From stefan at schloerholz.de Mon Jul 1 05:16:48 2013 From: stefan at schloerholz.de (=?UTF-8?Q?=22Stefan_Schl=C3=B6rholz=22?=) Date: Mon, 1 Jul 2013 11:16:48 +0200 (CEST) Subject: No subject Message-ID: An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20130701/7342ba27/attachment.html From simon.matter at invoca.ch Mon Jul 1 08:11:03 2013 From: simon.matter at invoca.ch (Simon Matter) Date: Mon, 1 Jul 2013 14:11:03 +0200 Subject: In-Reply-To: References: Message-ID: <255779d6116aad3b5078af26153de543.squirrel@webmail.bi.corp.invoca.ch> Hi, Did you try running "reconstruct -r -f ..."? BTW, could you change your mailer so you don't send HTML only? Regards, Simon >

Hello,

> >

I had a cyrus installation running that worked perfectly. After a > crash of a hard drive I had to reinstall the system. The hard drive > contained the cyrus installation. However, the cyrus > "partition-default" was on a different hard drive.

> >

I reinstalled cyrus with the "partition-default" as > before. The user mailbox can be seen in cyradmin [user.paul > (\HasChildren)]. However all subfolders exept user.paul.Drafts, > user.paul.Mülleimer, user.paul.Versendete_Nachrichten can NOT be > seen. The directories and the messages in the storage partion under > /var/spool/imap/user/paul/ are all there.

> >

With a squirrelmail I do see all messages in INBOX and I do see the > three subfolders. However, the subfolders are being shown to be empty. The > files in the corresponding directories exist.

> >

I did a reconstruct -r with the cyrus admin allready, however I still > would have the data as they were after the hard drive crash.

> >

Is there anybody who can help to get the old subfolders and messages > back?

