[Dovecot] raw backtrace in masses
Hi
I've got huge amounts of
May 3 15:47:37 host dovecot: [ID 583609 mail.error] IMAP(user): close(/home/user/Mail/R-Help.lock) failed: Disc quota exceeded May 3 15:47:37 host dovecot: [ID 583609 mail.error] IMAP(user): Raw backtrace: 0xb0e58 -> 0xb0894 -> 0x552b4 -> 0x55508 -> 0x55774 -> 0x55d9c -> 0x5759c -> 0x57ad0 -> 0x713cc -> 0x64284 -> 0x57be0 -> 0x71178 -> 0x6841c -> 0x76a38 -> 0x76a6c -> 0x25a08 -> 0x2afd4 -> 0x2bee4 -> 0x2bea0 -> 0x2c1f8 -> 0x2cb4c -> 0xb9450 -> 0xb8d00 -> 0x35294 -> 0x246e8 May 3 15:47:37 host dovecot: [ID 583609 mail.error] dovecot: child 3892 (imap) killed with signal 6 (core dumps disabled) May 3 15:47:49 host dovecot: [ID 583609 mail.error] IMAP(user): Raw backtrace: 0xb0e58 -> 0xb0894 -> 0x552b4 -> 0x55508 -> 0x55774 -> 0x55d9c -> 0x5759c -> 0x57ad0 -> 0x713cc -> 0x64284 -> 0x57be0 -> 0x71178 -> 0x6841c -> 0x76a38 -> 0x76a6c -> 0x25a08 -> 0x2afd4 -> 0x2bee4 -> 0x2bea0 -> 0x2c1f8 -> 0x2cb4c -> 0xb9450 -> 0xb8d00 -> 0x35294 -> 0x246e8 May 3 15:47:49 host dovecot: [ID 583609 mail.error] dovecot: child 4277 (imap) killed with signal 6 (core dumps disabled)
The backtrace is always the same content. is this a disc quota problem?
Dovecot is 1.2.11, managesieve patched.
Thanks
May 3 15:47:37 host dovecot: [ID 583609 mail.error] IMAP(user): close(/home/user/Mail/R-Help.lock) failed: Disc quota exceeded May 3 15:47:37 host dovecot: [ID 583609 mail.error] IMAP(user): Raw backtrace: 0xb0e58 -> 0xb0894 -> 0x552b4 -> 0x55508 -> 0x55774 -> 0x55d9c -> 0x5759c -> 0x57ad0 -> 0x713cc -> 0x64284 -> 0x57be0 -> 0x71178 -> 0x6841c -> 0x76a38 -> 0x76a6c -> 0x25a08 -> 0x2afd4 -> 0x2bee4 -> 0x2bea0 -> 0x2c1f8 -> 0x2cb4c -> 0xb9450 -> 0xb8d00 -> 0x35294 -> 0x246e8 May 3 15:47:37 host dovecot: [ID 583609 mail.error] dovecot: child 3892 (imap) killed with signal 6 (core dumps disabled)
Your user has exceeded the disc quota, but still dovecot shouldn't crash when this happens. Could you enable core dumps and post a backtrace to the list?
On 05.05.2010 14:15, Mike Abbott wrote:
May 3 15:47:37 host dovecot: [ID 583609 mail.error] IMAP(user): close(/home/user/Mail/R-Help.lock) failed: Disc quota exceeded May 3 15:47:37 host dovecot: [ID 583609 mail.error] IMAP(user): Raw backtrace: 0xb0e58 -> 0xb0894 -> 0x552b4 -> 0x55508 -> 0x55774 -> 0x55d9c -> 0x5759c -> 0x57ad0 -> 0x713cc -> 0x64284 -> 0x57be0 -> 0x71178 -> 0x6841c -> 0x76a38 -> 0x76a6c -> 0x25a08 -> 0x2afd4 -> 0x2bee4 -> 0x2bea0 -> 0x2c1f8 -> 0x2cb4c -> 0xb9450 -> 0xb8d00 -> 0x35294 -> 0x246e8 May 3 15:47:37 host dovecot: [ID 583609 mail.error] dovecot: child 3892 (imap) killed with signal 6 (core dumps disabled)
Your user has exceeded the disc quota, but still dovecot shouldn't crash when this happens. Could you enable core dumps and post a backtrace to the list?
I am a bit new to this kind of debugging. What happens if I turn on core dumps on Solaris?
I am a bit new to this kind of debugging. What happens if I turn on core dumps on Solaris?
See the "Core Dumps" section of http://www.dovecot.org/bugreport.html.
Does this fix help against this sort of crashes? http://hg.dovecot.org/dovecot-2.0/rev/8f251e0bc02d
Probably not. You're running 1.2 not 2.0, and you did not report seeing that assertion failure in your logs.
On 05.05.2010 14:53, Mike Abbott wrote:
I am a bit new to this kind of debugging. What happens if I turn on core dumps on Solaris?
