Subject: Sudden core dump with SIGALRM
To: None <current-users@NetBSD.org>
From: Takahiro Kambe <taca@back-street.net>
List: current-users
Date: 09/16/2003 01:01:26
Hi.

I've tested to run a static linked binary (unknown version of NetBSD).
It works on 1.6.1_STABLE fine.  But on current, it core dumped just
after one minute from start.  (I hide real its name.)

  1046 program  CALL  socket(0x2,0x1,0)
  1046 program  RET   socket 8
  1046 program  CALL  setsockopt(0x8,0xffff,0x4,0xbfbff698,0x4)
  1046 program  RET   setsockopt 0
  1046 program  CALL  bind(0x8,0xbfbff664,0x10)
  1046 program  RET   bind 0
  1046 program  CALL  listen(0x8,0x5)
  1046 program  RET   listen 0
  1046 program  CALL  __sigaction14(0xe,0xbfbff60c,0xbfbff5f4)
  1046 program  RET   __sigaction14 0
  1046 program  CALL  setitimer(0,0xbfbff614,0xbfbff604)
  1046 program  RET   setitimer 0
  1046 program  CALL  accept(0x8,0xbfbff674,0xbfbff684)
  1046 program  RET   accept RESTART
  1046 program  PSIG  SIGALRM caught handler=0x8049900 mask=() code=0x0
  1046 program  CALL  __sigaction14(0xe,0xbfbff57c,0xbfbff564)
  1046 program  RET   __sigaction14 0
  1046 program  PSIG  SIGSEGV SIG_DFL
  1046 program  NAMI  "program.core"

Dose recent signal change affect it?

The system was:

NetBSD branch.back-street.net 1.6ZB NetBSD 1.6ZB (EPIA) #22: Sun Sep 14 23:26:54 JST 2003  taca@edge.a.back-street.net:/var/obj/sys/arch/i386/compile/EPIA i386

-- 
Takahiro Kambe <taca@back-street.net>