- dovecot - dovecot.org

[Dovecot] Released Pigeonhole v0.3.3 for Dovecot v2.1.10
by Stephan Bosch 18 Sep '12

18 Sep '12

Hello Dovecot users, Version 0.3.2 was a bit broken, because it failed to compile against installed Dovecot headers. Changelog v0.3.3: - Fixed compile against installed Dovecot headers. This was broken by the ld.gold fix in the previous release. The release is available as follows: http://www.rename-it.nl/dovecot/2.1/dovecot-2.1-pigeonhole-0.3.3.tar.gz http://www.rename-it.nl/dovecot/2.1/dovecot-2.1-pigeonhole-0.3.3.tar.gz.sig Refer to http://pigeonhole.dovecot.org and the Dovecot v2.x wiki for more information. Have fun testing this new release and don't hesitate to notify me when there are any problems. Regards, -- Stephan Bosch stephan(a)rename-it.nl

1 0

[Dovecot] lda bug: multiple passdb ignored
by Алексей Панкратов 18 Sep '12

18 Sep '12

Greetings. When dovecot configured to use multiple passdb and static userdb as: #password database in multiple ldap subtree passdb { driver = ldap args = /usr/local/etc/dovecot/dv-ldap-real.conf } passdb { driver = ldap args = /usr/local/etc/dovecot/dv-ldap-virtual.conf } userdb { driver = static args = uid=pop gid=mailnull } all works fine but... When LDA deliver mail it check passdb but only first appeared in configuration: 18.09.2012 15:00:18 auth: Debug: master in: USER 1 info(a)zzz.zz service=lda 18.09.2012 15:00:18 auth: Debug: cache(info(a)zzz.zz): miss 18.09.2012 15:00:18 auth: Debug: ldap(info(a)zzz.zz): pass search: base=ou=account,ou=it,dc=planar,dc=local scope=subtree filter=(&(objectClass=x-mailUser)(uid=info)) fields=x-mailPassword 18.09.2012 15:00:18 lda: Debug: auth input: 18.09.2012 15:00:18 auth: Debug: cache(info(a)planar.spb.ru): hit: 18.09.2012 15:00:18 auth: Debug: master out: NOTFOUND 1 1 P.S. dovecot version 2.1.7 sysytem x86 FreeBSD 9.0

3 4

[Dovecot] Proxy connection timeouts
by Kelsey Cummings 18 Sep '12

18 Sep '12

We are seeing a few (0-15) proxy failures like the following out of ~3m successful proxied connections a day. Average session creation load over our peak hour is about 47/sec. The backend servers aren't logging anything that would suggest any internal problem like insufficient processes to handle the load. It doesn't seem to happen when utilization is lowest at night. dovecot: imap-login: Error: proxy(foo): connect(1.1.1.1, 143) failed: Connection timed out (after 63 secs) I'm curious if anyone else has seen any similar problems or has any suggestions. # dovecot -n # 2.1.9: /etc/dovecot/dovecot.conf # OS: Linux 2.6.32-279.5.2.el6.x86_64 x86_64 Scientific Linux release 6.3 (Carbon) auth_master_user_separator = * auth_username_format = %Ln auth_verbose = yes auth_verbose_passwords = sha1 auth_worker_max_count = 64 mail_fsync = always mail_log_prefix = "%s(%u): session=%{session} " mail_plugins = stats zlib maildir_very_dirty_syncs = yes mmap_disable = yes passdb { args = /etc/dovecot/master-users driver = passwd-file master = yes } passdb { args = imap driver = pam } plugin { lazy_expunge = DELETED_MESSAGES. stats_refresh = 30 secs stats_track_cmds = yes } protocols = imap pop3 service anvil { client_limit = 10000 } service auth { client_limit = 10000 vsz_limit = 512 M } service doveadm { inet_listener { port = 1842 } unix_listener doveadm-server { mode = 0666 } } service imap-login { inet_listener imap { port = 143 } inet_listener imaps { port = 993 ssl = yes } process_limit = 7000 process_min_avail = 32 } service imap-postlogin { executable = script-login -d /etc/dovecot/bin/sonic-imap-postlogin user = $default_internal_user } service imap { executable = imap imap-postlogin process_limit = 4096 vsz_limit = 512 M } service pop3-login { inet_listener pop3 { port = 110 } inet_listener pop3s { port = 995 ssl = yes } process_limit = 2000 process_min_avail = 32 } service pop3-postlogin { executable = script-login -d /etc/dovecot/bin/sonic-pop3-postlogin user = $default_internal_user } service pop3 { executable = pop3 pop3-postlogin process_limit = 4096 } service stats { fifo_listener stats-mail { mode = 0666 } } shutdown_clients = no ssl = required ssl_ca = </etc/dovecot/ssl/gd_bundle.crt ssl_cert = </etc/dovecot/ssl/imap.sonic.net.crt ssl_key = </etc/dovecot/ssl/imap.sonic.net.key ssl_parameters_regenerate = 1 days syslog_facility = local0 userdb { driver = passwd } verbose_proctitle = yes protocol imap { imap_id_send = support-url support-email mail_max_userip_connections = 20 mail_plugins = stats zlib mwi_update mail_log notify imap_stats imap_zlib ssl_ca = </etc/dovecot/ssl/gd_bundle.crt ssl_cert = </etc/dovecot/ssl/imap.sonic.net.crt ssl_key = </etc/dovecot/ssl/imap.sonic.net.key } protocol pop3 { mail_plugins = stats zlib lazy_expunge pop3_fast_size_lookups = yes pop3_uidl_format = %f ssl_ca = </etc/dovecot/ssl/pop.sonic.net.ca-bundle ssl_cert = </etc/dovecot/ssl/pop.sonic.net.crt ssl_key = </etc/dovecot/ssl/pop.sonic.net.key }

