Re: Test version of the cyrus-imapd-2.4.6 Invoca RPM package
CONDSTORE is compulsary. It's always switched on. You have no choice. Welcome to the brave new world. Ops, thank you very much. So it could be useful remove 'condstore' from hints of mboxconfig: localhost.localdomain mboxcfg usage: mboxconfig mailbox [comment|condstore|expire|news2mail|sharedseen|sieve|squat|/explicit annotation] value name : Cyrus IMAPD version: v2.4.6-Invoca-RPM-2.4.6-0.9 35e0e72f 2010-12-21 Bye Marco Cyrus Home Page: http://www.cyrusimap.org/ List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
Re: Test version of the cyrus-imapd-2.4.6 Invoca RPM package
On Tue, Jan 18, 2011 at 11:38:51AM +0100, Marco wrote: CONDSTORE is compulsary. It's always switched on. You have no choice. Welcome to the brave new world. Ops, thank you very much. So it could be useful remove 'condstore' from hints of mboxconfig: localhost.localdomain mboxcfg usage: mboxconfig mailbox [comment|condstore|expire|news2mail|sharedseen|sieve|squat|/explicit annotation] value name : Cyrus IMAPD version: v2.4.6-Invoca-RPM-2.4.6-0.9 35e0e72f 2010-12-21 That's a very good point :) Can you make a bugzilla entry for this please (so you get the credit, and so we can track it into the next stable release) Thanks, Bron. Cyrus Home Page: http://www.cyrusimap.org/ List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
Documentation [Was Shared mailboxes doc]
On Fri, 2011-01-07 at 17:45 +, Jeroen van Meeuwen (Kolab Systems) wrote: Dan White wrote: I have a git repository at: http://git.cyrusimap.org/cyrus-imapd-docs/ and some output I regulary generate on: http://www.cyrusimap.org/~vanmeeuwen/cyrus-imapd-2.4-docs/ Really, at this point, *any* feedback is most appreciated ;-)) What are your plans with this documentation? Are you intending to use this build system to generate the contents of the /doc documentation? Assuming I'm understanding the question correctly; Ny intention is to make it easier to provide *more* content then is suitable for a man-page, or the doc/ directory, in a way that enables us to spit out the documentation in different formats, versioned and all. I added the following regarding the condstore annotation to my WMOGAG Cyrus section: QUOTE This annotation is only supported in the 2.3.x release series starting with 2.3.3 although its use is not recommended until 2.3.8. As of the 2.4.x release series CONDSTORE functionality is enabled on all mailboxes regardless of annotation and attempting to set this annotation will result in a permission denied message. On releases where this annotation is supported setting a value of “true” will enable CONDSTORE functionality [1]. [footnote:1] The CONDSTORE functionality is defined in RFC4551, http://tools.ietf.org/html/rfc4551 /QUOTE It might be nice to have that in the 'official' (?) documentation. I put this in the Administration/cyradm/mboxcfg section where the useful annotations are listed. jmeeuwen: If I create a documentation section or update something do you want me to ping you via private e-mail, on the -devel list, or not-at-all? Cyrus Home Page: http://www.cyrusimap.org/ List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
Re: Test version of the cyrus-imapd-2.4.6 Invoca RPM package
On Tue, Jan 18, 2011 at 11:38:51AM +0100, Marco wrote: CONDSTORE is compulsary. It's always switched on. You have no choice. Welcome to the brave new world. Ops, thank you very much. So it could be useful remove 'condstore' from hints of mboxconfig: localhost.localdomain mboxcfg usage: mboxconfig mailbox [comment|condstore|expire|news2mail|sharedseen|sieve|squat|/explicit annotation] value name : Cyrus IMAPD version: v2.4.6-Invoca-RPM-2.4.6-0.9 35e0e72f 2010-12-21 That's a very good point :) Can you make a bugzilla entry for this please (so you get the credit, and so we can track it into the next stable release) I've attached my version of a patch to the bugzilla entry. It is also in the just released 2.4.6-1 RPM. Simon Cyrus Home Page: http://www.cyrusimap.org/ List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
