- dovecot - dovecot.org

[Dovecot] Dovecot POP3 problem: Authentication suddenly fails until dovecot gets restarted
by Carsten Buchenau 01 Dec '08

01 Dec '08

Hi, I am running dovecot with postfix and OTRS2 on a fresh Ubuntu 8.10 Server system. OTRS2 is configured to poll new email to otrs-incoming@localhost via POP3 every 5 minutes. It works, but after approximately 20 hours it suddenly stops working and only a restart of dovecot helps. The outage starts at different times during night (00:15 on Thursday, 1:15 on Friday, 3:20 on Saturday). I checked all cron-jobs, but couldn't find anything matching. Anyone any idea? Thanks, carsten Mail.log: Nov 29 07:10:13 wombat dovecot: auth-worker(default): pam(otrs-incoming,127.0.0.1): pam_start() failed: Critical error - immediate abort Nov 29 07:10:14 wombat dovecot: pop3-login: Aborted login (auth failed, 1 attempts): user=<otrs-incoming>, method=PLAIN, rip=127.0.0.1, lip=127.0.0.1, secured Dovecot.conf: protocols = pop3 pop3s log_timestamp = "%Y-%m-%d %H:%M:%S " mail_location = mbox:~/mail:INBOX=/var/mail/%u mail_privileged_group = mail protocol imap { } protocol pop3 { pop3_uidl_format = %08Xu%08Xv } protocol managesieve { sieve=~/.dovecot.sieve sieve_storage=~/sieve } auth default { mechanisms = plain passdb pam { } userdb passwd { } user = root } dict { } plugin { } /etc/pam.d/dovecot: #%PAM-1.0 @include common-auth @include common-account @include common-session Manual test of POP3 Authentication: root@localhost:/etc/pam.d# telnet localhost 110 Trying 127.0.0.1... Connected to localhost. Escape character is '^]'. +OK Dovecot ready. user otrs-incoming +OK pass XXXXXX -ERR Temporary authentication failure.

2 3

[Dovecot] Relay-Ctrl, error message
by Wouter van der Schagt 30 Nov '08

30 Nov '08

Hi all, I'm trying to get Relay-Ctrl to work (as explained on http://wiki.dovecot.org/HowTo/PopBSMTPAndDovecot) however when i add: mail_executable: /usr/local/bin/envdir /etc/relay-ctrl /usr/local/bin/relay-ctrl-chdir /usr/local/bin/dovecot-settcpremoteip /usr/local/bin/relay-ctrl-allow /usr/local/libexec/dovecot/imap to /usr/local/etc/dovecot.conf I get the following error message when doing dovecot -n Error: Error in configuration file /usr/local/etc/dovecot.conf line 126: Expecting '=' Fatal: Invalid configuration in /usr/local/etc/dovecot.conf So I'm guessing i misplaced the line somewhere in the config file. Where should it be placed? Does it work for both pop3 and imap? Sincerely, - Wouter van der Schagt

2 1

[Dovecot] Cache details
by Warren Baker 30 Nov '08

30 Nov '08

Hi All, Is there any specific details we can extract from dovecots authentication cache (besides Authentication cache hits)? Reason i am asking is because we had a mysql user account which was returning a unknown user account. Fixing the one value in the sql result didnt have any affect and Dovecot was still returning an auth failure for the user after 24 hours had elapsed, even though auth_cache_negative_ttl was set to 120. Restarting Dovecot or a SIGHUP on dovecot-auth sorted out the problem. It would be useful if we could interrogate the cache somehow to see what cache entry is been hit. Auth cache settings are as follows: auth_cache_size = 51200 auth_cache_ttl = 2100 auth_cache_negative_ttl = 120 Any suggestions appreciated. Thanks, Warren

2 1

[Dovecot] INTERNALDATE call, timezone randomly changes
by Guy 30 Nov '08

30 Nov '08

Hi guys, I've run into an odd little problem. We use SquirrelMail as the basis for our webmail. Recently I've had a user complaining that the "Received" time on mail in the Sent folder has changed between logins. The change is always been by +1 hours or -5 hours. While checking, our developer had a time changed by +2 hours. On refresh it changed back. He poked through the code and it looks like SquirrelMail gets the received date via an IMAP command: A003 FETCH 1:10 (INTERNALDATE FLAGS BODY.PEEK[HEADER.FIELDS (To Content-Type Subject Date)]) I'm running dovecot 1.1.6, and I'm wondering how that function gets the date and time? If I can isolate the source of the timezone I can try and correct it. The server time shown by "date" is correct and /etc/localtime is /usr/share/zoneinfo/Europe/London (which is correct - I've also se the timezone in /etc/conf.d/clock is set to "Europe/London" now), so I'm wondering where else the information could be coming from? Thanks for any information you can give me to help isolate the problem. Guy -- Don't just do something...sit there!

2 1

[Dovecot] v1.2.alpha4 released
by Timo Sirainen 30 Nov '08

30 Nov '08

http://dovecot.org/releases/1.2/alpha/dovecot-1.2.alpha4.tar.gz http://dovecot.org/releases/1.2/alpha/dovecot-1.2.alpha4.tar.gz.sig Note that there are significant changes in this release from alpha3 and I haven't tested everything extensively. Be careful. The next release will hopefully be v1.2.beta1 with a feature freeze. Most importantly shared mailboxes and IMAP ACLs are finally supported. This is mainly thanks to the Kolab people (looks like I forgot your names from NEWS file, the next release will have them). I'll have to write a wiki page about ACLs, but the main thing you need to know is that if you want shared mailboxes to be LISTable by other users, you'll need to specify acl_dict. If you don't expect many shared mailboxes, a file backend is the easiest, but you could also use SQL or Berkeley DB. With a file backend you could also create per-domain files, which will improve performance if you don't need cross-domain shared mailboxes. You could do this with e.g.: acl_file = /home/virtual/%d/shared-mailboxes Also full text searching works now better with virtual mailboxes. Previously the fts would have duplicated the indexes for virtual mailboxes. Now with Solr backend it searches the original mailboxes directly. Although this is still missing some optimizations and also it doesn't check first if mails are missing from the original mailboxes, so it may not find everything. Also this still doesn't work with Squat backend.. Anyway, a bit more code and Dovecot will have fast full text searching from all mailboxes (virtual * mailbox). I implemented also draft-gulbrandsen-imap-response-codes-04, which adds some [codes] for all IMAP command failures, which should make it possible for clients to figure out more easily what caused the failure. Hopefully the draft won't change much anymore. :)