2 1

[Dovecot] Mailbox aliases
by Timo Sirainen 18 Sep '12

18 Sep '12

I've been wondering about ways to handle all the various different Sent/Sent Messages/Sent Items/etc mailboxes that different clients create and if there could be a way to make them work at least somewhat better. I'm mainly thinking about ISP-like installations where users are using tons of different clients and there's no way to get everyone configured properly. Of course the eventual solution is hopefully that all the clients will support SPECIAL-USE extension, but while waiting for that maybe mailbox aliases could help some. So one possibility would be to force autocreation of wanted specific mailbox names, and create aliases for the other commonly used mailboxes. The main difference here for the current behavior is that all of the mailboxes would contain the same mails, instead of different mails going to different mailboxes with different clients. So for example: .. mailbox Sent { auto = create special_use = \Sent } mailbox "Sent Messages" { special_use = \Sent alias_for = Sent } .. If "Sent Messages" doesn't exist: - DELETE Sent works as usual - RENAME Sent works as usual If "Sent Messages" is created: - it's symlinked to Sent - DELETE "Sent messages" deletes the symlink, not the contents - RENAME "Sent messages" isn't allowed - DELETE/RENAME Sent fails with message: Delete/Rename for Sent not allowed before "Sent Messages" is deleted. Thoughts? The main disadvantage is that clients would still see two different mailboxes, and some would download contents from both of them. I was thinking about creating some wiki page which could list the best working configurations and also a client list, which would specify if it supports SPECIAL-USE and what the default mailbox names it uses are.

4 4

[Dovecot] Released Pigeonhole v0.3.2 for Dovecot v2.1.9
by Stephan Bosch 18 Sep '12

18 Sep '12

Hello Dovecot users, I've been busy recently with some wholly unrelated Dovecot IMAP patches, so Sieve development moved a bit into the background. This release therefore consists mostly of small bugfixes. The most important change is the fix of a bug in ManageSieve's CHECKSCRIPT command that causes a segfault. Quite a few people have encountered this problem recently, so it is about time I release this now. Changelog v0.3.2: + sieve-refilter tool: improved man page documentation by explicitly specifying the syntax used for mailbox arguments. + Sieve: spamtest and virustest extensions: improved trace debugging of score calculation. + Sieve: made error messages about exceeding the maximum number of actions more verbose. - Sieve tools: fixed problems with running as root: sievec and sieve-dump now ignore mail_uid and mail_gid settings when run as root. - Sieve: fixed bug in action accounting (for limit checking): increase action instance count only when an action is actually created. - Sieve: include extension: fixed namespace separation of :global and :personal scripts. - ManageSieve: fixed segfault bug triggered by CHECKSCRIPT command. - Fixed linking with ld.gold. - Fixed several Clang compile warnings and a few potential bugs. The release is available as follows: http://www.rename-it.nl/dovecot/2.1/dovecot-2.1-pigeonhole-0.3.2.tar.gz http://www.rename-it.nl/dovecot/2.1/dovecot-2.1-pigeonhole-0.3.2.tar.gz.sig Refer to http://pigeonhole.dovecot.org and the Dovecot v2.x wiki for more information. Have fun testing this new release and don't hesitate to notify me when there are any problems. Regards, -- Stephan Bosch stephan(a)rename-it.nl

