30 09:46 new drwx------ 2 2500 2500 4096 Aug 2 13:02 tmp The command used was: doveadm -v index -u admin Archive After search indexing on v2.1 resulted in: root@server:/home/admin/mails# ls -lh lucene-indexes total 390M -rw------- 1 2500 2500 390M Aug 18 07:03 _25.cfs -rw------- 1 2500 2500 20 Aug 18 07:03 segments.gen -rw------- 1 2500 2500 46 Aug 18 07:03 segments_4d Whereas dovecot v2.2 resulted in: root@server:/home/admin/mails# ls -lh lucene-indexes total 1.5G -rw------- 1 2500 2500 1.5G Aug 18 06:41 _5g.cfs -rw------- 1 2500 2500 20 Aug 18 06:41 segments.gen -rw------- 1 2500 2500 46 Aug 18 06:41 segments_az 390M vs 1.5G. That is a huge difference in size. Why is that? Thanks in advance. -Regards, Akash

2 5

Problems with dovecot 2.2.13 and monit
by Hanno Böck 20 Aug '14

20 Aug '14

Hello, When I upgraded my servers to dovecot 2.2.13 the monitoring tool monit started to send out warnings that it couldn't reach my imap/pop3 servers through ssl any more. The same problem didn't happen on non-ssl-connections. According to people on the monit list this is likely a dovecot issue: https://lists.gnu.org/archive/html/monit-general/2014-06/msg00031.html Let me quote: > the root cause of the error is, that dovecot 2.2.13 closes the > connection if SSL is used in response to LOGOUT command instead of > sending usual response. When no SSL is enabled, dovecot responses to > LOGOUT command normally. [...] > According to RFC 3501 (http://tools.ietf.org/html/rfc3501) LOGOUT is > any-state command, where the server MUST send response before closing > the connection: http://tools.ietf.org/html/rfc3501#section-3.4 > > => the problem is caused by dovecot 2.2.13 bug ... its behaviour is > inconsistent (LOGOUT in non-authenticated state works per RFC > requirement if no SSL is used and doesn't conform to RFC if SSL is > used). It is possible that the problem is related to their DoS-attack > modification, which has most probably unexpected side-effect. Maybe this is related to the DDoS-protection measures that have been added in dovecot 2.2.13. Would apprechiate if someone could have a look. cu, -- Hanno Böck http://hboeck.de/ mail/jabber: hanno(a)hboeck.de GPG: BBB51E42

3 3

Designing a Dovecot Replicate cluster
by Minhas Listas 19 Aug '14

19 Aug '14

Hi all! Do I need one Dovecot Proxy/Director for each Dovecot Replicate Cluster ? Isn't possible to put a Dovecot Proxy/Director on the border of multiple clusters ? Thanks! -- Thiago Henrique www.adminlinux.com.br

1 0

Health monitoring of backend servers
by Minhas Listas 19 Aug '14

19 Aug '14

Does Dovecot Director health monitoring of backend servers ? Or is poolmon (http://www.dovecot.org/list/dovecot/2010-August/051946.html) the best option for this? Maybe Ldirectord? Thanks! -- Thiago Henrique www.adminlinux.com.br