See the "Core Dumps" section of http://www.dovecot.org/bugreport.html.
Does this fix help against this sort of crashes? http://hg.dovecot.org/dovecot-2.0/rev/8f251e0bc02d
Probably not. You're running 1.2 not 2.0, and you did not report seeing that assertion failure in your logs.
Well, I do have assertion failure. May 3 15:47:37 host dovecot: [ID 583609 mail.crit] IMAP(user): Panic: file istream-raw-mbox.c: line 583 (istream_raw_mbox_get_body_size): assertion failed: (rstream->body_offset != (uoff_t)-1)
On 05.05.2010 14:15, Mike Abbott wrote:
May 3 15:47:37 host dovecot: [ID 583609 mail.error] IMAP(user): close(/home/user/Mail/R-Help.lock) failed: Disc quota exceeded May 3 15:47:37 host dovecot: [ID 583609 mail.error] IMAP(user): Raw backtrace: 0xb0e58 -> 0xb0894 -> 0x552b4 -> 0x55508 -> 0x55774 -> 0x55d9c -> 0x5759c -> 0x57ad0 -> 0x713cc -> 0x64284 -> 0x57be0 -> 0x71178 -> 0x6841c -> 0x76a38 -> 0x76a6c -> 0x25a08 -> 0x2afd4 -> 0x2bee4 -> 0x2bea0 -> 0x2c1f8 -> 0x2cb4c -> 0xb9450 -> 0xb8d00 -> 0x35294 -> 0x246e8 May 3 15:47:37 host dovecot: [ID 583609 mail.error] dovecot: child 3892 (imap) killed with signal 6 (core dumps disabled)
Your user has exceeded the disc quota, but still dovecot shouldn't crash when this happens. Could you enable core dumps and post a backtrace to the list?
Does this fix help against this sort of crashes? http://hg.dovecot.org/dovecot-2.0/rev/8f251e0bc02d
On Tue, 2010-05-04 at 20:38 +0200, John Adams wrote:
Hi
I've got huge amounts of
May 3 15:47:37 host dovecot: [ID 583609 mail.error] IMAP(user): close(/home/user/Mail/R-Help.lock) failed: Disc quota exceeded
This shouldn't happen if you set mail_nfs_storage=yes.
May 3 15:47:37 host dovecot: [ID 583609 mail.error] dovecot: child 3892 (imap) killed with signal 6 (core dumps disabled)
But yeah, this still shouldn't happen.
On 26.05.2010 18:32, Timo Sirainen wrote:
On Tue, 2010-05-04 at 20:38 +0200, John Adams wrote:
Hi
I've got huge amounts of
May 3 15:47:37 host dovecot: [ID 583609 mail.error] IMAP(user): close(/home/user/Mail/R-Help.lock) failed: Disc quota exceeded
This shouldn't happen if you set mail_nfs_storage=yes.
It is not set to yes. However, it is a mixed environment. 1/3 of the mailboxes are accessible directly on the storage, 2/3 are mounted NFS. What happens to the 1/3 if I set mail_nfs_storage=yes?
Anyway, was dovecot designed for this kind of setup?
On 5.6.2010, at 6.20, John Adams wrote:
May 3 15:47:37 host dovecot: [ID 583609 mail.error] IMAP(user): close(/home/user/Mail/R-Help.lock) failed: Disc quota exceeded
This shouldn't happen if you set mail_nfs_storage=yes.
It is not set to yes. However, it is a mixed environment. 1/3 of the mailboxes are accessible directly on the storage, 2/3 are mounted NFS. What happens to the 1/3 if I set mail_nfs_storage=yes?
Slightly worse performance, but probably not noticeable. It would also be possible to set this value per-user by returning the setting from userdb. http://wiki.dovecot.org/UserDatabase/ExtraFields
Anyway, was dovecot designed for this kind of setup?
Dovecot was designed to be as flexible as possible :) Not perfect yet, but slowly getting there.
On 05.06.2010 16:26, Timo Sirainen wrote:
On 5.6.2010, at 6.20, John Adams wrote:
May 3 15:47:37 host dovecot: [ID 583609 mail.error] IMAP(user): close(/home/user/Mail/R-Help.lock) failed: Disc quota exceeded
This shouldn't happen if you set mail_nfs_storage=yes.
It is not set to yes. However, it is a mixed environment. 1/3 of the mailboxes are accessible directly on the storage, 2/3 are mounted NFS. What happens to the 1/3 if I set mail_nfs_storage=yes?
Slightly worse performance, but probably not noticeable. It would also be possible to set this value per-user by returning the setting from userdb. http://wiki.dovecot.org/UserDatabase/ExtraFields
Changed the config. Thanks. Also thanks for the link. I like this very much :)
Anyway, was dovecot designed for this kind of setup?
Dovecot was designed to be as flexible as possible :) Not perfect yet, but slowly getting there.
participants (3)
-
John Adams
-
Mike Abbott
-
Timo Sirainen