> >

Best regards

> >

Stefan

> >

> ---- > Cyrus Home Page: http://www.cyrusimap.org/ > List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/ > To Unsubscribe: > https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus From murch at andrew.cmu.edu Mon Jul 1 14:09:43 2013 From: murch at andrew.cmu.edu (Ken Murchison) Date: Mon, 01 Jul 2013 14:09:43 -0400 Subject: cyrus-imapd-2.4.17-caldav-beta6 released Message-ID: <51D1C5E7.7080308@andrew.cmu.edu> We are pleased to announce the sixth beta release of Cyrus IMAP with integrated calendaring and contacts. This is a bug fix release, with only a few minor features added. The biggest changes are support for unixhierarchysep in all HTTP modules, and the switch from RSS 2.0 to Atom 1.0 format for RSS feeds. This code is based on the stable Cyrus 2.4.17 release with support for CalDAV, CardDAV and RSS added. All of the standard Cyrus IMAP daemons and utilities should be considered production quality in this release, but the HTTP support (CalDAV, CardDAV and RSS) is in beta status. You can download via HTTP or FTP: http://cyrusimap.org/releases/cyrus-imapd-2.4.17-caldav-beta6.tar.gz ftp://ftp.cyrusimap.org/cyrus-imapd/cyrus-imapd-2.4.17-caldav-beta6.tar.gz Installation documentation will be found in doc/install-http.html in the distribution. Thanks for your continued support, and we look forward to any and all feedback. -- Kenneth Murchison Principal Systems Software Engineer Carnegie Mellon University From lst_hoe02 at kwsoft.de Thu Jul 4 06:13:05 2013 From: lst_hoe02 at kwsoft.de (lst_hoe02 at kwsoft.de) Date: Thu, 04 Jul 2013 12:13:05 +0200 Subject: MDOSEQ per Mailbox setting Message-ID: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> Hello, we use Cyrus as our central IMAP store and have a Webmail system which uses IMAP to access user mail and provide some sync access via ActiveSync. The documentation of the webmail system mentioned the following: "Cyrus support QRESYNC but do not enable per mailbox MODSEQ by default. Enabling this on these servers will greatly improve performance." But i failed to see how configure MODSEQ in Cyrus. Any pointers for this one? System is Cyrus 2.4.17 running on Ubuntu 12.04 x64 Many Thanks Andreas -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 6144 bytes Desc: S/MIME Cryptographic Signature Url : http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20130704/265a4157/attachment.bin From lst_hoe02 at kwsoft.de Fri Jul 5 03:41:29 2013 From: lst_hoe02 at kwsoft.de (lst_hoe02 at kwsoft.de) Date: Fri, 05 Jul 2013 09:41:29 +0200 Subject: MDOSEQ per Mailbox setting In-Reply-To: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> References: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> Message-ID: <20130705094129.Horde.xTLv8naeNZHNgRMNIxBFaA2@webmail.kwsoft.de> Is this list dead or the question not clear/interesting? I have not seen a single post beside me?? Regards Andreas -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 6144 bytes Desc: S/MIME Cryptographic Signature Url : http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20130705/7e83cbcd/attachment.bin From awilliam at whitemice.org Fri Jul 5 06:45:35 2013 From: awilliam at whitemice.org (Adam Tauno Williams) Date: Fri, 05 Jul 2013 06:45:35 -0400 Subject: MDOSEQ per Mailbox setting In-Reply-To: <20130705094129.Horde.xTLv8naeNZHNgRMNIxBFaA2@webmail.kwsoft.de> References: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> <20130705094129.Horde.xTLv8naeNZHNgRMNIxBFaA2@webmail.kwsoft.de> Message-ID: <1373021135.2820.0.camel@workstation.wmmi.net> On Fri, 2013-07-05 at 09:41 +0200, lst_hoe02 at kwsoft.de wrote: > Is this list dead or the question not clear/interesting? > I have not seen a single post beside me?? There are regular, at least daily [on average], posts to this list. What is the question? From lst_hoe02 at kwsoft.de Fri Jul 5 06:47:25 2013 From: lst_hoe02 at kwsoft.de (lst_hoe02 at kwsoft.de) Date: Fri, 05 Jul 2013 12:47:25 +0200 Subject: MDOSEQ per Mailbox setting In-Reply-To: <3278.313157c5.1373017021.nsm@mail.trader-internet.dk> References: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> <20130705094129.Horde.xTLv8naeNZHNgRMNIxBFaA2@webmail.kwsoft.de> <3278.313157c5.1373017021.nsm@mail.trader-internet.dk> Message-ID: <20130705124725.Horde.9kx_YPo8xJNS0ZivNcmRGQ1@webmail.kwsoft.de> Zitat von Mogens Melander : > A search on google for MDOSEQ return only your post to the > list. > > A grep for MDOSEQ in the source for cyrus-imapd-2.4.16 > does not reveal anything. The word MDOSEQ was not anywhere > in the source tree. > Sh... Typo in the subjet. It should be MODSEQ of course as described in the first mail to the list. http://tools.ietf.org/html/rfc4551 Sorry Andreas -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 6144 bytes Desc: S/MIME Cryptographic Signature Url : http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20130705/9b1ac76c/attachment.bin From awilliam at whitemice.org Fri Jul 5 06:53:38 2013 From: awilliam at whitemice.org (Adam Tauno Williams) Date: Fri, 05 Jul 2013 06:53:38 -0400 Subject: MDOSEQ per Mailbox setting In-Reply-To: <20130705124725.Horde.9kx_YPo8xJNS0ZivNcmRGQ1@webmail.kwsoft.de> References: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> <20130705094129.Horde.xTLv8naeNZHNgRMNIxBFaA2@webmail.kwsoft.de> <3278.313157c5.1373017021.nsm@mail.trader-internet.dk> <20130705124725.Horde.9kx_YPo8xJNS0ZivNcmRGQ1@webmail.kwsoft.de> Message-ID: <1373021618.2820.4.camel@workstation.wmmi.net> On Fri, 2013-07-05 at 12:47 +0200, lst_hoe02 at kwsoft.de wrote: > Zitat von Mogens Melander : > > A search on google for MDOSEQ return only your post to the > > list. > > A grep for MDOSEQ in the source for cyrus-imapd-2.4.16 > > does not reveal anything. The word MDOSEQ was not anywhere > > in the source tree. > Sh... > Typo in the subjet. It should be MODSEQ of course as described in the > first mail to the list. > http://tools.ietf.org/html/rfc4551 I believe this falls under the label "condstore" in Cyrus [???]. In 2.3.x you could toggle it on via mboxcfg in cyradm per-mailbox. In Cyrus 2.4.x it is just "on". From Hagedorn at uni-koeln.de Fri Jul 5 06:54:09 2013 From: Hagedorn at uni-koeln.de (Sebastian Hagedorn) Date: Fri, 05 Jul 2013 12:54:09 +0200 Subject: MDOSEQ per Mailbox setting In-Reply-To: <20130705094129.Horde.xTLv8naeNZHNgRMNIxBFaA2@webmail.kwsoft.de> References: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> <20130705094129.Horde.xTLv8naeNZHNgRMNIxBFaA2@webmail.kwsoft.de> Message-ID: --On 5. Juli 2013 09:41:29 +0200 lst_hoe02 at kwsoft.de wrote: > Is this list dead or the question not clear/interesting? > > I have not seen a single post beside me?? The list isn't dead, it just doesn't have high volume: To partially answer your question: " * Added support for CONDSTORE (must be enabled on a per-mailbox basis with the /vendor/cmu/cyrus-imapd/condstore mailbox annotation." I thought it should work like this, but I get "Permission denied": $ cyradm cyrus Password: imap> mboxconfig user/xxx /vendor/cmu/cyrus-imapd/condstore true mboxconfig: Permission denied imap> -- .:.Sebastian Hagedorn - Weyertal 121 (Geb?ude 133), Zimmer 2.02.:. .:.Regionales Rechenzentrum (RRZK).:. .:.Universit?t zu K?ln / Cologne University - ? +49-221-470-89578.:. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pkcs7-signature Size: 5313 bytes Desc: not available Url : http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20130705/80af0177/attachment-0001.bin From lst_hoe02 at kwsoft.de Fri Jul 5 07:02:08 2013 From: lst_hoe02 at kwsoft.de (lst_hoe02 at kwsoft.de) Date: Fri, 05 Jul 2013 13:02:08 +0200 Subject: MDOSEQ per Mailbox setting In-Reply-To: <1373021618.2820.4.camel@workstation.wmmi.net> References: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> <20130705094129.Horde.xTLv8naeNZHNgRMNIxBFaA2@webmail.kwsoft.de> <3278.313157c5.1373017021.nsm@mail.trader-internet.dk> <20130705124725.Horde.9kx_YPo8xJNS0ZivNcmRGQ1@webmail.kwsoft.de> <1373021618.2820.4.camel@workstation.wmmi.net> Message-ID: <20130705130208.Horde.R6cl2fNg3lufehP5mt2c4A1@webmail.kwsoft.de> Zitat von Adam Tauno Williams : > On Fri, 2013-07-05 at 12:47 +0200, lst_hoe02 at kwsoft.de wrote: >> Zitat von Mogens Melander : >> > A search on google for MDOSEQ return only your post to the >> > list. >> > A grep for MDOSEQ in the source for cyrus-imapd-2.4.16 >> > does not reveal anything. The word MDOSEQ was not anywhere >> > in the source tree. >> Sh... >> Typo in the subjet. It should be MODSEQ of course as described in the >> first mail to the list. >> http://tools.ietf.org/html/rfc4551 > > I believe this falls under the label "condstore" in Cyrus [???]