September 2012 - dovecot

[Dovecot] Changing Quota downward for special Mailboxes? (2.1.6)
by Jost Krieger 27 Sep '12

27 Sep '12

It seems there is no way to correct te quota limit downward for some mailboxes, like: quota_rule9 = UCE-TMP:storage=-10%% The idea is to not eat up the users storage with spam, so that spam stays in the queue (for some time) while real mail gets delivered. There *is* a line } else if (*value != '-' && relative_rule) { in quota.c, but nothing else handles it, starting with quota_rule_parse_percentage. Yours Jost Krieger -- | Jost.Krieger+sig(a)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 |

3 8

[Dovecot] 76Gb to 146Gb
by Spyros Tsiolis 27 Sep '12

27 Sep '12

Hello all, I have a DL360 G4 1U server that does a wonderfull job with dovecot horde, Xmail and OpenLDAP for a company and serving about 40 acouunts. The machine is wonderful. I am very happy with it. However, I am running out of disk space. It has two times 76Gb Drives in RAID1 (disk mirroring) and the capacity has reached 82%. I am starting of getting nervous. Does anyone know of a painless way to migrate the entire contents directly to another pair of 146Gb SCSI RAID1 disks ? I thought of downtime and using clonezilla, but my last experience with it was questionable. I remember having problems declaring disk re-sizing from the smaller capacity drives to the larger ones. CentOS 5.5 Manual install of : Mysql XMail (pop3/smtp) ASSP (anti spam) Apache / LAMP and last but by no means list : Dovecot Dovecot -n : # 1.2.16: /etc/dovecot/dovecot.conf # OS: Linux 2.6.18-194.17.4.el5 i686 CentOS release 5.5 (Final) ext3 base_dir: /var/run/dovecot/ log_path: /var/log/dovecot/dovecot.log info_log_path: /var/log/dovecot/dovecot-info.log ssl_parameters_regenerate: 48 verbose_ssl: yes login_dir: /var/run/dovecot//login login_executable: /usr/local/dovecot/libexec/dovecot/imap-login login_greeting: * Dovecot ready * login_max_processes_count: 96 mail_location: maildir:/var/MailRoot/domains/%d/%n/Maildir mail_plugins: zlib auth default: verbose: yes debug: yes debug_passwords: yes passdb: driver: passwd-file args: /etc/dovecot/passwd passdb: driver: pam userdb: driver: static args: uid=vmail gid=vmail home=/home/vmail/%u userdb: driver: passwd Any help would be appreciated or any ideas you might have. Regards, spyros ---- "I merely function as a channel that filters music through the chaos of noise" - Vangelis

10 18

[Dovecot] second dovecot imap server
by Spyros Tsiolis 26 Sep '12

26 Sep '12

Hello again all, I would like to thank everyone for rushing to help me with the dovecot migration issue. I am , currently, reading and studying what has been said by all the good and kind people who responded to this list. OK, on the same issue,I thought of adding a totally new server with dovecot acting as a secondary/backup server for the mail servicing. Would that be possible ? What if I manage to make it work; Can I retire the old(er) server ? I would think that this is more on-topic than my last post. If I am out of line again, please respond in private. Thank you all! spyros ---- "I merely function as a channel that filters music through the chaos of noise" - Vangelis

1 0

[Dovecot] duplicate all emails for 'backup' after user deletes them?
by Voytek Eymont 26 Sep '12

26 Sep '12

I have a user who occasionally reads, then deletes emails using his phone, on his and others email accounts, causing emails deleted on mail server, then he says 'oops, can you recover emails I've deleted?' what's the best way to duplicate all emails to duplicate accounts ? backup folders ? so, next time he deletes emails they can be 'recovered' on the server ? I was thinking along the lines of duplicating ALL emails...? any tips and pointers appreciated -- Voytek

5 5

[Dovecot] mail size needed in dovecot sieve log line
by Maciej Uhlig 26 Sep '12

26 Sep '12

This is a dovecot sieve log line: 2012-09-26T13:47:52+02:00 host/192.168.1.1 dovecot: [ID 583609 mail.info] lmtp(7565, user@domain): vTtJDTrrYlCNHQAAs1WjhA: sieve: msgid=<MSGID>: stored mail into mailbox 'INBOX' Would it be possible to have mail size logged here? MU

