May 2006 - dovecot - dovecot.org

[Dovecot] timeout_remove and "Timeout leak" problems
by David Friedlander 03 May '06

03 May '06

Hello, I have run into troubles trying to jump from 1.0alpha5 (basically running smoothly) to 1.0beta5 and/or beta7. I am seeing errors such as the following set for nearly every user. dovecot: May 01 16:21:02 Info: IMAP(joeuser): Disconnected: Logged out dovecot: May 01 16:21:02 Error: IMAP(joeuser): file ioloop.c: line 153 (timeout_remove): assertion failed: (*timeout != NULL) dovecot: May 01 16:21:03 Info: IMAP(joeuser): Disconnected: Logged out dovecot: May 01 16:21:03 Warning: IMAP(joeuser): Timeout leak: 0x111acc dovecot: May 01 16:21:03 Error: child 20607 (imap) killed with signal 6 (INBOX is on local RAID disks while home directory (hence, ~/Maildir) is on a NetApp NFS file server.) While I tweaked settings for the differences between the 1.0a5 conf file and the beta conf file, things are fundamentally the same between them. The other critical element to this is that imap-login sessions are being left around and the load average climbs inexorably. (It reached 50 on this dual dual-core machine before I killed things off and started the daemon from scratch.) This occurs for users using a whole array of clients: Thunderbird, SquirrelMail, Eudora, Apple Mail, etc. How can I track this down or can someone please help me figure out what is happening here? (I have been unable to find answers between the mailing list archives and Google.) Thanks in advance, David P.S. The only item of note from examining ioloop.c was the differences between the alpha and beta source code: alpha5: void timeout_remove(struct timeout *timeout) { i_assert(timeout != NULL); timeout->destroyed = TRUE; } beta7: void timeout_remove(struct timeout **timeout) { i_assert(*timeout != NULL); (*timeout)->destroyed = TRUE; *timeout = NULL; }

2 1

[Dovecot] Re: mail shown throught outlook, shows todays date
by richs＠whidbey.net 03 May '06

03 May '06

> bclark wrote: > > > The interesting thing is, once I have some mail on the dovecot, > the date > > displayed on Mozilla is correct, but the mail shown throught > outlook, > > shows todays date (thats for all mail). > > Each mail has (at least) two dates : the date it was sent (stored > in the > headers), and the date it arrived (in Maildir, the date of the > file). Most > likely, Mozilla and SquirrelMail are showing the date in the > headers, and Outloo > is showing the arrival date. > > This problem has come up a few times, I believe. As I recall, most > mail clients > allow you to select which date to sort by. If you can't find the > option, it may > be somewhere in this list's archives. > > > Almost seems as Outlook (office 2000) looks at the date / time > stamp of > > the file it self as opposed to opening and reading the mail and > reading > > the necassry tags. > > That's my assessment, also. > > > If anyone could help me understand this (is this a dovecot issue), I > > would really appreciate it. > > I don't believe it's a dovecot issue, per se. > > -- > Curtis Maloney > cmaloney <at> cardgate.net I've been testing Dovecot 1.0beta7 to implement IMAP for the first time, and have had a good experience so far. But like Curtis and Brent, I found that Outlook and Outlook Express display "incorrect" dates on messages uploaded via APPEND. That's because they sort messages using INTERNALDATE: wea3 SELECT "Test Folder" pqng UID FETCH 1:* (UID FLAGS RFC822.SIZE BODY.PEEK[HEADER] INTERNALDATE) Some have created scripts to sync file timestamps with "Date" headers, which conflicts with the concept of "INTERNALDATE". Others ask users to add the "Date Sent" column, but those dates are only revealed after a message is viewed. I'd like to propose the attached patch to add an optional "outlook- dates" IMAP workaround, which returns the message "Date:" even when INTERNALDATE is specified. Since "BODY.PEEK" is being called anyway, our tests didn't reveal any performance impact. Can anyone think of a scenario this would negatively impact? Another option might be to patch the APPEND function? Thanks! Rich Sandberg richs(a)whidbey.net

4 7

[Dovecot] Expunge is not expunging unless it is called twice
by Bill Boebel 03 May '06