2 4

[SOLVED] Re: Trying Dovecot Replication with dsync
by Minhas Listas 19 Aug '14

19 Aug '14

> > 1) > Aug 14 15:39:34 ceph1 dovecot: imap(account=domaintest.com): Error: > open(/var/run/dovecot/replication-notify-fifo) failed: Permission denied > > What is the correct perms for /var/run/dovecot/replication-notify-fifo ? > Currently: > # ll /var/run/dovecot/replication-notify-fifo > prw------- 1 root root 0 Aug 14 15:57 > /var/run/dovecot/replication-notify-fifo| > > Solved with: # chown dovemail:dovemail /var/run/dovecot/replication-notify-fifo dovemail is the user for imap processes. > 2) > Aug 14 16:02:34 ceph1 dovecot: dsync-server(account=domaintest.com): > Error: Couldn't create lock */var/lib/imap/account=domaintest.com > <http://domaintest.com>/*.dovecot-sync.lock: No such file or directory > Aug 14 16:02:34 ceph1 dovecot: dsync-local(account=domaintest.com): > Error: Couldn't create lock */var/lib/imap/account=domaintest.com > <http://domaintest.com>/*.dovecot-sync.lock: No such file or directory > > In my production cluster, it is automatically created by dovecot when the > user performs the first login: > */var/lib/imap/*<account> for INDEX > */var/spool/imap/*<account> for Mdbox > This was fixed with correct permission on /var/run/dovecot/replication-notify-fifo Thanks! -- Thiago Henrique www.adminlinux.com.br

2 1

IMAP on 993/SSL or 143/STARTTLS?
by Alex 18 Aug '14

18 Aug '14

Hi, I have a postfix+dovecot-2.2.13 system and have configured it to support IMAPS on 993 with SSL/TLS. I'm noticing with users using Thunderbird, the autodetect defaults to IMAPS on 143 with STARTTLS. Which is preferred? Which is more secure? Which is more common? Why would someone choose one over the other? Can I ask the same question about SMTP and submission? Why would one choose 587 with STARTTLS versus 465 with SSL/TLS? Thanks, Alex

4 3

Segmentation fault while indexing a large mailbox using doveadm..
by Akash 18 Aug '14

18 Aug '14

Hi, I am having the same issue as: http://www.dovecot.org/list/dovecot/2014-April/095681.html although I am using the latest dovecot-core and dovecot-lucene packages from the debian sid repo which already has the patch (http://hg.dovecot.org/dovecot-2.2/rev/5e42a8c18b51) mentioned in the above thread implemented. doveadm(admin): Info: Old.All: Caching mails seq=1..300332 9400/300332 Program received signal SIGSEGV, Segmentation fault. (gdb) bt #0 __wcslen ( s=s@entry=0xb93887a8 L" btnToggle: \"Change size\"}</script>\r\n<link href=\"http://s26 ucoz net/src/ulightbox/ulightbox css\" rel=\"stylesheet\" type=\"text/css\" /><script type=\"text/javascript\" src=\"http://s26 ucoz net/src/ulightb"...) at wcslen.c:36 #1 0xb7c026fd in wcsdup ( s=0xb93887a8 L" btnToggle: \"Change size\"}</script>\r\n<link href=\"http://s26 ucoz net/src/ulightbox/ulightbox css\" rel=\"stylesheet\" type=\"text/css\" /><script type=\"text/javascript\" src=\"http://s26 ucoz net/src/ulightb"...) at wcsdup.c:29 #2 0xb7a65660 in lucene::document::Field::Field(wchar_t const*, wchar_t const*, int, bool) () from /usr/lib/i386-linux-gnu/libclucene-core.so.1 #3 0xb7b5cd48 in lucene_index_build_more () from /usr/lib/dovecot/modules/lib21_fts_lucene_plugin.so #4 0xb7b5b7b9 in ?? () from /usr/lib/dovecot/modules/lib21_fts_lucene_plugin.so #5 0xb7b6ac2a in fts_backend_update_build_more () from /usr/lib/dovecot/modules/lib20_fts_plugin.so #6 0xb7b6b674 in ?? () from /usr/lib/dovecot/modules/lib20_fts_plugin.so #7 0xb7b6bbf4 in fts_build_mail () from /usr/lib/dovecot/modules/lib20_fts_plugin.so #8 0xb7b713cc in ?? () from /usr/lib/dovecot/modules/lib20_fts_plugin.so #9 0xb7e4969b in mail_precache () from /usr/lib/dovecot/libdovecot-storage.so.0 #10 0xb7fa1c7f in ?? () #11 0xb7f9d133 in _start () (gdb) # dovecot --version 2.2.13 Please help.. -Thanks, Akash