2 2

[Dovecot] Dovecot as proxy for M$ Exchange mailboxes
by Axel Schmalowsky 30 Nov '08

30 Nov '08

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi list, is there a way to use dovecot as proxy for m$ exchange mailboxes? Currently, our clients not only can login to their mailboxes with their e-mail address but also with their accountid. Both the clients e-mail address and accountid are stored in ldap. Now I need a way to let dovecot rewriting the client's accountid to the respective e-mail address. Is this possible with the proxy-feature of dovecot? Any hints would be appreciated. - -- Freundliche Grüße / Kind regards Axel Schmalowsky Platform Engineer ___________________________________ domainfactory GmbH Oskar-Messter-Str. 33 85737 Ismaning Germany Telefon: +49 (0)89 / 55266-356 Telefax: +49 (0)89 / 55266-222 E-Mail: aschmalowsky(a)df.eu Internet: www.df.eu Registergericht: Amtsgericht München HRB 150294, Geschäftsführer Tobias Marburg, Jochen Tuchbreiter -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFJL+JKsuqpduCyZM0RAoFUAJ9mPbv5eI5xP54YhwlmOs40F1RxDQCgir5l jCdaNeXKO8FnOCxaKSI7S4E= =m5WN -----END PGP SIGNATURE-----

3 2

Re: [Dovecot] Dovecot - Vpopmail subdirectories
by Yacoubi Mohamed 28 Nov '08

28 Nov '08

Hello, Thank you for this link, but this is not what I want. Dovecot is only used for imap. Currently, authentication works fine, but new fields are created in a subdirectory. Example: /var/vpopmail/domains/0/domain.com Once authenticated, I just want to be in chroot /var/vpopmail/domains/0/domain.com/user and not in /var/vpopmail/domains/domain.com/user I explained that vpopmail which manages the creation of areas and I can not influence the subdirectory. Sincerely, -- Mohamed YACOUBI

2 2

[Dovecot] ManageSieve and invalid scriptname
by Miguel Filho 28 Nov '08

28 Nov '08

Hello there, I have been using pysieved and avelsieve and it has been working great. I decided to do test with the ManageSieve patch and got this problem: Nov 27 17:21:29 cambui dovecot: MANAGESIEVE(miguel): sieve-storage: using active sieve script path: ~/.dovecot.sieve Nov 27 17:21:29 cambui dovecot: MANAGESIEVE(miguel): sieve-storage: using sieve script storage directory: ~/.sieve Nov 27 17:21:29 cambui dovecot: MANAGESIEVE(miguel): sieve-storage: relative path to sieve storage in active link: .sieve/ Nov 27 17:21:29 cambui dovecot: MANAGESIEVE(miguel): sieve-storage: Active sieve script symlink /home/admsis/miguel/.dovecot.sieve is broken: invalid scriptname (points to .sieve/phpscript). Well as you can see, a file without the .sieve is not welcome :-( I checked the RFC and there is no requirement for a .sieve file extension when considering scriptnames. http://tools.ietf.org/html/draft-martin-managesieve-12#section-1.6 Is this a misplaced restriction or it really should be enforced for any reason? I hope that this can be tolerable, or I will have to rename a lot of scripts and remove all hardcoded "phpscript" strings from avelsieve :-( Regards and thanks for the great work. Miguel

3 6

[Dovecot] Sieve and global scripts
by Allan Cassaro 28 Nov '08

28 Nov '08

Hi Stephan Bosch, there's any way to make the dovecot deliver agent (with sieve) load a "global sieve script" before load the user sieve file? I want to create some rules for every one, like: "if header contais: IS SPAM, move to SPAM BOX"... I made this today with a couple of "sed + awk" and a wraper for managesieve-login... but, this is definily not the best way to do this... This is possible to do? (Even with change in the code) Regards. -- "Não deixe a faculdade atrapalhar seus estudos." http://allan.cassaro.googlepages.com

4 3

Re: [Dovecot] UIDL question
by Timo Sirainen 28 Nov '08

28 Nov '08

On Nov 28, 2008, at 6:43 PM, Wouter van der Schagt wrote: >> This mixes v3 syntax with v1 syntax, won't work. > > You're right, after changing it to the v3 syntax i'm getting exactly > the same output in both servers after issuing the UIDL command in a > telnet session. So far so good. > > However, a 2nd copy of the files is still being downloaded. Is this > because the IP has changed? My steps are. > > - Download emails from IP.IP.IP.1 > - Migrate emails > - Confirm manually that UIDL output is the same > - If so, download emails from IP.IP.IP.2 > > And in the last step, im still getting a 2nd copy of all emails. > Where is the fault in my reasoning? Do i need to give dovecot a > message number that was last downloaded? If the UIDLs are exactly the same, I've no idea why the client would download a duplicate. I haven't heard before that a changed IP would be an issue. Of course that would still be a client-specific issue. Sounds a bit unlikely though. Ccing back to list in case someone else has an idea..

2 1