. In > 2.3.x you could toggle it on via mboxcfg in cyradm per-mailbox. In > Cyrus 2.4.x it is just "on". > Okay, so the statement from the webmail system only applies for Cyrus versions < 2.4 and in 2.4.x this is enabled for any mailbox without configuration/upgrade?? Thanks Andreas -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 6144 bytes Desc: S/MIME Cryptographic Signature Url : http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20130705/142c67db/attachment.bin From brong at fastmail.fm Sun Jul 7 06:37:56 2013 From: brong at fastmail.fm (Bron Gondwana) Date: Sun, 07 Jul 2013 20:37:56 +1000 Subject: MDOSEQ per Mailbox setting In-Reply-To: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> References: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> Message-ID: <1373193476.18664.140661252758777.07C3B33F@webmail.messagingengine.com> On Thu, Jul 4, 2013, at 08:13 PM, lst_hoe02 at kwsoft.de wrote: > "Cyrus support QRESYNC but do not enable per mailbox MODSEQ by > default. Enabling this on these servers will greatly improve > performance." > > But i failed to see how configure MODSEQ in Cyrus. Any pointers for this one? That was true of Cyrus 2.3. Cyrus 2.4 has MODSEQ enabled for all mailboxes. You can't even turn it off! It's embedded into all sorts of logic now. Bron. -- Bron Gondwana brong at fastmail.fm From lst_hoe02 at kwsoft.de Mon Jul 8 10:53:28 2013 From: lst_hoe02 at kwsoft.de (lst_hoe02 at kwsoft.de) Date: Mon, 08 Jul 2013 16:53:28 +0200 Subject: MDOSEQ per Mailbox setting In-Reply-To: <1373193476.18664.140661252758777.07C3B33F@webmail.messagingengine.com> References: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> <1373193476.18664.140661252758777.07C3B33F@webmail.messagingengine.com> Message-ID: <20130708165328.Horde.cfFClZ318dT5RTWdumMNTA6@webmail.kwsoft.de> Zitat von Bron Gondwana : > On Thu, Jul 4, 2013, at 08:13 PM, lst_hoe02 at kwsoft.de wrote: >> "Cyrus support QRESYNC but do not enable per mailbox MODSEQ by >> default. Enabling this on these servers will greatly improve >> performance." >> >> But i failed to see how configure MODSEQ in Cyrus. Any pointers for >> this one? > > That was true of Cyrus 2.3. Cyrus 2.4 has MODSEQ enabled for all mailboxes. > > You can't even turn it off! It's embedded into all sorts of logic now. > > Bron. Ok, so i can tell the wiki maintainer that the information should mention that this only applies to Cyrus <2.4. One more related question: I have got this warning in the logfile which also pointing to MODSEQ problems, no? "Jul 8 16:40:19 mailer cyrus/imap[32155]: inefficient qresync (1472 > 1) user.xxxxxx" From what i have read here http://git.kolabsys.com/cyrus-imapd.git/commit/?id=e3f72960b1349f1dffea0160d7bc7478f429cc54 it looks like syncronisation based on MODSEQ does not work as expected for this client? Any idea where to debug or what the actual problem could be. The warning appears on any sync attempt as far as i can tell. Regards Andreas -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 6144 bytes Desc: S/MIME Cryptographic Signature Url : http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20130708/94c518e6/attachment.bin From brong at fastmail.fm Mon Jul 8 21:45:04 2013 From: brong at fastmail.fm (Bron Gondwana) Date: Tue, 09 Jul 2013 11:45:04 +1000 Subject: MDOSEQ per Mailbox setting In-Reply-To: <20130708165328.Horde.cfFClZ318dT5RTWdumMNTA6@webmail.kwsoft.de> References: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> <1373193476.18664.140661252758777.07C3B33F@webmail.messagingengine.com> <20130708165328.Horde.cfFClZ318dT5RTWdumMNTA6@webmail.kwsoft.de> Message-ID: <1373334304.28267.140661253395274.0F5207BA@webmail.messagingengine.com> On Tue, Jul 9, 2013, at 12:53 AM, lst_hoe02 at kwsoft.de wrote: > Ok, so i can tell the wiki maintainer that the information should > mention that this only applies to Cyrus <2.4. Correct. Maybe I should have left the annotation in, just forced on! > One more related question: > > I have got this warning in the logfile which also pointing to MODSEQ > problems, no? > > "Jul 8 16:40:19 mailer cyrus/imap[32155]: inefficient qresync (1472 > > 1) user.xxxxxx" Why on earth would you send a MODSEQ of 1 and ask for all changes since then? That's crazy. Seems like a bogus client. > From what i have read here > http://git.kolabsys.com/cyrus-imapd.git/commit/?id=e3f72960b1349f1dffea0160d7bc7478f429cc54 it looks like syncronisation based on MODSEQ does not work as expected for this client? Any idea where to debug or what the actual problem could be. The warning appears on any sync attempt as far as i can > tell. Well - it works fine. The client is just being stupid. What a client is SUPPOSED to do is make an initial request for the data it wants, and then note the HIGHESTMODSEQ in the response. Next time it connects, it can "resync" by asking for changes since the last value it saw. But asking for changes since '1' - obviously that's everything. The reason we log that it's "inefficient" is that we don't have full information for every deleted message any more, so we have to assume every gap in the UID listing is a deleted message. It's still fine - the response from Cyrus is correct, and the client will work correctly too. I guess maybe I should make that message not be logged if the request is "since modseq == 1", since clients do that. Then it will only happen if you reconnect after a week and the expunged messages have been expired. Regards, Bron. -- Bron Gondwana brong at fastmail.fm From lst_hoe02 at kwsoft.de Tue Jul 9 04:06:48 2013 From: lst_hoe02 at kwsoft.de (lst_hoe02 at kwsoft.de) Date: Tue, 09 Jul 2013 10:06:48 +0200 Subject: MDOSEQ per Mailbox setting In-Reply-To: <1373334304.28267.140661253395274.0F5207BA@webmail.messagingengine.com> References: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> <1373193476.18664.140661252758777.07C3B33F@webmail.messagingengine.com> <20130708165328.Horde.cfFClZ318dT5RTWdumMNTA6@webmail.kwsoft.de> <1373334304.28267.140661253395274.0F5207BA@webmail.messagingengine.com> Message-ID: <20130709100648.Horde.ZT_vE5HJI8W1ei5wkK9YjA1@webmail.kwsoft.de> Zitat von Bron Gondwana : > On Tue, Jul 9, 2013, at 12:53 AM, lst_hoe02 at kwsoft.de wrote: >> Ok, so i can tell the wiki maintainer that the information should >> mention that this only applies to Cyrus <2.4. > > Correct. Maybe I should have left the annotation in, just forced on! > >> One more related question: >> >> I have got this warning in the logfile which also pointing to MODSEQ >> problems, no? >> >> "Jul 8 16:40:19 mailer cyrus/imap[32155]: inefficient qresync (1472 > >> 1) user.xxxxxx" > > Why on earth would you send a MODSEQ of 1 and ask for all changes > since then? That's crazy. Seems like a bogus client. > Maybe it was an attempt to get in sync again because its ActiveSync e-Mail trouble which pointed to this error. >> From what i have read here >> http://git.kolabsys.com/cyrus-imapd.git/commit/?id=e3f72960b1349f1dffea0160d7bc7478f429cc54 it looks like syncronisation based on MODSEQ does not work as expected for this client? Any idea where to debug or what the actual problem could be. The warning appears on any sync attempt as far as i >> can >> tell. > > Well - it works fine. The client is just being stupid. What a > client is SUPPOSED to do is make an initial request for the data it > wants, and then note the HIGHESTMODSEQ in the response. Next time > it connects, it can "resync" by asking for changes since the last > value it saw. > > But asking for changes since '1' - obviously that's everything. The > reason we log that it's "inefficient" is that we don't have full > information for every deleted message any more, so we have to assume > every gap in the UID listing is a deleted message. > > It's still fine - the response from Cyrus is correct, and the client > will work correctly too. I guess maybe I should make that message > not be logged if the request is "since modseq == 1", since clients > do that. Then it will only happen if you reconnect after a week and > the expunged messages have been expired. > > Regards, > > Bron. I see, thanks for explaining. But i found out that additionally most of the time the warning is followed by a "signaled to death" from master like this: Jul 8 07:39:04 mailer cyrus/imap[28235]: inefficient qresync (1364 > 1) user.xxxxxx Jul 8 07:39:04 mailer cyrus/master[23881]: process 28235 exited, signaled to death by 6 This shouldn't happen, no? Regards Andreas -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 6144 bytes Desc: S/MIME Cryptographic Signature Url : http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20130709/536ee54f/attachment.bin From brong at fastmail.fm Tue Jul 9 05:49:11 2013 From: brong at fastmail.fm (Bron Gondwana) Date: Tue, 09 Jul 2013 19:49:11 +1000 Subject: MDOSEQ per Mailbox setting In-Reply-To: <20130709100648.Horde.ZT_vE5HJI8W1ei5wkK9YjA1@webmail.kwsoft.de> References: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> <1373193476.18664.140661252758777.07C3B33F@webmail.messagingengine.com> <20130708165328.Horde.cfFClZ318dT5RTWdumMNTA6@webmail.kwsoft.de> <1373334304.28267.140661253395274.0F5207BA@webmail.messagingengine.com> <20130709100648.Horde.ZT_vE5HJI8W1ei5wkK9YjA1@webmail.kwsoft.de> Message-ID: <1373363351.21402.140661253515534.035D52A9@webmail.messagingengine.com> On Tue, Jul 9, 2013, at 06:06 PM, lst_hoe02 at kwsoft.de wrote: > I see, thanks for explaining. But i found out that additionally most > of the time the warning is followed by a "signaled to death" from > master like this: > > Jul 8 07:39:04 mailer cyrus/imap[28235]: inefficient qresync (1364 > > 1) user.xxxxxx > Jul 8 07:39:04 mailer cyrus/master[23881]: process 28235 exited, > signaled to death by 6 > > This shouldn't happen, no? Definitely not! SIGABRT 6 Core Abort signal from abort(3) Don't suppose you have a core file that can give me a backtrace do you? The only things that could "abort" there are memory allocation or invalid sequence parsing (perhaps). Also - can you enable telemetry for the user and give me the exact command which is causing the abort? You can enable telemetry by creating the directory $confdir/log/$username (i.e. /var/imap/log/brong - I think, it's a while since I did any distro-standard work) and making sure that the directory is owned by user 'cyrus'. It also only works for NEW connections, so you'll need to create a new connection before it logs anything. Regards, Bron. -- Bron Gondwana brong at fastmail.fm From lst_hoe02 at kwsoft.de Tue Jul 9 12:49:01 2013 From: lst_hoe02 at kwsoft.de (lst_hoe02 at kwsoft.de) Date: Tue, 09 Jul 2013 18:49:01 +0200 Subject: MDOSEQ per Mailbox setting In-Reply-To: <1373363351.21402.140661253515534.035D52A9@webmail.messagingengine.com> References: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> <1373193476.18664.140661252758777.07C3B33F@webmail.messagingengine.com> <20130708165328.Horde.cfFClZ318dT5RTWdumMNTA6@webmail.kwsoft.de> <1373334304.28267.140661253395274.0F5207BA@webmail.messagingengine.com> <20130709100648.Horde.ZT_vE5HJI8W1ei5wkK9YjA1@webmail.kwsoft.de> <1373363351.21402.140661253515534.035D52A9@webmail.messagingengine.com> Message-ID: <20130709184901.Horde.SWnEpA7kih1pbf2XV8mkOA1@webmail.kwsoft.de> Zitat von Bron Gondwana : > On Tue, Jul 9, 2013, at 06:06 PM, lst_hoe02 at kwsoft.de wrote: >> I see, thanks for explaining. But i found out that additionally most >> of the time the warning is followed by a "signaled to death" from >> master like this: >> >> Jul 8 07:39:04 mailer cyrus/imap[28235]: inefficient qresync (1364 > >> 1) user.xxxxxx >> Jul 8 07:39:04 mailer cyrus/master[23881]: process 28235 exited, >> signaled to death by 6 >> >> This shouldn't happen, no? > > Definitely not! > > SIGABRT 6 Core Abort signal from abort(3) > > Don't suppose you have a core file that can give me a backtrace do you? > > The only things that could "abort" there are memory allocation or > invalid sequence parsing (perhaps). > > Also - can you enable telemetry for the user and give me the exact command > which is causing the abort? You can enable telemetry by creating > the directory > $confdir/log/$username (i.e. /var/imap/log/brong - I think, it's a > while since > I did any distro-standard work) and making sure that the directory > is owned by > user 'cyrus'. It also only works for NEW connections, so you'll > need to create > a new connection before it logs anything. > > Regards, > > Bron. In the telemetry log the last entry is something about "Replacing SYNC_CACHE" and looks like it doesn't really finish with it. I will try to get a backtrace soon and send you the whole telemetry log if desired. Regards Andreas -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 6144 bytes Desc: S/MIME Cryptographic Signature Url : http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20130709/f4e87f40/attachment.bin From swball at uwaterloo.ca Tue Jul 9 15:04:54 2013 From: swball at uwaterloo.ca (Shawn Winnington-Ball) Date: Tue, 9 Jul 2013 15:04:54 -0400 Subject: mailboxes.db discrepancies between mailbox and mupdate servers Message-ID: <20130709190454.GI59773@locohost.uwaterloo.ca> Hi all, I'm having an issue with a Cyrus murder wherein the mupdate server believes that a set of mailboxes are in mid-transfer, when in fact they don't exist on any of the downstream mailbox servers. Here's an example of a lone entry gleaned from the output of `ctl_mboxlist -d' run on the mupdate server: user.foo 3 mailbox-03-internal!spool If I login to mailbox-03 and use `cyradm' to create the mailbox directory for user.foo, I get createmailbox: unable to reserve mailbox on mupdate server It seems that I need to get these sorts of entries removed from the mupdate server's mailboxes.db file so that I can go about creating the mailboxes afresh. My question is how to do this? Looking through this list's archives, I see that the cyr_dbtool command can be used to query and manipulate the mailboxes.db file itself, but someone mentioned that it's best not to modify the mupdate server's mailboxes.db file directly. In my case, if I were to try running cyr_dbtool /path/to/mailboxes.db skiplist delete 'user.foo' on the mupdate server, would I be taking an unnecessary risk? Are there other means by which to force mailbox-03 to report a complete and accurate list of its mailboxes to the mupdate server, thereby overwriting those mailboxes.db entries marked `in-transit' ? Thanks for any assistance, Shawn From dave64 at andrew.cmu.edu Tue Jul 9 15:32:42 2013 From: dave64 at andrew.cmu.edu (Dave McMurtrie) Date: Tue, 9 Jul 2013 19:32:42 +0000 Subject: mailboxes.db discrepancies between mailbox and mupdate servers In-Reply-To: <20130709190454.GI59773@locohost.uwaterloo.ca> Message-ID: <5EBEFD277BF8A84793A86CC78C44FFE3BBFE50@PGH-MSGMB-03.andrew.ad.cmu.edu> On 7/9/13 3:04 PM, "Shawn Winnington-Ball" wrote: >Hi all, > >I'm having an issue with a Cyrus murder wherein the mupdate server >believes that a set of mailboxes are in mid-transfer, when in fact >they don't exist on any of the downstream mailbox servers. Here's >an example of a lone entry gleaned from the output of `ctl_mboxlist >-d' run on the mupdate server: > >user.foo 3 mailbox-03-internal!spool > >If I login to mailbox-03 and use `cyradm' to create the mailbox >directory for user.foo, I get > >createmailbox: unable to reserve mailbox on mupdate server > >It seems that I need to get these sorts of entries removed from the >mupdate server's mailboxes.db file so that I can go about creating >the mailboxes afresh. If you log in to mailbox-03-internal and run: # ctl_mboxlist -d | grep "^user\.foo" is anything returned? For that matter, run this on each of your backend servers and see if it exists anywhere. > >My question is how to do this? Looking through this list's archives, >I see that the cyr_dbtool command can be used to query and manipulate >the mailboxes.db file itself, but someone mentioned that it's best >not to modify the mupdate server's mailboxes.db file directly. In >my case, if I were to try running > >cyr_dbtool /path/to/mailboxes.db skiplist delete 'user.foo' > >on the mupdate server, would I be taking an unnecessary risk? It might be safer to clear the reserved state via mupdate protocol. You can do this manually by using mupdatetest. You can google for the mupdate RFC if you want more details about the protocol, but basically: $ mupdatetest your.mupdate.server.com. 1 ACTIVATE "user.foo" "mailbox-03-internal!spool" "foo" "lrswipcda" > Are >there other means by which to force mailbox-03 to report a complete >and accurate list of its mailboxes to the mupdate server, thereby >overwriting those mailboxes.db entries marked `in-transit' ? You can force a backend to push all of its mailboxes to the mupdate master by running "ctl_mboxlist -m" on the backend. If you're not 100% sure whether you want to push every mailbox before you know what state things are in, you can individually push mailboxes, again using mupdate protocol. Log in to the backend and run $ mupdatetest your.mupdate.server.com. 1 MUPDATEPUSH user.foo I don't think a mupdate push will change the RESERVED state of the mailbox, though. HTH, Dave From swball at uwaterloo.ca Tue Jul 9 17:20:51 2013 From: swball at uwaterloo.ca (Shawn Winnington-Ball) Date: Tue, 9 Jul 2013 17:20:51 -0400 Subject: mailboxes.db discrepancies between mailbox and mupdate servers In-Reply-To: <5EBEFD277BF8A84793A86CC78C44FFE3BBFE50@PGH-MSGMB-03.andrew.ad.cmu.edu> References: <20130709190454.GI59773@locohost.uwaterloo.ca> <5EBEFD277BF8A84793A86CC78C44FFE3BBFE50@PGH-MSGMB-03.andrew.ad.cmu.edu> Message-ID: <20130709212051.GM59773@locohost.uwaterloo.ca> Hi Dave, Thanks for the quick, helpful response. > If you log in to mailbox-03-internal and run: > > # ctl_mboxlist -d | grep "^user\.foo" > > is anything returned? For that matter, run this on each of your backend > servers and see if it exists anywhere. None of the mailbox servers have any record of user.foo, only the mupdate server does. > It might be safer to clear the reserved state via mupdate protocol. You > can do