2 2

Dovecot failing with exit code 19200 in cron job
by Murray Trainer 18 Aug '14

18 Aug '14

Hi All, I am synchronizing a few hundred dovecot mailboxes with a perl script via a crontab. It works for a while (ie exit status=0) but the rest then start failing with exit code 19200. I tried redirecting STDERR but showed nothing. I suspect the cron job is hitting some sort of resource limit? If I run the perl script from the command line it appears to dsync all the accounts OK. The relevant part of my Perl code is below. Any ideas? Thanks, Murray foreach (@userlist) { my $email = $_; chomp ($email); # get rid of CR logit("Started sync of mailbox $email") if $debug; my $dsync_command="dsync -v -D -u $email mirror ssh -C root@$mailstore dsync -u $email"; logit("dsync_command=$dsync_command"); # Run dsync of mailbox logit("Started secondary sync of $mailbox to $mailstore") if $debug; my $exit_status = system ( $dsync_command ); logit("Dsync exit status: $exit_status") if $debug; logit("Finished secondary sync of $mailbox to $mailstore") if $debug; }

2 1

NFS INTR mount option
by Murray Trainer 18 Aug '14

18 Aug '14

Hi again, We have our mailboxes on several NFS filesystems mounted with the options below. 10.11.0.238:/mailbox_store_01 on /home1 type nfs4 (rw,noatime,vers=4.0,rsize=65536,wsize=65536,namlen=255,hard,nordirplus,proto=tcp,port=0,timeo=600,retrans=2,sec=sys,clientaddr=10.11.0.95,local_lock=none,addr=10.110.238) On the rare occasions that the NFS mounts hang, the system load goes through the roof and I often have to reboot the mailstore server as dovecot process are hung and can't be killed. Will the NFS INTR mount option help in this situation? Thanks Murray Send EMAIL [1] Links: ------ [1] https://webmail.westnet.com.au/index.php/mail#

1 0

Manage Sieve Security?
by Chris 18 Aug '14

18 Aug '14

Dear All, can I expose Dovecot's Manage Sieve port to the internet or should I restrict access (e.g. by IP address or VPN)? -- Christian

2 1

virtual plugin in HG unaware of namespaces
by Thomas Leuxner 17 Aug '14

17 Aug '14

The latest changes to the virtual plugin seem to have a negative effect on name space rules: $ cat dovecot-virtual * Public/* flagged While flagged messages inside the mailbox still show up, messages flagged in 'Public' are not picked up anymore. Regards Thomas

1 1

INBOX subfolders not showing up
by Matthias Sitte 16 Aug '14

16 Aug '14

Hi, I have a problem with INBOX subfolders and subscriptions which I can't solve... Maybe somebody has seen this, too... I can create regular subfolders (not in INBOX) and (un-)subscribe to them with Thunderbird. If I create subfolders in the INBOX with TB this seems to work now in v31 (just found this out), but with v24.6 neither INBOX subfolder creation nor subscriptions worked. When I modified the `subscriptions' file manually on the server, the folders would show up in TB, though. Maybe this was a TB bug, but quite similarly INBOX subfolders don't show up on iPhones (iOS up-to-date), but all other subfolders do. Any ideas what's going on? Best, Matthias ====================================================================== # dovecot -n # 2.1.7: /etc/dovecot/dovecot.conf # OS: Linux 3.2.0-4-amd64 x86_64 Debian 7.6 ext4 hostname = familie-sitte.org log_path = /var/log/dovecot.log log_timestamp = "%Y-%m-%d %H:%M:%S " mail_location = maildir:/srv/mail/%d/%n/Maildir:LAYOUT=fs 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 { inbox = yes list = 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 = / subscriptions = yes type = private } passdb { args = /etc/dovecot/dovecot-sql.conf.ext driver = sql } plugin { quota = maildir:User quota quota_warning = storage=95%% quota-warning 95 %u quota_warning2 = storage=80%% quota-warning 80 %u sieve = ~/.dovecot.sieve sieve_dir = ~/sieve } postmaster_address = postmaster(a)familie-sitte.org protocols = " imap sieve pop3" service auth { unix_listener /var/spool/postfix/private/auth { group = postfix mode = 0660 user = postfix } } service imap-login { inet_listener imaps { port = 993 ssl = yes } } service managesieve-login { inet_listener sieve { port = 4190 } } service quota-warning { executable = script /usr/local/bin/quota-warning.sh unix_listener quota-warning { user = vmail } user = dovecot } ssl_cert = </etc/dovecot/ssl/mail_cert.pem ssl_cipher_list = SSDH+ECDSA+AESGCM:EECDH+aRSA+AESGCM:EECDH+ECDSA+SHA256:EECDH+aRSA+RC4:EDH+aRSA:EECDH:RC4:!aNULL:!eNULL:!LOW:!3DES:!MD5:!EXP:!PSK:!SRP:!DSS ssl_key = </etc/dovecot/ssl/mail_key.pem userdb { args = uid=vmail gid=vmail home=/srv/mail/%d/%n driver = static } protocol lda { mail_plugins = " quota sieve" } protocol imap { mail_plugins = " imap_quota quota" }