deleting old trash mails
Hello list Many of my clients are use imap accounts and they usually delete mails but don't purge them. Most of them use SquirrelMail, others Thunderbird. So I'm trying to make Cyrus to delete automatically after some days. I use cyrus-imapd-2.3.16-3.fc12.i686. In the imap.conf I put: ... expunge_mode: delayed ... And in the cyrus.conf: ... delprune cmd=cyr_expire -a -E 3 -X 10 at=0400 ... In maillog everyday I found the following records: ... Jan 17 04:00:02 ns1 cyr_expire[12994]: mydelete: starting txn 2147487755 Jan 17 04:00:02 ns1 cyr_expire[12994]: mydelete: committing txn 2147487755 Jan 17 04:00:02 ns1 cyr_expire[12994]: duplicate_prune: purged 284 out of 786 entries but no mails deleted from trash. I believe there is some misunderstanding here. What I' doing wrong? Thank's in advance Nikos Cyrus Home Page: http://www.cyrusimap.org/ List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
Re: deleting old trash mails
On 18/01/11 14:01 +0200, Nikos Gatsis - Qbit wrote: Hello list Many of my clients are use imap accounts and they usually delete mails but don't purge them. Most of them use SquirrelMail, others Thunderbird. So I'm trying to make Cyrus to delete automatically after some days. I use cyrus-imapd-2.3.16-3.fc12.i686. In the imap.conf I put: ... expunge_mode: delayed ... And in the cyrus.conf: ... delprune cmd=cyr_expire -a -E 3 -X 10 at=0400 ... In maillog everyday I found the following records: ... Jan 17 04:00:02 ns1 cyr_expire[12994]: mydelete: starting txn 2147487755 Jan 17 04:00:02 ns1 cyr_expire[12994]: mydelete: committing txn 2147487755 Jan 17 04:00:02 ns1 cyr_expire[12994]: duplicate_prune: purged 284 out of 786 entries but no mails deleted from trash. Cyrus does not have any inherent understanding of what the Trash folder is... it's just another IMAP folder. 'Deleted' in the IMAP sense refers to an email that has the '\Deleted' flag set. A subsequent EXPUNGE will remove the emails from user view altogether. Since the emails that are contained within the Trash folder probably do not have the '\Deleted' flag set, they will not be expunged automatically by Cyrus. You could use ipurge to automatically delete messages from your users' Trash folders. Adding something like this to the EVENTS section of your cyrus.conf should work (please test as a typo could lead to loss of data): purgetrash cmd=/usr/sbin/ipurge -f -d 14 */Trash at=0301 Which purges all messages older than 14 days, in all users' Trash folders and runs every morning at 3:01am. See the man pages for ipurge and cyrus.conf for more details. -- Dan White Cyrus Home Page: http://www.cyrusimap.org/ List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
Cyrus 2.4.6: RENAME/XFER problems
Hello list,
I've tried to test Cyrus-imapd 2.4.6 (v2.4.6-Invoca-RPM-2.4.6-0.9,
thanks, Simon!) in a test environment with no problems until I tried
moving user mailboxes to different partitions.
In 2.3.16, the procedure works almost but replication is broken if you
move a user to a different spool on a same backend (but works when
using different backend).
In 2.4.6, every renm/xfer command I can come up with in cyradm fails
somehow. Below is my test mailbox and failed commands attached with
their respective log entries. In particular, the Mailbox name too
long error msg is a bit funny. Why is it trying to find
socket/mupdate.target in the first place as I'm using standard murder?
Now, what do you suggest? How should I move half of my ~5 user
mailboxes to 24 newly created 500G partitions? Is there a bug in my
syntax/configuration or in Cyrus/cyradm?
BTW, according to the cyradm's help, one shouldn't be able to give the
backend hostname at all, see:
cyradm help renm
renm, alias for renamemailbox [--partition partition] oldname newname
[partition]
rename (and optionally relocate) mailbox
That is somewhat contradictory to what we see below (getaddrinfo).