this manually by using mupdatetest. You can google for the mupdate > RFC if you want more details about the protocol, but basically: > > $ mupdatetest your.mupdate.server.com. > 1 ACTIVATE "user.foo" "mailbox-03-internal!spool" "foo" "lrswipcda" This worked to change the status of user.foo on the mupdate server: user.foo 1 mailbox-03-internal!spool foo lrswipkxtecda > You can force a backend to push all of its mailboxes to the mupdate master > by running "ctl_mboxlist -m" on the backend. If you're not 100% sure > whether you want to push every mailbox before you know what state things > are in, you can individually push mailboxes, again using mupdate protocol. > Log in to the backend and run > > $ mupdatetest your.mupdate.server.com. > 1 MUPDATEPUSH user.foo However, I tried running this command and got B01 MUPDATEPUSH user.foo B01 BAD "Unrecognized command" I can't find MUPDATEPUSH in RFC 3656 either. Shawn From simon at leaky.org Tue Jul 9 17:28:29 2013 From: simon at leaky.org (Simon Amor) Date: Tue, 9 Jul 2013 22:28:29 +0100 Subject: mailboxes.db discrepancies between mailbox and mupdate servers In-Reply-To: <20130709212051.GM59773@locohost.uwaterloo.ca> References: <20130709190454.GI59773@locohost.uwaterloo.ca> <5EBEFD277BF8A84793A86CC78C44FFE3BBFE50@PGH-MSGMB-03.andrew.ad.cmu.edu> <20130709212051.GM59773@locohost.uwaterloo.ca> Message-ID: <440_1373405312_r69LSVH0012868_070B18D2-7BCE-4038-BC67-8131628B70F0@leaky.org> On 9 Jul 2013, at 22:20, Shawn Winnington-Ball wrote: > >> You can force a backend to push all of its mailboxes to the mupdate master >> by running "ctl_mboxlist -m" on the backend. If you're not 100% sure >> whether you want to push every mailbox before you know what state things >> are in, you can individually push mailboxes, again using mupdate protocol. >> Log in to the backend and run >> >> $ mupdatetest your.mupdate.server.com. >> 1 MUPDATEPUSH user.foo > > However, I tried running this command and got > > B01 MUPDATEPUSH user.foo > B01 BAD "Unrecognized command" > > I can't find MUPDATEPUSH in RFC 3656 either. > According to my notes, MUPDATEPUSH is used through an admin user's IMAP connection (via imtest) not through mupdatetest. If the mailbox doesn't actually exist, you might need to use LOCALCREATE via IMAP to the backend prior to MUPDATEPUSH. LC1 LOCALCREATE user.foo ACL0 SETACL user.foo foo lrswipkxtea MP1 MUPDATEPUSH user.foo I use virtual domains so the syntax of the SETACL command may be slightly off ('foo' is the authenticated user as far as I can remember). If you've already told the mupdate server about the mailbox, you probably don't need the last 2 commands, just the localcreate should be enough to let them connect to the mailbox. Simon Disclaimer: I'm not a cyrus expert - I just break it a lot :) From swball at uwaterloo.ca Tue Jul 9 17:38:28 2013 From: swball at uwaterloo.ca (Shawn Winnington-Ball) Date: Tue, 9 Jul 2013 17:38:28 -0400 Subject: mailboxes.db discrepancies between mailbox and mupdate servers In-Reply-To: <440_1373405312_r69LSVH0012868_070B18D2-7BCE-4038-BC67-8131628B70F0@leaky.org> References: <20130709190454.GI59773@locohost.uwaterloo.ca> <5EBEFD277BF8A84793A86CC78C44FFE3BBFE50@PGH-MSGMB-03.andrew.ad.cmu.edu> <20130709212051.GM59773@locohost.uwaterloo.ca> <440_1373405312_r69LSVH0012868_070B18D2-7BCE-4038-BC67-8131628B70F0@leaky.org> Message-ID: <20130709213828.GO59773@locohost.uwaterloo.ca> > According to my notes, MUPDATEPUSH is used through an admin user's IMAP connection (via imtest) not through mupdatetest. > > If the mailbox doesn't actually exist, you might need to use LOCALCREATE via IMAP to the backend prior to MUPDATEPUSH. > > LC1 LOCALCREATE user.foo Thank you Simon, this did the trick to recreate the directory structure on-disk and populate the mailboxes.db file on the backend. Shawn From brong at fastmail.fm Tue Jul 9 19:12:44 2013 From: brong at fastmail.fm (Bron Gondwana) Date: Wed, 10 Jul 2013 09:12:44 +1000 Subject: MDOSEQ per Mailbox setting In-Reply-To: <20130709184901.Horde.SWnEpA7kih1pbf2XV8mkOA1@webmail.kwsoft.de> References: <20130704121305.Horde.yCnTXmFeHicBtwbo_D1Vjg7@webmail.kwsoft.de> <1373193476.18664.140661252758777.07C3B33F@webmail.messagingengine.com> <20130708165328.Horde.cfFClZ318dT5RTWdumMNTA6@webmail.kwsoft.de> <1373334304.28267.140661253395274.0F5207BA@webmail.messagingengine.com> <20130709100648.Horde.ZT_vE5HJI8W1ei5wkK9YjA1@webmail.kwsoft.de> <1373363351.21402.140661253515534.035D52A9@webmail.messagingengine.com> <20130709184901.Horde.SWnEpA7kih1pbf2XV8mkOA1@webmail.kwsoft.de> Message-ID: <1373411564.20203.140661253808426.034C2874@webmail.messagingengine.com> On Wed, Jul 10, 2013, at 02:49 AM, lst_hoe02 at kwsoft.de wrote: > In the telemetry log the last entry is something about "Replacing > SYNC_CACHE" and looks like it doesn't really finish with it. I will > try to get a backtrace soon and send you the whole telemetry log if > desired. That would be very helpful - along with the exact version you're using. The backtrace is even better of course, but even telemetry and syslog will help a lot :) Regards, Bron. -- Bron Gondwana brong at fastmail.fm From l.schimmer at cgv.tugraz.at Wed Jul 10 03:52:26 2013 From: l.schimmer at cgv.tugraz.at (Lars Schimmer) Date: Wed, 10 Jul 2013 09:52:26 +0200 Subject: Cyrus 2.4.x logging issue Message-ID: <51DD12BA.2040303@cgv.tugraz.at> Hi! I do run debian cyrus impad 2.4.12-2+b1 on my box. But for a long time, the logging did annoy me big time. Now with a central logfile server, it annoys me even more. How can I reduce the stuff cyrus is logging? Currently it logs all login/prune/lookup/... nearly everything. Or any other tip on howto I can redeuce the logging of cyrus in conjunction with rsyslog? thank you. MfG, Lars Schimmer -- ------------------------------------------------------------- TU Graz, Institut f?r ComputerGraphik & WissensVisualisierung Tel: +43 316 873-5405 E-Mail: l.schimmer at cgv.tugraz.at Fax: +43 316 873-5402 PGP-Key-ID: 0x4A9B1723 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: OpenPGP digital signature Url : http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20130710/939dbf96/attachment.bin From dwhite at olp.net Wed Jul 10 09:21:06 2013 From: dwhite at olp.net (Dan White) Date: Wed, 10 Jul 2013 08:21:06 -0500 Subject: Cyrus 2.4.x logging issue In-Reply-To: <51DD12BA.2040303@cgv.tugraz.at> References: <51DD12BA.2040303@cgv.tugraz.at> Message-ID: <20130710132106.GA5386@dan.olp.net> On 07/10/13?09:52?+0200, Lars Schimmer wrote: >Hi! > >I do run debian cyrus impad 2.4.12-2+b1 on my box. > >But for a long time, the logging did annoy me big time. >Now with a central logfile server, it annoys me even more. > >How can I reduce the stuff cyrus is logging? >Currently it logs all login/prune/lookup/... nearly everything. > >Or any other tip on howto I can redeuce the logging of cyrus in >conjunction with rsyslog? See: http://cyrusimap.org/docs/cyrus-imapd/2.4.17/install-configure.php /usr/share/doc/cyrus-imapd-x.x/README.Debian.debug.gz Debian packages use syslog facility mail (for cyrus-imap) instead of local6. And Debian configures a verbose level by default. -- Dan White From bgould at crewcorpinc.com Wed Jul 10 20:58:21 2013 From: bgould at crewcorpinc.com (Brandon Gould) Date: Wed, 10 Jul 2013 20:58:21 -0400 Subject: TLS server engine: cannot load CA data Message-ID: <51DE032D.9090400@crewcorpinc.com> Hello all, I'm encountering some peculiar behaviour with my present configuration. Sometimes, I'll get "TLS server engine: cannot load CA data," a certificate warning on the client (Certificate validation failed for unknown reasons?). Other times, it will succeed non problemo. No warning on the client, nothing. The server is set up to force encryption... if the client isn't capable, it will refuse. Here's a log output when it fails: Jul 11 00:41:37 ip-10-0-0-201 cyrus/imap[32711]: executed Jul 11 00:41:37 ip-10-0-0-201 cyrus/imap[32711]: accepted connection Jul 11 00:41:37 ip-10-0-0-201 cyrus/imap[32711]: TLS server engine: cannot load CA data Jul 11 00:41:37 ip-10-0-0-201 cyrus/imap[32711]: imapd:Loading hard-coded DH parameters Jul 11 00:41:37 ip-10-0-0-201 cyrus/imap[32711]: SSL_accept() incomplete -> wait Jul 11 00:41:37 ip-10-0-0-201 cyrus/imap[32711]: SSL_accept() succeeded -> done Jul 11 00:41:37 ip-10-0-0-201 cyrus/imap[32711]: starttls: TLSv1 with cipher AES128-SHA (128/128 bits reused) no authentication Jul 11 00:41:38 ip-10-0-0-201 cyrus/imap[32711]: fetching user_deny.db entry for '[REDACTED]' Jul 11 00:41:38 ip-10-0-0-201 cyrus/imap[32711]: login: [REDACTED] plaintext+TLS User logged in SESSIONID= Jul 11 00:41:38 ip-10-0-0-201 cyrus/imap[32711]: fetching user_deny.db entry for '[REDACTED]' Jul 11 00:41:38 cyrus/imap[32711]: last message repeated 2 times Jul 11 00:41:38 ip-10-0-0-201 cyrus/imap[31285]: fetching user_deny.db entry for '[REDACTED]' Jul 11 00:41:38 ip-10-0-0-201 cyrus/master[32712]: about to exec /usr/lib/cyrus/bin/imapd Jul 11 00:41:38 ip-10-0-0-201 cyrus/imap[32712]: executed Jul 11 00:41:38 ip-10-0-0-201 cyrus/imap[32712]: accepted connection Jul 11 00:41:38 ip-10-0-0-201 cyrus/imap[31285]: fetching user_deny.db entry for '[REDACTED]' Jul 11 00:41:39 ip-10-0-0-201 cyrus/imap[32712]: TLS server engine: cannot load CA data Jul 11 00:41:39 ip-10-0-0-201 cyrus/imap[32712]: imapd:Loading hard-coded DH parameters Jul 11 00:41:39 ip-10-0-0-201 cyrus/imap[32712]: SSL_accept() incomplete -> wait Jul 11 00:41:39 ip-10-0-0-201 cyrus/imap[32712]: SSL_accept() succeeded -> done Jul 11 00:41:39 ip-10-0-0-201 cyrus/imap[32712]: starttls: TLSv1 with cipher AES128-SHA (128/128 bits reused) no authentication Jul 11 00:41:39 ip-10-0-0-201 cyrus/imap[32712]: fetching user_deny.db entry for '[REDACTED]' Jul 11 00:41:39 ip-10-0-0-201 cyrus/imap[32712]: login: [REDACTED] plaintext+TLS User logged in SESSIONID= Jul 11 00:41:39 ip-10-0-0-201 cyrus/imap[32712]: fetching user_deny.db entry for '[REDACTED]' Jul 11 00:41:39 ip-10-0-0-201 cyrus/imap[32712]: fetching user_deny.db entry for '[REDACTED]' Jul 11 00:41:40 ip-10-0-0-201 cyrus/imap[32712]: open: user bgould opened INBOX.Apache Directory Server Jul 11 00:41:40 ip-10-0-0-201 cyrus/imap[32712]: fetching user_deny.db entry for '[REDACTED]' Annd configuration tls_ca_file: /var/www/crets/gd_bundle.crt tls_ca_path: /etc/ssl/certs imap_tls_key_file: /var/www/certs/[REDACTED 1].key tls_cert_file: /var/www/certs/[REDACTED 2].crt tls_key_file: /var/www/certs/[REDACTED 1].key imap_tls_key_file: /var/www/certs/[REDACTED 2].key There are seperate files for the cert and key as well as my CA (GoDaddy... can't go wrong since they have a sale!) (The above /var/www directories are not actually accessible to the web server. I just happened to store my web certs in there as well, so why not use the same directory for mail certs?) The certificate has the domain as well as the FQDN of the server specified. It's interesting to add that even when I get this error, the connection will still succeed encrypted. A google search of this issue indicates that it is more or less not commonly encountered. From marty at maui-systems.co.uk Wed Jul 17 09:33:12 2013 From: marty at maui-systems.co.uk (Marty Lee) Date: Wed, 17 Jul 2013 14:33:12 +0100 Subject: Cyrus+CalDAV Message-ID: Hi, I've been playing with the latest Cyrus beta which includes the CalDAV & CardDAV additions - from a personal perspective, almost all seems ok. Server is a Solaris 10 (x86) box; clients are mainly Mac OSX (Mountain Lion) and some PCs (Thunderbird/Lightning). One question that Ken or someone may already know and one issue that I need to track down further. The question first: I've got two users that have permission to read each others Default calendar (lr9) - but I'm guessing that the list of calendars returned to the Mac calendar app only includes calendars for the actual user, not shared ones, as the shared calendars can't be seen? does this sound right, or should I be able to see the shared calendars (or need to do something to make it work)? I've also seem similar with the CardDAV interface - I use a DAV client to pull down all my contacts and put them into a local LDAP server for address book lookups for a number of other apps. This works if I use my username+password, but not if I use a different account with permissions to read my Default address book (lr). The issue I've seen revolves around adding pictures to vCards - some existing cards have pictures (copied from existing Mac address book), but changing pictures or adding new cards with photos seems to cause problems - I suspect it's 'segfaulting' the server process, but I'm not 100% certain of that yet, so I won't log a bug just yet? Anyone else tried any of these scenario's and able to say whether they've had success or not - maybe I'm just too bleeding edge and dive into the code myself (which I'll do anyway, I just don't want to spend time doing something someone has already worked out!). Cheers Marty From murch at andrew.cmu.edu Thu Jul 18 14:59:44 2013 From: murch at andrew.cmu.edu (Ken Murchison) Date: Thu, 18 Jul 2013 14:59:44 -0400 Subject: Cyrus+CalDAV In-Reply-To: References: Message-ID: <51E83B20.6040906@andrew.cmu.edu> Hi Marty, On 07/17/2013 09:33 AM, Marty Lee wrote: > Hi, > > I've been playing with the latest Cyrus beta which includes the CalDAV & CardDAV > additions - from a personal perspective, almost all seems ok. > > Server is a Solaris 10 (x86) box; clients are mainly Mac OSX (Mountain Lion) and > some PCs (Thunderbird/Lightning). > > One question that Ken or someone may already know and one issue that I need to track > down further. > > The question first: I've got two users that have permission to read each others > Default calendar (lr9) - but I'm guessing that the list of calendars returned to > the Mac calendar app only includes calendars for the actual user, not shared ones, > as the shared calendars can't be seen? does this sound right, or should I be able > to see the shared calendars (or need to do something to make it work)? Even though you set the ACLs properly, this won't help you by itself. There is a separate CALDAV sharing spec which essentially makes shared calendars appear under a user's own calendar-home-set. Unfortunately, Cyrus doesn't support this spec yet. I don't know if there are other ways for a user to access someone else's calendars. > > I've also seem similar with the CardDAV interface - I use a DAV client to pull down > all my contacts and put them into a local LDAP server for address book lookups for > a number of other apps. This works if I use my username+password, but not if I use a different account with permissions to read my Default address book (lr). There has been talk of a sharing spec for CardDAV, but I don't believe any work has been done on it yet. > > The issue I've seen revolves around adding pictures to vCards - some existing cards have pictures (copied from existing Mac address book), but changing pictures or adding new cards with photos seems to cause problems - I suspect it's 'segfaulting' the server process, but I'm not 100% certain of that yet, so I won't log a bug just yet? Hmm. Which client are you using to edit the vCard? I know that I used the CardDAVMate javascript client to add a picture and it worked. But that's not to say that I don't have a bug somewhere. Thanks for the feedback! -- Kenneth Murchison Principal Systems Software Engineer Carnegie Mellon University From murch at andrew.cmu.edu Thu Jul 18 16:12:08 2013 From: murch at andrew.cmu.edu (Ken Murchison) Date: Thu, 18 Jul 2013 16:12:08 -0400 Subject: Cyrus+CalDAV In-Reply-To: <51E83B20.6040906@andrew.cmu.edu> References: <51E83B20.6040906@andrew.cmu.edu> Message-ID: <51E84C18.2070209@andrew.cmu.edu> On 07/18/2013 02:59 PM, Ken Murchison wrote: >> The issue I've seen revolves around adding pictures to vCards - some existing cards have pictures (copied from existing Mac address book), but changing pictures or adding new cards with photos seems to cause problems - I suspect it's 'segfaulting' the server process, but I'm not 100% certain of that yet, so I won't log a bug just yet? > Hmm. Which client are you using to edit the vCard? I know that I used > the CardDAVMate javascript client to add a picture and it worked. But > that's not to say that I don't have a bug somewhere. I did some more testing with Apple Contacts and it looks like the vCard parser that I'm using doesn't like the data. The server doesn't segfault, its returns a 403 (Forbidden) with the precondition. I'm looking into this. -- Kenneth Murchison Principal Systems Software Engineer Carnegie Mellon University From robn at fastmail.fm Sun Jul 21 20:53:49 2013 From: robn at fastmail.fm (Robert Norris) Date: Mon, 22 Jul 2013 10:53:49 +1000 Subject: CalDAV and virtual domains not working Message-ID: <20130722005349.GA7086@pyro.melbourne.osa> It looks like the CalDAV stuff is not doing the right thing with virtual domains (beta6 and git e415f906) 2013-07-21T20:44:55.580301-04:00 calendar1 calendar/http[30755]: login: vpn94.mail.srv.osa [10.203.0.94] robn at fastmail.fm Basic User logged in 2013-07-21T20:44:55.580929-04:00 calendar1 calendar/http[30755]: mlookup(user.robn at fastmail^fm.#calendars.Default) failed: Mailbox does not exist 2013-07-21T20:44:55.581141-04:00 calendar1 calendar/http[30755]: vpn94.mail.srv.osa [10.203.0.94] as "robn at fastmail.fm" with "Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130630 Icedove/17.0.7 Lightning/1.9b1"; "PROPFIND /dav/calendars/user/robn%40fastmail.fm/Default/ HTTP/1.1" (depth=0) => "404 Not Found" (error=Mailbox does not exist) Mailbox structure via IMAP is: . LIST "" * * LIST (\HasChildren) "." user.robn at fastmail.fm * LIST (\HasChildren) "." user.robn.#calendars at fastmail.fm * LIST (\HasNoChildren) "." user.robn.#calendars.Default at fastmail.fm * LIST (\HasNoChildren) "." user.robn.