3 4

Server dovecot: auth: Warning: sql: Ignoring changed iterate_quer
by Joseph Malloy 16 Aug '14

16 Aug '14

Hello, I am having an issue with Dovecot. I am using Roundcube to access my mailboxes, which I have set up using ViMbAdmin. However when I go to the Roundcube, I get login Failed. So looking in the mail.log I can see the following: Aug 15 13:20:09 Server dovecot: auth: Warning: sql: Ignoring changed iterate_query in /etc/dovecot/dovecot-sql.conf.ext, because userdb sql not used. (If this is intentional, set userdb_warning_disable=yes) Aug 15 13:20:11 Server dovecot: imap-login: Disconnected (auth failed, 1 attempts in 2 secs): user=<joe(a)anoblequest.co.uk>, method=PLAIN, rip=62.49.15.89, lip=192.168.0.90, TLS, session=<su7WCqoALAA+MQ9Z> I am most greatful for the support Joe :)

2 1

namspace management
by A. Schulze 15 Aug '14

15 Aug '14

Hello, since some weeks I'm playing with namespaces. But I still did nod found a solution for all faces of different problems. Current issue: Public namespace. I have users inbox as follow: mail_home = /data/mail/%Ln/ mail_location = maildir:~/Maildir:INDEX=~/.dovecot.index namespace { inbox = yes separator = / prefix = INBOX/ } Public namespace: namespace { location = maildir:/data/mail_public/Maildir/:INDEXPVT=~/.dovecot.index.public/ prefix = PUBLIC/ separator = / type = public } that way only one subscription file is possible. Any use could subscribe/unsubscribe any folder and that match all users. If I add "subscription=no" the parent namespace' subscription is used. In fact there is no parent namespace an no subscription management is possible at all. On the other side I tried to change the PUBLIC namespace prefix to "INBOX/PUBLIC" but now the namespace is *below* the inbox. That's also not what I want. How do other users organise the namespaces INBOX, PUBLIC, SHARED and VIRTUAL (via virtual plugin) ? Thanks Andreas

