v2.3.6 message delete causes signal 10

[Rosenbaum, Larry M.]

> -----Original Message-----
> From: Ken Murchison [mailto:murch at andrew.cmu.edu]
...
> >>>> Rosenbaum, Larry M. wrote:
> >>>>> I have just installed Cyrus IMAP v2.3.6 on a test system.
> > Whenever
> >>> I
> >>>>> try to delete a message in my INBOX, the delete fails and I see
> > the
> >>>>> following message in the log file:
> >>>>>
> >>>>> May 26 11:41:01 xxxx master[10821]: [ID 970914 local6.error]
> > process
> >>>>> 10841 exited, signaled to death by 10
> >>>>>
> >>>>> This happens whether or not expunge_mode: delayed is set.
> >>>>>
> >>>>> This is on Solaris 8 with the following setup:
> >>>>>
> >>>>> name       : Cyrus IMAPD
> >>>>> version    : v2.3.6 2006/05/25 15:21:49
> >>>>> vendor     : Project Cyrus
> >>>>> support-url: http://asg.web.cmu.edu/cyrus
> >>>>> os         : SunOS
> >>>>> os-version : 5.8
> >>>>> environment: Built w/Cyrus SASL 2.1.22
> >>>>>              Running w/Cyrus SASL 2.1.22
> >>>>>              Built w/Sleepycat Software: Berkeley DB 4.4.20:
> >>> (January
> >>>>> 10, 2006)
> >>>>>              Running w/Sleepycat Software: Berkeley DB 4.4.20:
> >>> (January
> >>>>> 10, 2006)
> >>>>>              Built w/OpenSSL 0.9.8a 11 Oct 2005
> >>>>>              Running w/OpenSSL 0.9.8a 11 Oct 2005
> >>>>>              CMU Sieve 2.3
> >>>>>              mmap = shared
> >>>>>              lock = fcntl
> >>>>>              nonblock = fcntl
> >>>>>              idle = idled
> >>>>>
> >>>>> Cyrus IMAP v2.3.3 (with sasl v2.1.21) ran fine.  Any ideas?
> >>>> Can you get a backtrace from a core dump?
> >>>
> >>> I was unable to get a core dump, but I was able to determine that
it
> >>> appears to be choking on the following statement (line 1000 of
> >>> mailbox.c):
> >>>
> >>>     record->modseq = htonll(*((bit64 *)(buf+OFFSET_MODSEQ_64)));
> >> This only occurs when you delete a message?  Which part, flagging
it
> >> with \Deleted, or doing the expunge?
> >
> > Flagging it with \Deleted.
> >
> >> It would seem strange that this only occurs for only one operation,
> > and
> >> this function is called in a lot of places.
> >
> > This is not a production system; the only user is me.  It's entirely
> > likely that I haven't tried anything else that uses the same
function.
> >
> >> Just curious, does pop3d crash at login?
> >
> > Sonofagun.... yes it does.
> 
> OK.  32-bit or 6-bit Solaris kernel?

64 bit:

# uname -a
SunOS emroute1x 5.8 Generic_117350-16 sun4u sparc SUNW,Ultra-250
# isainfo -kv
64-bit sparcv9 kernel modules

> --
> Kenneth Murchison
> Systems Programmer
> Project Cyrus Developer/Maintainer
> Carnegie Mellon University