03 May '06

Dovecot is requiring two "expunge" calls before it deletes an email. It just seems to ignore the first expunge call, leaving the email file in the maildir cur/ folder. Any idea why? I even waited 10 seconds after setting the Deleted flag, but still the first expunge call did nothing.... 1 OK User logged in 2 select "INBOX" * FLAGS (\Answered \Flagged \Deleted \Seen \Draft) * OK [PERMANENTFLAGS (\Answered \Flagged \Deleted \Seen \Draft \*)] Flags permitted. * 12 EXISTS * 0 RECENT * OK [UIDVALIDITY 1144082550] UIDs valid * OK [UIDNEXT 1509] Predicted next UID 2 OK [READ-WRITE] Select completed. 3 UID STORE 1511 +FLAGS (\Deleted) * 12 FETCH (UID 1508 FLAGS (\Deleted \Seen)) 3 OK Store completed. 4 expunge 4 OK Expunge completed. 5 expunge * 12 EXPUNGE 5 OK Expunge completed.

2 1

[Dovecot] errors while testing with imaptest
by Jens Laas 03 May '06

03 May '06

I get lots of these (in error log): dovecot: Apr 28 14:50:03 Error: IMAP(jenslt): Corrupted index cache file /home0/jenslt/.imap/INBOX/dovecot.index.cache: record points outside file And some of these: dovecot: Apr 28 14:50:03 Error: IMAP(jenslt): Corrupted index cache file /home0/jenslt/.imap/INBOX/dovecot.index.cache: field index too large (2037149797 >= 19) Should I worry or not? (Assuming its not due to me having broken imaptest). This was with command: imaptest jenslt@xxxx:yyy pass=xxxx clients=40 msgs=300 Local software raid 5 on linux 2.6.16.11 on reiserfs. Cheers, Jens ----------------------------------------------------------------------- 'Old C programmers don't die ... they're just cast into void*' ----------------------------------------------------------------------- Jens Låås Email: jens.laas(a)data.slu.se Department of Computer Services, SLU Phone: +46 18 67 35 15 Vindbrovägen 1 P.O. Box 7079 S-750 07 Uppsala SWEDEN -----------------------------------------------------------------------

2 1

[Dovecot] Dovecot LDA - Performance/issues versus postfix delivery? Also, installing on OS X?
by Dean Blackburn 03 May '06

03 May '06

Can anyone give me an idea of whether or not dovecot-lda can solve the following issue: We seem to have constantly-corrupting dovecot* files for people who fit one (or more) of the following categories: 1) 1000+ message inboxes; 2) multiple clients simultaneously checking mail; 3) people receiving a lot of mail, and/or the server receiving and trying to deliver a single message to all users at once We're using straight postfix maildir delivery, and while the mail keeps coming, we start to see a lot of "temp.xxxx.xxxx.xxx" type lockfiles build up after any inconsistencies appear in the Inbox indexes, and clients start to report timeouts and/or crash as a result. Would switching to LDA by itself resolve this issue (on the argument that delivery would ALSO update the indexes every time), or is there some more fundamental issue we are having that is causing the inconsistency to arise? On a related note: I noticed the darwinports version of 1.0beta7 does not include the code for dovecot-lda... Does anyone have experience setting this up? Thanks! -deano

3 3

[Dovecot] madvise() errors
by Mark E. Mallett 02 May '06

02 May '06

(I think this has been mentioned before.) Do these hurt?: May 2 13:41:30 sysname dovecot: POP3(xxxx): madvise() failed with file /users/xx/xxxx/Maildir/dovecot.index.log: Operation not supported I think not, since I've been seeing them forever. But they do muddy the logs. This is on an old BSD/OS server, where the man page for madvise ends with: BUGS This call is not yet supported. The messages are *only* about index.log files. Would it be possible to either suppress the log message if the error is specifically "not supported" or, better yet, detect it at configure time? Completely understandable if it's not worth the effort, though. mm

2 2

Re: [Dovecot] managesieve daemon
by Gerald Dachs 02 May '06

02 May '06

On Tue, 02 May 2006 15:10:16 -0500 Richard Laager <rlaager(a)wiktel.com> wrote: > On Tue, 2006-05-02 at 13:00 +0200, Gerald Dachs wrote: > > Is this interesting > > for somebody else? > > I would be interested in this for a migration at work. I may be able to > provide financial just one hour ago I broke an AMD 2700+ XP CPU (yes, it's physically broken), because I am stupid, so one of my development machines is down and my finance minister (my wife) would kill me (or would be angry with me, that would be more worse than being killed) if I would ask for a new CPU. So a contribution in form of another fast XP CPU would speed up the process ;), but is not absolutely necessary, because I can set up another machine. > or code contributions as well. only if you have something ready. The server itself is running, I use libut for this, I highly recomend this http://libut.sourceforge.net/primer.html. Last thing I was working on before the machine died (code is safe), was the command parser with flex, not ready yet. Iater I could need help with the autoconf, automake stuff, this is not one of my strength. Gerald

