Re: few programs now getting sig 6

On Feb 14, 2005, at 11:36 PM, Daniel O'Connor wrote:

> On Tue, 15 Feb 2005 18:02, Sean McNeil wrote:
>> I've also seen a core dump on a recompile attempt for howl (port that
>> has mDNSResponder).  It wasn't repeatable, so I thought my system had
>> memory issues.  After some reboots and testing, I've found that the
>> above aborts are consistent and not memory related.
>
> signal 6 is almost invariably caused by the process calling abort() - 
> ie it's
> shooting itself in the head because some expectation it had has been
> violated.
>
> I'd check log messages (if the programs generate them..)
>
> The big hammer approach would be to just rebuild the broken apps and 
> see if
> that fixes it :)
>
That was the first thing I tried.  I recompiled howl and it fails the 
same way.  Even did a portupgrade -fR on the port.  No joy.

I forgot to mention that this is an amd64 system.  Here is a traceback 
of the core file:

#0  0x00000008009abcfc in kse_thr_interrupt () at kse_thr_interrupt.S:2
#1  0x000000080099c4fd in _thr_sig_add (pthread=0x50d800, sig=6, 
info=0x0)
     at /usr/src/lib/libpthread/thread/thr_sig.c:974
#2  0x000000080099c971 in _thr_sig_send (pthread=0x50d800, sig=6)
     at /usr/src/lib/libpthread/thread/thr_sig.c:1106
#3  0x000000080099675d in _pthread_kill (pthread=0x50d800, sig=6)
     at /usr/src/lib/libpthread/thread/thr_kill.c:60
#4  0x0000000800995fd0 in _raise (sig=6)
     at /usr/src/lib/libpthread/thread/thr_raise.c:46
#5  0x0000000800b80456 in abort () at 
/usr/src/lib/libc/stdlib/abort.c:69
#6  0x00000008009ac588 in _thr_exit (
     fname=0x8009adb60 "/usr/src/lib/libpthread/thread/thr_kern.c",
     lineno=1101, msg=0x8009add70 "Thread has returned from 
_thread_switch")
     at /usr/src/lib/libpthread/thread/thr_exit.c:56
#7  0x00000008009a48a0 in kse_sched_multi (kmbx=0x0)
     at /usr/src/lib/libpthread/thread/thr_kern.c:1101
#8  0x0000000000000000 in ?? ()
#9  0x000000080052d040 in ?? ()
#10 0x000000000050a000 in ?? ()
#11 0x0000000000505000 in ?? ()
#12 0x0000000000000000 in ?? ()
#13 0x0000000000000000 in ?? ()
#14 0x0000000000000000 in ?? ()
#15 0x0000000000000000 in ?? ()
#16 0x0000000000000000 in ?? ()
#17 0x0000000000000000 in ?? ()
#18 0x00000008009a42f0 in _thr_sched_switch_unlocked ()
     at /usr/src/lib/libpthread/thread/thr_kern.c:885
(gdb) info threads
   3 Thread 3 (runnable)  0x000000080099c202 in _thr_sig_rundown (
     curthread=0x50d000, ucp=0x7fffffffe660)
     at /usr/src/lib/libpthread/thread/thr_sig.c:870
   2 Thread 2 (LWP 100133)  0x00000008009abce4 in kse_release ()
     at kse_release.S:2
* 1 Thread 1 (LWP 100181)  0x00000008009abcfc in kse_thr_interrupt ()
     at kse_thr_interrupt.S:2
(gdb) thread 3
[Switching to thread 3 (Thread 3 (runnable))]#0  0x000000080099c202 in 
_thr_sig_rundown (curthread=0x50d000, ucp=0x7fffffffe660)
     at /usr/src/lib/libpthread/thread/thr_sig.c:870
870             errno = err_save;
Current language:  auto; currently c
(gdb) bt
#0  0x000000080099c202 in _thr_sig_rundown (curthread=0x50d000,
     ucp=0x7fffffffe660) at /usr/src/lib/libpthread/thread/thr_sig.c:870
#1  0x000000080099c29c in _thr_sig_check_pending (curthread=0x50d000)
     at /usr/src/lib/libpthread/thread/thr_sig.c:908
#2  0x00000001ffffea38 in ?? ()
#3  0x00000000005020c0 in ?? ()
#4  0x000000080099e08b in _thr_rtld_lock_release (lock=0x50d060)
     at /usr/src/lib/libpthread/thread/thr_rtld.c:263
#5  0x000000080050c3ca in rlock_release (lock=0x80062bb20, 
locked=5296264)
     at /usr/src/libexec/rtld-elf/rtld_lock.c:211
#6  0x0000000800508c96 in _rtld_bind (obj=0x800b0a144, 
reloff=34370953728)
     at /usr/src/libexec/rtld-elf/rtld.c:480
#7  0x0000000800507efd in _rtld_bind_start ()
     at /usr/src/libexec/rtld-elf/amd64/rtld_start.S:99
#8  0x0000000000000202 in ?? ()
#9  0x0000000000000000 in ?? ()
#10 0x0000000000502080 in ?? ()
#11 0x000000080064c260 in sw_mdns_servant_refresh ()
    from /usr/local/lib/libmDNSResponder.so.0
Previous frame inner to this frame (corrupt stack?)
(gdb)

I tried with a feb 8th kernel and got the same error, so the problem 
must be in a library like libc or libpthread.

Cheers,
Sean