#calendars.Inbox at fastmail.fm * LIST (\HasNoChildren) "." user.robn.#calendars.Outbox at fastmail.fm Bron tells me (if I understood him correctly) that the domain splitting is broken, and it should be producing eg "fastmail.fm!user.robn.#calendars.Default" instead. Cheers, Rob N. From murch at andrew.cmu.edu Mon Jul 22 15:58:28 2013 From: murch at andrew.cmu.edu (Ken Murchison) Date: Mon, 22 Jul 2013 15:58:28 -0400 Subject: CalDAV and virtual domains not working In-Reply-To: <20130722005349.GA7086@pyro.melbourne.osa> References: <20130722005349.GA7086@pyro.melbourne.osa> Message-ID: <51ED8EE4.2000309@andrew.cmu.edu> Hi Robert, Bron is correct. The code isn't doing the right thing for virtdomains, mainly because I hadn't bothered to add support for it as I was coding. Obviously, this needs to be fixed. I guess the big question is what the URLs should look like. Should we go with the current URL format and encode the '@' as your log entry shows below, or should we do something like: /dav/calendars/domain/fastmail.fm/user/robn/Default/ I think I prefer /dav/calendars/user/robn%40fastmail.fm/Default/ as long as clients support it. Bron, any thoughts? On 07/21/2013 08:53 PM, Robert Norris wrote: > It looks like the CalDAV stuff is not doing the right thing with virtual > domains (beta6 and git e415f906) > > 2013-07-21T20:44:55.580301-04:00 calendar1 calendar/http[30755]: login: vpn94.mail.srv.osa [10.203.0.94] robn at fastmail.fm Basic User logged in > 2013-07-21T20:44:55.580929-04:00 calendar1 calendar/http[30755]: mlookup(user.robn at fastmail^fm.#calendars.Default) failed: Mailbox does not exist > 2013-07-21T20:44:55.581141-04:00 calendar1 calendar/http[30755]: vpn94.mail.srv.osa [10.203.0.94] as "robn at fastmail.fm" with "Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130630 Icedove/17.0.7 Lightning/1.9b1"; "PROPFIND /dav/calendars/user/robn%40fastmail.fm/Default/ HTTP/1.1" (depth=0) => "404 Not Found" (error=Mailbox does not exist) > > Mailbox structure via IMAP is: > > . LIST "" * > * LIST (\HasChildren) "." user.robn at fastmail.fm > * LIST (\HasChildren) "." user.robn.#calendars at fastmail.fm > * LIST (\HasNoChildren) "." user.robn.#calendars.Default at fastmail.fm > * LIST (\HasNoChildren) "." user.robn.#calendars.Inbox at fastmail.fm > * LIST (\HasNoChildren) "." user.robn.#calendars.Outbox at fastmail.fm > > Bron tells me (if I understood him correctly) that the domain splitting > is broken, and it should be producing eg > "fastmail.fm!user.robn.#calendars.Default" instead. > > Cheers, > Rob N. > ---- > Cyrus Home Page: http://www.cyrusimap.org/ > List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/ > To Unsubscribe: > https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus -- Kenneth Murchison Principal Systems Software Engineer Carnegie Mellon University From robn at fastmail.fm Tue Jul 23 01:31:41 2013 From: robn at fastmail.fm (Robert Norris) Date: Tue, 23 Jul 2013 15:31:41 +1000 Subject: CalDAV and virtual domains not working In-Reply-To: <51ED8EE4.2000309@andrew.cmu.edu> References: <20130722005349.GA7086@pyro.melbourne.osa> <51ED8EE4.2000309@andrew.cmu.edu> Message-ID: <1374557501.1399.461459.68AB151C@webmail.messagingengine.com> On Tue, Jul 23, 2013, at 05:58 AM, Ken Murchison wrote: > I think I prefer > > /dav/calendars/user/robn%40fastmail.fm/Default/ > > as long as clients support it. Yeah, we like this one. Its much clearer for the user. Can't think of any reason that clients wouldn't support it - URI escaping is hardly new. Unless calendar clients are especially braindead? :) Cheers, Rob N. From buecker at phil.hhu.de Tue Jul 23 10:33:12 2013 From: buecker at phil.hhu.de (=?ISO-8859-1?Q?Peter_B=FCcker?=) Date: Tue, 23 Jul 2013 16:33:12 +0200 Subject: Moving mailboxes in Cyrus Murder leaves mails behind on source server Message-ID: <51EE9428.40707@phil.hhu.de> Hi all, We're having an issue with our Cyrus Murder environment when moving mailboxes from one backend to another backend. Some mails are left behind on the source server while we think they should be removed. We issue the rename by using "cyradm" on the source server with the following command: oldserver> rename user.foo user.foo newserver!default Cyrus completes the migration without any error messages, and all mails are successfully copied to the new backend. However, some mails are left behind in the spool directory of the old backend server. According to section "Moving mailboxes between backends", seen state should be the only thing left behind on the source server. The mailbox database is updated correctly, e.g. the old backend now longer believes to own the affected mailboxes. It happens to be that only complete mailboxes are left behind. For example, the whole INBOX might be left behind, and maybe also some subfolders. But if a folder is left behind, it's always completely left behind, not just partially. Is this expected behavior or are we doing something wrong? Please note that the affected user was logged in via IMAP during the migration. All servers are running vanilla Cyrus 2.4.17 compiled from source, running on Debian Squeeze or Debian Wheezy. All backend servers just have one partition. The backend configuration looks like this: ------------------------------------------------------ configdirectory: /var/lib/cyrus-imapd/imap defaultpartition: default partition-default: /var/spool/cyrus-imapd admins: cyrus-admin cyrus-murder allowallsubscribe: 1 expunge_mode: delayed delete_mode: delayed allowanonymouslogin: no hashimapspool: 1 duplicatesuppression: 0 sendmail: /usr/sbin/sendmail unix_group_enable: 0 imapidresponse: 0 lmtp_downcase_rcpt: 1 sievedir: /var/lib/cyrus-imapd/sieve sieve_maxscriptsize: 32 sieve_maxscripts: 5 sasl_mech_list: plain login sasl_pwcheck_method: saslauthd allowapop: 0 allowplaintext: 1 sync_log: 1 sync_authname: cyrus-admin sync_password: xxxxx sync_host: cyrus-replica sync_repeat_interval: 5 mupdate_server: cyrus-mupdate mupdate_authname: cyrus-murder mupdate_username: cyrus-murder mupdate_password: xxxxx proxy_authname: cyrus-murder proxy_password: xxxxx proxyservers: cyrus-murder allowusermoves: 1 disconnect_on_vanished_mailbox: 1 ------------------------------------------------------ Thanks for any support :-) Regards, Peter From stefan at schloerholz.de Wed Jul 24 17:13:37 2013 From: stefan at schloerholz.de (=?UTF-8?Q?=22Stefan_Schl=C3=B6rholz=22?=) Date: Wed, 24 Jul 2013 23:13:37 +0200 (CEST) Subject: Aw: Re: In-Reply-To: <255779d6116aad3b5078af26153de543.squirrel@webmail.bi.corp.invoca.ch> References: , <255779d6116aad3b5078af26153de543.squirrel@webmail.bi.corp.invoca.ch> Message-ID: Hello Simon, > Did you try running "reconstruct -r -f ..."? I did try to run "reconstruct -r user.paul". The -f switch is not known/accepted by my cyradm. > BTW, could you change your mailer so you don't send HTML only? Sorry for that. My whole mail system is down so I have to use a web mailer where I have to set that option individually for each mail. I cleaned my original post for better reading. Stefan >> I had a cyrus installation running that worked perfectly. After a >> crash of a hard drive I had to reinstall the system. The hard drive >> contained the cyrus installation. However, the cyrus >> "partition-default" was on a different hard drive. >> >> I reinstalled cyrus with the "partition-default" as >> before. The user mailbox can be seen in cyradmin [user.paul >> (\HasChildren)]. However all subfolders exept user.paul.Drafts, >> user.paul.M?lleimer, user.paul.Versendete_Nachrichten can NOT be >> seen. The directories and the messages in the storage partion under >> /var/spool/imap/user/paul/ are all there. >> >> With a squirrelmail I do see all messages in INBOX and I do see the >> three subfolders. However, the subfolders are being shown to be empty. The >> files in the corresponding directories exist. >> >> I did a reconstruct -r with the cyrus admin allready, however I still >> would have the data as they were after the hard drive crash. >> >> Is there anybody who can help to get the old subfolders and messages >> back? From morgan at orst.edu Wed Jul 24 17:29:52 2013 From: morgan at orst.edu (Andrew Morgan) Date: Wed, 24 Jul 2013 14:29:52 -0700 (PDT) Subject: Aw: Re: In-Reply-To: References: , <255779d6116aad3b5078af26153de543.squirrel@webmail.bi.corp.invoca.ch> Message-ID: On Wed, 24 Jul 2013, "Stefan Schl?rholz" wrote: > Hello Simon, > >> Did you try running "reconstruct -r -f ..."? > > I did try to run "reconstruct -r user.paul". The -f switch is not > known/accepted by my cyradm. Use the command-line program named "reconstruct", not the command "reconstruct" inside of cyradm. Andy From simon at leaky.org Wed Jul 24 17:30:25 2013 From: simon at leaky.org (Simon Amor) Date: Wed, 24 Jul 2013 22:30:25 +0100 Subject: In-Reply-To: References: , <255779d6116aad3b5078af26153de543.squirrel@webmail.bi.corp.invoca.ch> Message-ID: <11606_1374701427_r6OLUQZ5021631_B10017FC-8A89-4F36-A400-D409E088AF8C@leaky.org> Hi Stefan, On 24 Jul 2013, at 22:13, "Stefan Schl?rholz" wrote: > Hello Simon, > >> Did you try running "reconstruct -r -f ..."