1 0

[Dovecot] create_dovecot_shares
by Gavin Henry 02 May '06

02 May '06

Dear All, We've been working on this for a client to help create and manage maildir shares, via the dovecot-shared file. Our problems are creating lots of dovecot-shared files and permissions and monitoring new Maildir folders that get added by the user (maybe a better way, would love to know). The tool is almost done, with the docs at: http://www.suretecsystems.com/our_docs/create_dovecot_shares.html Maybe this can be of use to others when it's finished. Thanks, Gavin. P.S> Might be redundant a little when the ACLs are complete, as per Timos' e-mail earlier today: http://dovecot.org/pipermail/dovecot/2006-May/012991.html -- Kind Regards, Gavin Henry. Managing Director. T +44 (0) 1224 279484 M +44 (0) 7930 323266 F +44 (0) 1224 742001 E ghenry(a)suretecsystems.com Open Source. Open Solutions(tm). http://www.suretecsystems.com/

2 3

[Dovecot] Assertion failed
by Jonathan Knight 02 May '06

02 May '06

We're in a bit of a panic here as we're getting a lot of errors along the lines of: May 2 12:39:00 lenny dovecot: IMAP(XXXXX): file mbox-sync-rewrite.c: line 106 (mbox_sync_headers_add_space): assertion failed: (start_pos < data_size) We didn't see this until our mail servers went under a full load and now we've trashed the ability of nearly all our users to read email. We saw this first under 1.0Beta3 and we've tried building 1.0Beta7 and its made no difference. I've tried fiddling with the "dirty" settings for the indices and banning mmap to see if there's any way way of avoiding the broken code, but with no luck. We're running White Box EL4 (2.6 Linux kernel) Does anyone know a workaround or whether there is an older version that doesn't have the bug? We desperately need to get an email service running and I don't want to go back to (shudder) UW-imap if I can help it but my options are limited at the moment. -- ______ J.Knight(a)kis.keele.ac.uk Jonathan Knight, / Keele Information Services / _ __ Telephone: +44 1782 583478 University of Keele, Keele, (_/ (_) / / Fax : +44 1782 713082 Staffordshire. ST5 5BG. U.K.

5 9

[Dovecot] Dovecot and NFS with file locking
by Nate Sanders 02 May '06

02 May '06

Greetings all, I'm trying to get an understanding of a problem we are facing here. We're currently running dovecot 1.0-beta3 and have a long standing issue of system crashes on our mail server (Debian Linux 2.4.27-2-k7-smp). Here's what is happening: The machine hangs and the system load climbs as high as 80.0+. Yet, the system response is not effected. Command line is still responds instantly. There are multiple running dovecot PIDs, even if I stop the service. If I try to kill or -9 the PIDs, they will not die. The machine is DOA and must be forcefully restarted. Issuing a reboot will cause the machine to hang when it attempts to unmount network shares. Here's the setup: - Dovecot 1.0-beta3 - lock_method = dotlock - mmap_disable = yes /var/mail is store locally on the mail server and access via NFS to ALL remote machines. All remote machines have /var/mail sym linked to the NFS share on Mail. /home on Mail is NFS'd to another set of servers where IMAP mail folders reside in mbox format. All client machines have /home sym linked to the second NFS server. In other words, there's a lot of NFS shares and one mail transaction can involve 3 machines. What I'm trying to find out is the current state of NFS locking with Dovecot. This system hang happens 1-3 times a week. The current /home NFS mounts are running from SGI machines on IRIX 6.5. Clients are all Linux (debian) 2.4 or Linux (ubuntu) 2.6. Is our setup too much for Dovecot to handle? Are there other variables we're not looking at here? Thanks everyone. -- ============================================== Nate Sanders nate(a)ima.umn.edu Associate Systems Manager (612) 624 - 4353 http://www.ima.umn.edu/ ============================================== Institute for Mathematics and its Applications University of Minnesota 400 Lind Hall, 207 Church St. SE Minneapolis, MN 55455-0463 ==============================================

3 4