[Dovecot] imap cores

[Anil]

Timo,

Its not just imap, it seems deliver is dieing too.
I created a brand new Maildir and I still see it crap out with the
same error message:

Jan  5 23:28:51 vds5 deliver(admin at domain.com): [ID 480647 mail.crit]
file index-mail.c: line 1042: assertion failed:
(!mail->data.destroying_stream)
Jan  5 23:28:51 vds5 deliver(admin at domain.com): [ID 398108 mail.error]
Raw backtrace:
Jan  5 23:28:51 vds5 sendmail[25792]: [ID 801593 mail.crit]
m067Sp2s025788: SYSERR(root): mailer dovecot died with signal 6


On Jan 5, 2008 12:39 PM, Anil <replicase at gmail.com> wrote:
> Experts,
>
> (Now that I am able to run imap...)
>
> Here's what I am seeing with imap, it seems to die in the latest
> nightly. (tss, I am still trying to get dbx to work properly, maybe it
> won't, but I got some more details here.)
>
> ...
> 2 select "INBOX"
> * FLAGS (\Answered \Flagged \Deleted \Seen \Draft)
> * OK [PERMANENTFLAGS (\Answered \Flagged \Deleted \Seen \Draft \*)]
> Flags permitted.
> * 1 EXISTS
> * 0 RECENT
> * OK [UNSEEN 1] First unseen.
> * OK [UIDVALIDITY 1199514743] UIDs valid
> * OK [UIDNEXT 2] Predicted next UID
> 2 OK [READ-WRITE] Select completed.
> 3 UID fetch 1 ( BODY.PEEK[HEADER.FIELDS (From)])
> Connection to mail closed by foreign host.
>
> openat(-3041965, "/var/mail/apps/domain/admin/Maildir/cur",
> O_RDONLY|O_NDELAY|O_LARGEFILE) = 9
> fcntl(9, F_SETFD, 0x00000001)                   = 0
> fstat64(9, 0x080474C0)                          = 0
> stat64("/var/mail/apps/domain/admin/Maildir/cur", 0x080475C0) = 0
> time()                                          = 1199564680
> getdents64(9, 0xFEFA4000, 8192)                 = 112
> getdents64(9, 0xFEFA4000, 8192)                 = 0
> close(9)                                        = 0
> fcntl(6, F_SETLK64, 0x08047680)                 = 0
> lstat64("/var/mail/apps/domain/admin/Maildir/dovecot-uidlist.lock",
> 0x080476C0) = 0
> unlink("/var/mail/apps/domain/admin/Maildir/dovecot-uidlist.lock") = 0
> open64("/var/mail/apps/domain/admin/Maildir/cur/1199514786.M867961P7368.vds5,W=1325:2,",
> O_RDONLY) = 9
> fstat64(9, 0x080477A0)                          = 0
> pread64(9, " R e t u r n - P a t h :".., 4096, 0) = 1298
> pread64(9, " R e t u r n - P a t h :".., 4096, 0) = 1298
> close(9)                                        = 0
> write(2, "01 P f i l e   i n d e x".., 82)      = 82
> getcontext(0x08047510)
> open("/proc/self/as", O_RDONLY)                 = 9
> close(9)                                        = 0
> write(2, "01 E R a w   b a c k t r".., 18)      = 18
> sigaction(SIGABRT, 0x00000000, 0x08047710)      = 0
> sigaction(SIGABRT, 0x08047650, 0x080476D0)      = 0
> schedctl()                                      = 0xFED32000
> lwp_sigmask(SIG_SETMASK, 0x00000000, 0x00000000) = 0xFFBFFEFF [0x0000FFFF]
> lwp_kill(1, SIGABRT)                            = 0
>     Received signal #6, SIGABRT [default]
>       siginfo: SIGABRT pid=3257 uid=500 code=-1
>
>
> This is all I can seem to get with dbx:
> program terminated by signal ABRT (Abort)
> 0xfee55687: __lwp_kill+0x0007:  jae      __lwp_kill+0x15        [
> 0xfee55695, .+0xe ]
> (dbx) where
> =>[1] __lwp_kill(0x1, 0x6), at 0xfee55687
>   [2] _thr_kill(0x1, 0x6), at 0xfee52dee
>   [3] raise(0x6), at 0xfee011bb
>   [4] abort(0x817e3d8, 0x817e3d8, 0xfee7f5b8, 0x816e898, 0xffffffff,
> 0x81759e0), at 0xfede15d9
> (dbx)
>
>
> # pstack core.imap.3479
> core 'core.imap.3479' of 3479:  imap
>  fee55687 _lwp_kill (1, 6) + 7
>  fee011bb raise    (6) + 1f
>  fede15d9 abort    (817e3d8, 817e3d8, fee7f5b8, 816e898, ffffffff, 81759e0) + cd
>  08118884 i_internal_fatal_handler () + 54
>
>
> What do you think?
>