1 0

[Dovecot] trash plugin together with sieve_before ?
by Jan-Frode Myklebust 18 Sep '12

18 Sep '12

We have a sieve script doing sieve_before to sort spam to spam-folders. Now I'm trying to configure the "Trash" plugin, but it doesn't seem to work.. I noticed my config file says: # Space separated list of plugins to load (none known to be useful so far). Do NOT # try to load IMAP plugins here. #mail_plugins = and that doveconf doesn't list any plugins loaded for "protocol sieve". Should we load quota and trash here ? -jf

2 1

[Dovecot] Shared Mailboxes in a multi domain environment
by Alessio Cecchi 18 Sep '12

18 Sep '12

Hi, I would like to enable mailbox sharing for my users. I'm using dovecot 2.1.9 with vpopmail authentication. For security reason shared mailboxes should be available only for users of the same domain. Dovecot's wiki suggest to enable acl like this: plugin { # assumes mailboxes are in /var/mail/%d/%n: acl_shared_dict = file:/var/mail/%d/shared-mailboxes.db } but in vpopmail, and also in many others configuration, mailboxes are, for example, in /var/mail/nas1/%d, /var/mail/nas2/%d and so on, so it's difficult to restrict shared mailboxes available only for the same domain. A trick could be set acl_shared_dict to /var/mail/%d-shared-mailboxes.db, but I don't know if it's safe on an NFS environment. So, why don't provide a way to restrict shared mailboxes also for dict in SQL? One way could be to add to "user_shares" table a column "domains" which is the same domain of the mailbox'owner and a config option, for acl, like acl_only_for_same_domain =yes/no so dovecot can add a "WHERE %n = domain" for the SELECT. Are there other possible solutions? Thanks -- Alessio Cecchi is: @ ILS -> http://www.linux.it/~alessice/ on LinkedIn -> http://www.linkedin.com/in/alessice Assistenza Sistemi GNU/Linux -> http://www.cecchi.biz/ @ PLUG -> ex-Presidente, adesso senatore a vita, http://www.prato.linux.it

2 2

[Dovecot] dovecot 1.x: problem with quota
by Toni Mueller 17 Sep '12

17 Sep '12

Hi, I have now verified that there is a problem with quota (and sieve?). Summary: If the "maildirsize" file gets garbled (= every once in a while), mail is getting lost randomly. Details: I have a Debian box (6.0.5/amd64) with a Postfix server (2.7.1), which delivers email via Dovecot (1.2.15) to local disk (ext3). My email is filtered via a sieve script (wish I could have something better, eg. procmail). The account data is fetched from an SQL database, and I have a default quota set in dovecot.conf (200MB). My own mailbox has no quota (*:storage=0, mentioned in the database). A client sent an email to me, which is logged as being delivered like normal to my mailbox, but doesn't show up there. After poking around, I decided to simply delete the maildirsize file and have the client send that email again, and voila, it shows up in my mailbox. At the same time, however, other emails get delivered just fine, so the heinous thing is that you don't immediately notice that something fishy is going on - only when the client actually calls you (sort of defeats some uses of email)... The problem is worsened by the fact that you can't tell *which* emails get lost - you only end up with a random subset of "your" messages. The problem occurred now after running for a few months. FWIW, the problem with the garbled maildirsize file resulted in a different user's email getting bounced due to "mailbox full", when it wasn't. Questions: * Is this a known problem (Google was unhelpful/unwieldy)? * How can I prevent the problem, short of setting up a cron job to delete that file every week? * How can I log the delivery of messages in dovecot's log, not only the POP3/IMAP activity when people fetch their email? TIA! Kind regards, --Toni++

