gnokii-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Gnokii smsd SIGSEGV


From: Federico Sevilla III
Subject: Gnokii smsd SIGSEGV
Date: Mon, 23 Sep 2002 08:35:22 +0800
User-agent: Mutt/1.4i

Hi everyone,

I'm using Gnokii smsd from the 0.4.3 release, which is running on a
Debian GNU/Linux system (I am testing -smsd packages I'm building for
submission to the main package maintainer). I built both pq and mysql
modules, but am only using the pq module, and am connecting to a
PostgreSQL server that is on another machine in the LAN. The unit is a
Nokia 3210.

I've noticed that thrice already so far, smsd would segfault after
awhile. "Awhile" being a few days, with seemingly nothing peculiar
happening (eg: no active inbound or outbound messages or calls). During
the third session I ran smsd through strace, generating a full log of
system calls.

A bzip2-compressed log is available (273KB compressed, 74MB extracted)
online at <http://jijo.free.net.ph/smsd.strace.log.bz2>.

The last 50 lines of this log are:

    23324 <... nanosleep resumed> {3, 0})   = 0
    23324 rt_sigaction(SIGPIPE, {SIG_IGN}, {SIG_DFL}, 8) = 0
    23324 send(8, "QBEGIN\0", 7, 0)         = 7
    23324 rt_sigaction(SIGPIPE, {SIG_DFL}, {SIG_IGN}, 8) = 0
    23324 select(9, [8], [], [8], NULL)     = 1 (in [8])
    23324 recv(8, "CBEGIN\0Z", 16384, 0)    = 8
    23324 rt_sigaction(SIGPIPE, {SIG_IGN}, {SIG_DFL}, 8) = 0
    23324 send(8, "QSELECT id, number, text FROM ou"..., 94, 0) = 94
    23324 rt_sigaction(SIGPIPE, {SIG_DFL}, {SIG_IGN}, 8) = 0
    23324 select(9, [8], [], [8], NULL)     = 1 (in [8])
    23324 recv(8, "Pblank\0T\0\3id\0\0\0\0\27\0\4\377\377\377\377number\0"..., 
16384, 0) = 64
    23324 rt_sigaction(SIGPIPE, {SIG_IGN}, {SIG_DFL}, 8) = 0
    23324 send(8, "QCOMMIT\0", 8, 0)        = 8
    23324 rt_sigaction(SIGPIPE, {SIG_DFL}, {SIG_IGN}, 8) = 0
    23324 select(9, [8], [], [8], NULL)     = 1 (in [8])
    23324 recv(8, "CCOMMIT\0Z", 16384, 0)   = 9
    23324 rt_sigprocmask(SIG_BLOCK, [CHLD], [RTMIN], 8) = 0
    23324 rt_sigaction(SIGCHLD, NULL, {SIG_DFL}, 8) = 0
    23324 rt_sigprocmask(SIG_SETMASK, [RTMIN], NULL, 8) = 0
    23324 nanosleep({3, 0}, {3, 0})         = 0
    23324 rt_sigaction(SIGPIPE, {SIG_IGN}, {SIG_DFL}, 8) = 0
    23324 send(8, "QBEGIN\0", 7, 0)         = 7
    23324 rt_sigaction(SIGPIPE, {SIG_DFL}, {SIG_IGN}, 8) = 0
    23324 select(9, [8], [], [8], NULL)     = 1 (in [8])
    23324 recv(8, "CBEGIN\0Z", 16384, 0)    = 8
    23324 rt_sigaction(SIGPIPE, {SIG_IGN}, {SIG_DFL}, 8) = 0
    23324 send(8, "QSELECT id, number, text FROM ou"..., 94, 0) = 94
    23324 rt_sigaction(SIGPIPE, {SIG_DFL}, {SIG_IGN}, 8) = 0
    23324 select(9, [8], [], [8], NULL)     = 1 (in [8])
    23324 recv(8, "Pblank\0T\0\3id\0\0\0\0\27\0\4\377\377\377\377number\0"..., 
16384, 0) = 64
    23324 rt_sigaction(SIGPIPE, {SIG_IGN}, {SIG_DFL}, 8) = 0
    23324 send(8, "QCOMMIT\0", 8, 0)        = 8
    23324 rt_sigaction(SIGPIPE, {SIG_DFL}, {SIG_IGN}, 8) = 0
    23324 select(9, [8], [], [8], NULL)     = 1 (in [8])
    23324 recv(8, "CCOMMIT\0Z", 16384, 0)   = 9
    23324 rt_sigprocmask(SIG_BLOCK, [CHLD], [RTMIN], 8) = 0
    23324 rt_sigaction(SIGCHLD, NULL, {SIG_DFL}, 8) = 0
    23324 rt_sigprocmask(SIG_SETMASK, [RTMIN], NULL, 8) = 0
    23324 nanosleep({3, 0},  <unfinished ...>
    23322 wait4(-1, [WIFSIGNALED(s) && WTERMSIG(s) == SIGSEGV], 
WNOHANG|__WCLONE, NULL) = 23323
    23322 kill(23324, SIGSEGV <unfinished ...>
    23324 <... nanosleep resumed> 0xbf5ff960) = -1 EINTR (Interrupted system 
call)
    23322 <... kill resumed> )              = 0
    23324 --- SIGSEGV (Segmentation fault) ---
    23322 kill(23321, SIGSEGV)              = 0
    23321 --- SIGSEGV (Segmentation fault) ---
    23322 --- SIGRT_1 (Real-time signal 1) ---
    23321 +++ killed by SIGSEGV +++
    23322 sigreturn()                       = ? (mask now ~[TRAP KILL STOP])
    23322 _exit(0)                          = ?

smsd is running again, albeit not being logged by strace anymore. It
normally runs -- and works really well -- for a few days or even a week
or so before crashing. I hope the information in the strace will be
useful. Please let me know if you need any more information or if this
has been fixed in CVS.

Cheers! :)

 --> Jijo

-- 
Federico Sevilla III   :  http://jijo.free.net.ph
Network Administrator  :  The Leather Collection, Inc.
GnuPG Key ID           :  0x93B746BE

Attachment: pgpqcY2Hb19IX.pgp
Description: PGP signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]