Here's some info about our (test) config:
- mupdate_config: standard
- frontend: m2v1t.mappi.helsinki.fi
- mupdate master: lcluster.mappi.helsinki.fi
- backends: m2v2t.mappi.helsinki.fi m2v3t.mappi.helsinki.fi
- spool partitions: m2v2t, m2v2t-b m2v3t, m2v3t-b accordingly
- running on clustered CentOS 5.5
m2v1t.mappi.helsinki.fi info user.jtest400
{user.jtest400}:
partition: m2v3t
server: m2v3t.mappi.helsinki.fi
[...]
m2v1t.mappi.helsinki.fi renm user.jtest400 user.jtest400
m2v2t.mappi.helsinki.fi
renamemailbox: The remote Server(s) denied the operation
Jan 18 12:30:16 m2cn1t m2v2t/imap[23386]: kick_mupdate: can't connect to
target: No such file or directory
Jan 18 12:30:16 m2cn1t m2v2t/imap[23386]: IOERROR: Mailbox name too long
(user.jtest400)
Jan 18 12:30:16 m2cn1t m2v3t/imap[23784]: Could not move mailbox:
user.jtest400, LOCALCREATE failed
m2v3t.mappi.helsinki.fi renm user.jtest400 user.jtest400
m2v2t.mappi.helsinki.fi
renamemailbox: The remote Server(s) denied the operation
Jan 18 12:37:31 m2cn1t m2v2t/imap[23392]: kick_mupdate: can't connect to
target: No such file or directory
Jan 18 12:37:31 m2cn1t m2v2t/imap[23392]: IOERROR: Mailbox name too long
(user.jtest400)
Jan 18 12:37:31 m2cn1t m2v3t/imap[23876]: Could not move mailbox:
user.jtest400, LOCALCREATE failed
m2v3t.mappi.helsinki.fi renm user.jtest400 user.jtest400 m2v3t-2
renamemailbox: Server(s) unavailable to complete operation
Jan 18 12:38:53 m2cn1t m2v3t/imap[23876]: getaddrinfo(m2v3t-2) failed: Name or
service not known
m2v3t.mappi.helsinki.fi renm user.jtest400 user.jtest400 m2v3t-2
renamemailbox: Server(s) unavailable to complete operation
Jan 18 12:38:53 m2cn1t m2v3t/imap[23876]: getaddrinfo(m2v3t-2) failed: Name or
service not known
m2v3t.mappi.helsinki.fi renm user.jtest400 user.jtest400
m2v3t.mappi.helsinki.fi!m2v3t-2
renamemailbox: Invalid server requested
m2v3t.mappi.helsinki.fi xfer user.jtest400 m2v3t.mappi.helsinki.fi m2v3t-2
xfermailbox: Invalid server requested
m2v3t.mappi.helsinki.fi xfer user.jtest400 m2v3t.mappi.helsinki.fi!m2v3t-2
xfermailbox: Server(s) unavailable to complete operation
Jan 18 12:42:38 m2cn1t m2v3t/imap[23876]:
getaddrinfo(m2v3t.mappi.helsinki.fi!m2v3t-2) failed: Name or service not known
m2v3t.mappi.helsinki.fi xfer user.jtest400 m2v3t
xfermailbox: The remote Server(s) denied the operation
Jan 18 12:43:28 m2cn1t m2v3t/imap[23876]: Could not move mailbox:
user.jtest400, LOCALCREATE failed
m2v3t.mappi.helsinki.fi xfer user.jtest400 m2v2t-2
xfermailbox: Server(s) unavailable to complete operation
Jan 18 12:44:13 m2cn1t m2v3t/imap[23876]: getaddrinfo(m2v2t-2) failed: Name or
service not known
m2v3t.mappi.helsinki.fi xfer user.jtest400 m2v2t.mappi.helsinki.fi!m2v2t-2
xfermailbox: Server(s) unavailable to complete operation
Jan 18 12:44:25 m2cn1t m2v3t/imap[23876]:
getaddrinfo(m2v2t.mappi.helsinki.fi!m2v2t-2) failed: Name or service not known
m2v3t.mappi.helsinki.fi xfer user.jtest400 m2v2t
xfermailbox: The remote Server(s) denied the operation
Jan 18 12:44:07 m2cn1t m2v2t/imap[23411]: kick_mupdate: can't connect to
target: No such file or directory
Jan 18 12:44:07 m2cn1t m2v2t/imap[23411]: IOERROR: Mailbox name too long
(user.jtest400)
Jan 18 12:44:07 m2cn1t m2v3t/imap[23876]: Could not move mailbox:
user.jtest400, LOCALCREATE failed
m2v1t.mappi.helsinki.fi xfer user.jtest400 m2v3t.mappi.helsinki.fi m2v3t-2
xfermailbox: bad parameters to function
Jan 18 12:47:32 m2cn3t im/mupdate[16394]: accepted connection
Jan 18 12:47:32 m2cn3t im/mupdate[16394]: login: lcluster.mappi.helsinki.fi
[...] User logged in
Jan 18 12:47:32 m2cn3t m2v1t/imapm2v1t[9845]: MUPDATE: lost connection, retrying
Jan 18 12:47:32 m2cn3t im/mupdate[16394]: accepted connection
Jan 18 12:47:32 m2cn3t im/mupdate[16394]: login:
Some advantage in prefork several imapds in startup ?
In tests, i configure /etc/cyrus.conf to start 5000 process in startup.
SERVICES {
imapcmd=imapdlisten=imap prefork=5000
}
Consulting the system:
imap_server:/var/lib/imap/procps axwww | grep imapd | wc -l
5177
But in: ls -l /var/lib/imap/proc | wc -l i see only
173
Why, the master start new process and don't use prefork process ?
Exist any advantage startup with several process ?
Regards
Zinato
Cyrus Home Page: http://www.cyrusimap.org/
List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
Re: Test version of the cyrus-imapd-2.4.6 Invoca RPM package
On Tue, Jan 18, 2011 at 02:59:53PM +0100, Simon Matter wrote: I've attached my version of a patch to the bugzilla entry. It is also in the just released 2.4.6-1 RPM. Woot :) Have you considered learning a bit of git and making patches directly? Bron ( how lazy can I get anyway??? ) Cyrus Home Page: http://www.cyrusimap.org/ List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
Re: Test version of the cyrus-imapd-2.4.6 Invoca RPM package
On Tue, Jan 18, 2011 at 02:59:53PM +0100, Simon Matter wrote: I've attached my version of a patch to the bugzilla entry. It is also in the just released 2.4.6-1 RPM. Woot :) Have you considered learning a bit of git and making patches directly? Yes I have considered it - like so many other things. I guess I'll do it just after digging into IPv6 a bit deeper :) Regards, Simon Cyrus Home Page: http://www.cyrusimap.org/ List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
2.4.6 reconstruct: timestamp mismatch which doesn't go away
Hi, I've done some upgrades from 2.3.16 to 2.4.6 and it seems sometimes a reconstruct is the only way to get a mailbox work after upgrade. While running reconstruct two things have shown up which I have not seen wirh older releases: 1) on a mailbox which was used by a client reconstruct was just sitting there doing nothing. Seems like it was waiting for a lock to be freed? After terminating the client reconstruct would immediately resume it's work. Maybe that's expected, I'm just wondering how those with large number of users will reconstruct while the server is online? I mean, IIRC with older releases it was always possible to do a full reconstruct while users had access to their mailboxes. Is it not possible that way anymore? 2) I got some timestamp mismatch errors while reconstructing my own mailbox. I've read here http://www.mail-archive.com/cyrus-devel@lists.andrew.cmu.edu/msg01481.html that this should happen only once because the timestamp of the file should be corrected. That doesn't seem to be true in my case. Any ideas why? Regards, Simon Cyrus Home Page: http://www.cyrusimap.org/ List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