1 1

[Dovecot] Can`t get over 1024 processes
by Tomáš Randa 17 Sep '12

17 Sep '12

Hello, I have configuration question. Running FreeBSD 9.0 / AMD64 / Dovecot 2.1.8 version and cannot get over 1024 processes. I`ve increased process_limit in imap and pop3 and still have messages in syslog like: maxproc limit exceeded by uid 89, please see tuning(7) and login.conf(5). Did I forget to increase some other value? System values are higher - kern.maxprocperuid: 8192 and kern.maxproc: 12288 I`d like to have more secure configuration > one process to one connection. Thanks for help. Tomas My config: # OS: FreeBSD 9.0-STABLE amd64 auth_mechanisms = plain login digest-md5 cram-md5 default_client_limit = 6200 default_process_limit = 4096 disable_plaintext_auth = no first_valid_gid = 89 first_valid_uid = 89 info_log_path = /var/log/dovecot-info.log last_valid_gid = 89 last_valid_uid = 89 listen = * log_timestamp = "%Y-%m-%d %H:%M:%S " login_greeting = Mail Toaster (Dovecot) ready. mail_location = maildir:~/Maildir mail_privileged_group = mail maildir_broken_filename_sizes = yes passdb { driver = vpopmail } plugin { quota = maildir quota_rule = Trash:ignore } protocols = imap pop3 service auth { unix_listener auth-client { mode = 0660 } unix_listener auth-master { mode = 0600 } } service imap-login { process_limit = 1536 service_count = 1 } service imap { process_limit = 2048 } service pop3-login { service_count = 1 } service pop3 { process_limit = 2048 } shutdown_clients = no ssl_cert = </var/qmail/control/servercert.pem ssl_key = </var/qmail/control/servercert.pem userdb { args = blocking=yes driver = passwd } userdb { driver = vpopmail } verbose_proctitle = yes protocol imap { imap_client_workarounds = delay-newmail tb-extra-mailbox-sep mail_max_userip_connections = 1024 mail_plugins = quota imap_quota } protocol lda { mail_plugins = quota imap_quota sendmail_path = /var/qmail/bin/sendmail } -- Tomáš Randa, Blueboard.cz s.r.o. ---------------------------------- http://hosting.blueboard.cz GSM: +420 775 086 575 Jabber: tomas(a)blueboard.cz Tel: +420 245 008 678 -- Tomáš Randa, Blueboard.cz s.r.o. ---------------------------------- http://hosting.blueboard.cz GSM: +420 775 086 575 Jabber: tomas(a)blueboard.cz Tel: +420 245 008 678

3 9

[Dovecot] clucene build warnings?
by Charles Marcus 17 Sep '12

17 Sep '12

Hi all/Timo, Is clucene still the best/recommended method for enabling full text search? Although I haven't enabled FTS yet, I plan on doing sometime in the future, and was going to be using clucene, but before I do, I'd like to see if I need to worry about an issue I've been seeing ever since I first installed clucene (and still get every time it is upgraded)... I run gentoo linux on an amd64 server, and have always gotten this warning whenever it is updated (and also got it when I first installed it): QA Notice: Package triggers severe warnings which indicate that it may exhibit random runtime failures. /var/tmp/portage/dev-cpp/clucene-2.3.3.4-r4/work/clucene-core-2.3.3.4/src/core/CLucene/index/DocumentsWriter.cpp:129:33: warning: passing NULL to non-pointer argument 2 of ‘void* memset(void*, int, size_t)’ Please do not file a Gentoo bug and instead report the above QA issues directly to the upstream developers of this software. Homepage:http://clucene.sourceforge.net/ Has anyone else ever seen similar warnings when building clucene? Is the gentoo ebuild output correct that this is not a gentoo bug, but is in fact a problem with upstream? If so, is it harmless? Or should I try to get upstream to fix it? That said, being that ianap, I probably wouldn't be much help to them, so, Timo, any chance you can see if this is a problem that needs to be addressed with them? Thanks, -- Best regards, Charles

3 10