3 3

Re: [Dovecot] 76Gb to 146Gb
by Stan Hoeppner 26 Sep '12

26 Sep '12

On 9/25/2012 12:29 AM, Spyros Tsiolis wrote: > My clients don't invest in hardware. I live in Greece. Things are really > bad right now. Not that if they were better they (the client) would invest > in their infrastructure. ... > Thank you very much Stan for you kind reply. > > I reply to you in private since I was told to stop the thread > on the dovecot list because it's off-topic. I'm going to bring it back on topic. To recap the issue: 1. 73GB volume, ~82% full. Volume contains OS, spool, and mailboxes 2. Two 146GB drives on hand that are impossible to employ in either a live or offline migration due to hardware limitations--drive slots (0.82*73GB=60GB)-5GB(OS)= ~55GB/40= ~1.375GB/user mailbox (This math assumes the box has already been scoured and things like source files and other temporary stuff has already been removed, along with anything and everything not necessary to the primary functions of this box) 1.375GB/user suggests these users seem to horde old emails, or simply don't keep their mailboxes tidy. Normally I'd guess that a lot of space in a situation like this is taken up by full text search indexes, but you don't appear to have these enabled. I notice in your 'dovecot -n' output that you are already employing zlib, but to what degree is currently unknown to the list. Are you compressing maildirs with a nightly script? If not, and you're compressing on an ad hoc basis, realtime compression during write could save you quite a bit of additional space, extending the operating life of your current drives. Of course a simple request to your users to clean up their mailboxes would surely help as well, if that's a possibility--some users would flog one for making such a suggestion. If you're not fully taking advantage of compression currently, you may want to consider upgrading to Dovecot 2.x which allows transparent compression to dbox and maildir storage. You may want to migrate to mdbox message store format while still using zlib, and enable single instance attachment storage (SIS). As the name suggests, SIS saves an attachment only once, regardless how many user emails in which it exists. See: http://wiki2.dovecot.org/Plugins/Zlib I'd point you to the docs for SIS but I'm unable to find them in the Wiki. Surely someone else can provide the settings and/or update the wiki. -- Stan

1 0

[Dovecot] Can`t get over 1024 processes on FreeBSD - possible bug?
by Tomáš Randa 24 Sep '12

24 Sep '12

Hello, I still cannot get dovecot running with more then 1000 processes, but hard limit is 8192 per user in box. I tried everything, including modifying startup script of dovecot to set ulimit -u 8192. Could it be some dovecot bug or dovecot<>freebsd bug? I also tried to set client_limit=2 in imap service to spawn more imap clients in one process, but still I am over 1000 processes with kernel message: maxproc limit exceeded by uid 89 Could anybody help? Many thanks Tomas My system have following settings: FreeBSD 9.0 / AMD64 Dovecot 2.1.8 kern.maxproc: 12288 kern.maxfilesperproc: 36864 kern.maxprocperuid: 8192 no limit for uid 89: # limit cputime unlimited filesize unlimited datasize 33554432 kbytes stacksize 524288 kbytes coredumpsize unlimited memoryuse unlimited vmemoryuse unlimited descriptors 36864 memorylocked unlimited maxproc 8192 sbsize unlimited swapsize unlimited My dovecot.conf: # 2.1.8: /usr/local/etc/dovecot/dovecot.conf # OS: FreeBSD 9.0-STABLE amd64 auth_mechanisms = plain login digest-md5 cram-md5 default_client_limit = 2048 default_process_limit = 2048 disable_plaintext_auth = no first_valid_gid = 89 first_valid_uid = 89 info_log_path = /data/logfiles/dovecot/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_plugins = " quota" mail_privileged_group = mail maildir_broken_filename_sizes = yes passdb { driver = vpopmail } plugin { quota = maildir quota_rule = Trash:ignore sieve = ~/.sieve/dovecot.sieve sieve_dir = ~/.sieve } protocols = imap pop3 sieve service anvil { client_limit = 6147 } service auth { client_limit = 8192 unix_listener auth-client { mode = 0660 } unix_listener auth-master { mode = 0600 } } service imap-login { process_limit = 2048 service_count = 1 } service imap { client_limit = 1 process_limit = 2048 } service managesieve { process_limit = 2048 } service pop3-login { process_limit = 2048 service_count = 1 } service pop3 { client_limit = 1 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 = 2048 mail_plugins = quota imap_quota } protocol lda { mail_plugins = quota sieve sendmail_path = /var/qmail/bin/sendmail } protocol sieve { managesieve_implementation_string = dovecot managesieve_max_line_length = 65536 }