3 4

(Again) Some trouble with dovecot 2.1(.7) and SASL
by admin 15 Aug '14

15 Aug '14

The parentheses in the subject are due to referral to this discussion on the mailing list: http://dovecot.org/list/dovecot/2012-February/063851.html At first I thought I found the problem, but then realized this was 2012 and the suggested patch already is applied in dovecot. I also have segault messages in my kernel log; not too many, but it was enough to dig for it: > Wed Jun 25 08:29:05 2014 auth[17590]: segfault at 10 ip 00007f02ad5e8e56 sp 00007fff932f9250 error 4 in auth[7f02ad5d3000+40000] > Mon Jun 30 16:11:41 2014 auth[12501]: segfault at 10 ip 00007fc6a1fefe56 sp 00007fff276e6ea0 error 4 in auth[7fc6a1fda000+40000] > Sat Jul 5 13:11:32 2014 auth[15603]: segfault at 10 ip 00007f6447a6be56 sp 00007fff09c48020 error 4 in auth[7f6447a56000+40000] > Tue Jul 8 05:08:21 2014 auth[30071]: segfault at 10 ip 00007ff22656ae56 sp 00007fff26b017f0 error 4 in auth[7ff226555000+40000] > Mon Jul 21 14:25:28 2014 auth[30918]: segfault at 10 ip 00007fe1fc817e56 sp 00007fff7e654f20 error 4 in auth[7fe1fc802000+40000] <--- > Wed Aug 13 18:56:31 2014 auth[27593]: segfault at 10 ip 00007ffbf8ce8e56 sp 00007fffe0157f70 error 4 in auth[7ffbf8cd3000+40000] > Wed Aug 13 18:57:59 2014 auth[24264]: segfault at 10 ip 00007f2df31bde56 sp 00007fff094263c0 error 4 in auth[7f2df31a8000+40000] The last 2 entries are a result of me being able to reproduce the problem, the 3rd last is the only one I was able to track down (due to log rotation) I tracked the problem down to a contact form of a website of one of my customers. whenever used, the segfault gets triggered. Nevertheless, the message always gets delivered. So we are talking about the marked line; the corresponding log entries from mail.info: > Jul 21 14:25:29 mail postfix/smtpd[27590]: connect from unknown[<webserver ip>] > Jul 21 14:25:29 mail dovecot: auth: Fatal: master: service(auth): child 30918 killed with signal 11 (core dumps disabled) > Jul 21 14:25:29 mail postfix/smtpd[27590]: warning: unknown[<webserver ip>]: SASL DIGEST-MD5 authentication failed: Connection lost to authentication server > Jul 21 14:25:29 mail dovecot: auth-worker(27595): mysql(127.0.0.1): Connected to database mail > Jul 21 14:25:29 mail postfix/smtpd[27590]: D4D8E44752: client=unknown[<webserver ip>], sasl_method=CRAM-MD5, sasl_username=<existing account and working account> > Jul 21 14:25:29 mail postfix/cleanup[27598]: D4D8E44752: message-id=<id(a)doma.in> > Jul 21 14:25:29 mail postfix/qmgr[13349]: D4D8E44752: from=<server(a)doma.in>, size=2680, nrcpt=1 (queue active) > Jul 21 14:25:29 mail postfix/smtpd[27590]: disconnect from unknown[<webserver ip>] > Jul 21 14:25:29 mail dovecot: lmtp(27600): Connect from local > Jul 21 14:25:30 mail postfix/lmtp[27599]: D4D8E44752: to=<info(a)doma.in>, orig_to=<contact(a)doma.in>, relay=<mailserver fqdn>[private/dovecot-lmtp], delay=0.19, delays=0.07/0.04/0.04/0.04, dsn=2.0.0, status=sent (250 2.0.0 <info(a)doma.in> MrY+OdkizVPQawAAMrkmwA Saved) > Jul 21 14:25:30 mail postfix/qmgr[13349]: D4D8E44752: removed > Jul 21 14:25:30 mail dovecot: lmtp(27600, info(a)doma.in) MrY+OdkizVPQawAAMrkmwA: msgid=<id(a)doma.in>: saved mail to INBOX > Jul 21 14:25:30 mail dovecot: lmtp(27600): Disconnect from local: Client quit (in reset) contact(a)doma.in is a virtual address which gets forwarded to info(a)doma.in (/etc/postfix/virtual). The used CMS is PyroCMS (v2.2.3) - the version is not up2date, but nevertheless the CMS might be the one to blame, a service should not segfault. Some packages are to be updated, as well as the CMS; but I am willing to wait to provide information, which leads to the problem, as I cannot switch between different versions on my productive system for the sake of gaining debug information. I hope that someone has any kind of clue and is willing to resolve that issue with me, asking me for the necessary information to do so. Thanks in advance!

2 1