? > > I did try to run "reconstruct -r user.paul". The -f switch is not known/accepted by my cyradm. Use the binary not the cyradm command. Depending on the system, it could be somewhere like /usr/lib/cyrus-imapd/reconstruct It's in the same place as imapd, lmtpd, and cyrus-master live (at least on CentOS) so you might be able to locate it using something like this command: ps ax | grep cyrus-master 32268 ? Ss 0:07 /usr/lib/cyrus-imapd/cyrus-master -d The -f argument to reconstruct according to the man page: Examine the filesystem underneath mailbox, adding all directories with a cyrus.header found there as new mailboxes. Useful for restoring mailboxes from backups. Regards, Simon (a different one) -- Simon Amor simon at leaky.org From Hagedorn at uni-koeln.de Thu Jul 25 11:51:36 2013 From: Hagedorn at uni-koeln.de (Sebastian Hagedorn) Date: Thu, 25 Jul 2013 17:51:36 +0200 Subject: Remove quota using Cyrus::IMAP::Admin? Message-ID: <6E66FB5AE9A68EE05DA4A319@tyrion.rrz.uni-koeln.de> Hi, you can remove the quota for a mailbox with cyradm by setting the quota to none: $ cyradm cyrus Password: server> sq user/xxx none remove quota server> But when I try to do the same using Cyrus::IMAP::Admin I get an error, because a number is expected as argument. Is that a bug or a feature? Here's my code: $result = $imapcon->setquota("user/$user", "STORAGE", $quota); If I set $quota to 'none', I get: Error: xxx, STORAGE: none: not a number -- .:.Sebastian Hagedorn - Weyertal 121 (Geb?ude 133), Zimmer 2.02.:. .:.Regionales Rechenzentrum (RRZK).:. .:.Universit?t zu K?ln / Cologne University - ? +49-221-470-89578.:. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pkcs7-signature Size: 5313 bytes Desc: not available Url : http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20130725/3efc05fb/attachment.bin From nic at onlight.com Thu Jul 25 12:13:34 2013 From: nic at onlight.com (Nic Bernstein) Date: Thu, 25 Jul 2013 11:13:34 -0500 Subject: Remove quota using Cyrus::IMAP::Admin? In-Reply-To: <6E66FB5AE9A68EE05DA4A319@tyrion.rrz.uni-koeln.de> References: <6E66FB5AE9A68EE05DA4A319@tyrion.rrz.uni-koeln.de> Message-ID: <51F14EAE.1030206@onlight.com> Sebastian, Please take a look at this thread from January of last year, which discusses the issue in full. http://comments.gmane.org/gmane.mail.imap.cyrus/36194 -nic On 07/25/2013 10:51 AM, Sebastian Hagedorn wrote: > Hi, > > you can remove the quota for a mailbox with cyradm by setting the > quota to none: > > $ cyradm cyrus > Password: > server> sq user/xxx none > remove quota > server> > > But when I try to do the same using Cyrus::IMAP::Admin I get an error, > because a number is expected as argument. Is that a bug or a feature? > Here's my code: > > $result = $imapcon->setquota("user/$user", "STORAGE", $quota); > > If I set $quota to 'none', I get: > > Error: xxx, STORAGE: none: not a number > > > ---- > Cyrus Home Page: http://www.cyrusimap.org/ > List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/ > To Unsubscribe: > https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus -- Nic Bernstein nic at onlight.com Onlight, Inc. www.onlight.com 219 N. Milwaukee St., Suite 2a v. 414.272.4477 Milwaukee, Wisconsin 53202 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20130725/9758085e/attachment.html From Hagedorn at uni-koeln.de Thu Jul 25 13:17:24 2013 From: Hagedorn at uni-koeln.de (Sebastian Hagedorn) Date: Thu, 25 Jul 2013 19:17:24 +0200 Subject: Remove quota using Cyrus::IMAP::Admin? In-Reply-To: <51F14EAE.1030206@onlight.com> References: <6E66FB5AE9A68EE05DA4A319@tyrion.rrz.uni-koeln.de> <51F14EAE.1030206@onlight.com> Message-ID: <01B33A898A82EEC1E7C3ADA2@vpn83-230.vpn.uni-koeln.de> Thanks Nic, I should've checked the archives first :( -- Nic Bernstein is rumored to have mumbled on 25. Juli 2013 11:13:34 -0500 regarding Re: Remove quota using Cyrus::IMAP::Admin?: > Sebastian, > Please take a look at this thread from January of last year, which > discusses the issue in full. > > http://comments.gmane.org/gmane.mail.imap.cyrus/36194 -- Sebastian Hagedorn - Weyertal 121, Zimmer 2.02 Regionales Rechenzentrum (RRZK) Universit?t zu K?ln / Cologne University - Tel. +49-221-470-89578 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pkcs7-signature Size: 5313 bytes Desc: not available Url : http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20130725/f9db64a0/attachment.bin From ka-el at laposte.net Fri Jul 26 08:11:54 2013 From: ka-el at laposte.net (kael) Date: Fri, 26 Jul 2013 14:11:54 +0200 Subject: Cyrus version with METADATA support Message-ID: <51F2678A.8020807@laposte.net> Hello, I've installed Cyrus 2.4.17 but can't find METADATA support in the capabilities, only ANNOTATEMORE is displayed. Is there a specific version somewhere in git for using METADATA or did I miss something in the configuration ? -- kael From ka-el at laposte.net Fri Jul 26 08:12:46 2013 From: ka-el at laposte.net (kael) Date: Fri, 26 Jul 2013 14:12:46 +0200 Subject: Cyrus version with METADATA support Message-ID: Hello, I've installed Cyrus 2.4.17 but can't find METADATA support in the capabilities, only ANNOTATEMORE is displayed. Is there a specific version somewhere in git for using METADATA or did I miss something in the configuration ? -- kael From jpb at bordengrammar.kent.sch.uk Wed Jul 31 06:05:04 2013 From: jpb at bordengrammar.kent.sch.uk (Julian Pilfold-Bagwell) Date: Wed, 31 Jul 2013 11:05:04 +0100 Subject: Moving to dotted usernames Message-ID: <51F8E150.5010008@bordengrammar.kent.sch.uk> Hi list, I have a running cyrus server that I need to switch to unixhierarchysep as I'm being asked to change our mail scheme to include full stops. The authentication backend for this is sasldb2. As I understand it from the documentation, setting "unixhierarchysep 1" in imapd.conf should be enough as the backend handling of folders stays the same but cyrus translates the . to ^ . I've backed up the mailboxes and the databases and set the 1 in imap.conf but when restarting, cyrus fails on importing cyrus-imapd databases. I've searched for solutions but the suggestions always seem to be that there will be no problem. Is there anything I'm missing? Thanks, Julian -- Borden Grammar School, Avenue of Remembrance, Sittingbourne, Kent, ME10 4DB. Tel: 01795 424192 **************************************************************************** This e-mail is from Borden Grammar School Trust. This e-mail, together with any files transmitted with it, are confidential, and are intended solely for the use of the individual or entity to whom they are addressed. Any unauthorised dissemination or copying of this e-mail or its attachments, and any use or disclosure of any information contained in them, is strictly prohibited, and may also be illegal. If you are not the intended recipient you must not use, disclose, distribute, copy, print or relay this e-mail. Please note that any views expressed by an individual within this e-mail, do not necessarily reflect the views of the Borden Grammar School Trust. Borden Grammar School Trust has taken reasonable precautions to ensure no viruses are present in this e-mail, the Academy cannot accept responsibility for any loss or damage arising from the use of this e-mail and/or files attached. Registered office: Borden Grammar School, Avenue of Remembrance, Sittingbourne, Kent, ME10 4DB Registered in England: 07827591 From magiza83 at hotmail.com Wed Jul 31 09:34:19 2013 From: magiza83 at hotmail.com (=?iso-8859-1?B?TWFuZWwgR2ltZW5vIFphcmFnb3rh?=) Date: Wed, 31 Jul 2013 15:34:19 +0200 Subject: digest.sha1 cyrus-imapd 2.5 problem Message-ID: Hello, i'm trying to apply the patch digest.sha1 to cyrus-imapd-2.5 kolab version You can find it in pastebin: http://pastebin.com/raw.php?i=kXKKGbdu But when I try to compile it, I'm getting the following error: imap/index.c: In function 'index_fetchreply': imap/index.c:2981: error: 'struct index_map' has no member named 'record' imap/index.c:3006: error: 'struct index_map' has no member named 'record' make[2]: *** [imap/imap_libcyrus_imap_la-index.lo] Error 1 make[2]: Leaving directory `/root/rpmbuild/BUILD/cyrus-imapd-2.5' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/root/rpmbuild/BUILD/cyrus-imapd-2.5' make: *** [all] Error 2 error: Bad exit status from /var/tmp/rpm-tmp.nN45fY (%build) The sources have been downloaded from: http://mirror.kolabsys.com/pub/redhat/kolab-3.1/el6/development/SRPMS/cyrus-imapd-2.5-0.1.dev6.git534066.el6.kolab_3.1.src.rpm Could anyone help me with this? This same question has been asked in kolab-dev maillist. Thanks! Manel -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20130731/c81ba679/attachment.html