5 4

[Dovecot] Traffic Accounting
by M. Naumann 24 Sep '12

24 Sep '12

Hi, I'm trying to find out how to do traffic accounting with Dovecot 2.x, preferrably v2.0.9, preferrably on CentOS 6. I've previously asked on IRC, but there was little feedback, and my understanding is now this list is the preferred media for such inquiries. If I recall correctly, some weeks ago I was told that traffic accounting is not officially supported on Dovecot 2, but that there could still be ways to get it to work, but no details were provided. I can think of the following approaches: * rawlog, preferrably piped (if that's possible?) into something like wc to prevent privacy issues and to reduce the I/O overhead * maildrop filtering in front of dovecot LDA (for mail inbound to mail storage) * sieve filtering Unfortunately I have little experience with either so far, so it's hard to make a good choice. I would appreciate hints on these approaches, and on any other approaches you can think of, as well as any related documentation / how-to you could point me to. While I'm subscribed to the list (for mail authentication purposes), I've disabled receiving any e-mail form the list, so please CC me on any replies. Thanks in advance, Moritz

2 1

[Dovecot] Patches and dovecot releases
by Jean Michel 24 Sep '12

24 Sep '12

I'd like to know if after a release, for example the recently release 2.1.10, its common to see few days from the release some bug reports and algo some patches, are theses patches applyed on the daily builds ? ------------------------------ Jean Michel Feltrin

2 1

[Dovecot] Logging question regarding delete actions
by Ralf Hildebrandt 24 Sep '12

24 Sep '12

A user is logged in via imap from multiple devices. The log has this: Sep 21 11:46:32 postamt dovecot: imap(awxxxxer): delete: box=INBOX, uid=15347, msgid=<1341851741.4ffb085d2e2b7(a)swift.generated>, size=15675 Sep 21 11:46:32 postamt dovecot: imap(awxxxxer): delete: box=INBOX, uid=15739, msgid=<b23b2e42f6ae9ba1602690be42b7b5c7.squirrel(a)webmail.charite.de>, size=18134 Sep 21 11:46:32 postamt dovecot: imap(awxxxxer): delete: box=INBOX, uid=15740, msgid=<3509D8B3F9054C4BB26C85F3E4B96563036822877538(a)EXCHANGE21.charite.de>, size=22226 Sep 21 11:46:32 postamt dovecot: imap(awxxxxer): delete: box=INBOX, uid=15741, msgid=<C2E306F4-8622-4D0F-98BD-C67F31F2FF08(a)charite.de>, size=5160 How can I find out WHICH CLIENT caused the deletion? Same issue: Sep 21 09:36:05 postamt dovecot: imap-login: Login: user=<awxxxxer>, method=PLAIN, rip=109.45.0.37, lip=141.42.206.36, mpid=30773, TLS, session=<e8TfSjHKgwBtLQAl> Sep 21 10:06:17 postamt dovecot: imap(awxxxxer): Disconnected for inactivity in=2255 out=4398 How can I be sure that the log entry from 10:06:17 "belongs" to the log entry from "09:36:05"? Also, what is the meaning of the "session=<e8TfSjHKgwBtLQAl>"? -- Ralf Hildebrandt Geschäftsbereich IT | Abteilung Netzwerk Charité - Universitätsmedizin Berlin Campus Benjamin Franklin Hindenburgdamm 30 | D-12203 Berlin Tel. +49 30 450 570 155 | Fax: +49 30 450 570 962 ralf.hildebrandt(a)charite.de | http://www.charite.de

2 2