--
We have a program that is 'automatically' invoked by init when the system
boots that provides access to an application for all user's on the system...
Here's what we are suddenly seeing (AIX 4.3.1) (it worked just fine
until recently) ...When the 'automatic' program is killed (by SIGINT
on the console)...it cleans up normally and then calls
killpg(SIGKILL, getpid())...(btw: it is the process group leader)...
...It is respawned by 'init' (normal operation)...but what we then see
is the 'kernel (gil)' process begins consuming substantial CPU and the
original 'automatic' process has not COMPLETELY died, and the system is
at zero idle...
If you run a 'ps -ef' and grep for the pid it yields:
root 8570 1 102 Dec 31 - 0:00 []
A 'crash stack trace' yields:
===========================================================================
dssts1h1:/>echo trace 39 | crash
WARNING: Using crash on a live system can potentially
cause a system crash and/or data corruption.
.soclose () 2ff3b0a8Quote:> STACK TRACE:
Does anyone have a clue as to what the problem may be?
TIA,
Tony
--
-Deanna
> We have a program that is 'automatically' invoked by init when the system
> boots that provides access to an application for all user's on the system...
> Here's what we are suddenly seeing (AIX 4.3.1) (it worked just fine
> until recently) ...When the 'automatic' program is killed (by SIGINT
> on the console)...it cleans up normally and then calls
> killpg(SIGKILL, getpid())...(btw: it is the process group leader)...
> ...It is respawned by 'init' (normal operation)...but what we then see
> is the 'kernel (gil)' process begins consuming substantial CPU and the
> original 'automatic' process has not COMPLETELY died, and the system is
> at zero idle...
> If you run a 'ps -ef' and grep for the pid it yields:
> root 8570 1 102 Dec 31 - 0:00 []
> A 'crash stack trace' yields:
> ===========================================================================
> dssts1h1:/>echo trace 39 | crash
> WARNING: Using crash on a live system can potentially
> cause a system crash and/or data corruption.
> > STACK TRACE:
> .soclose () 2ff3b0a8
> .soo_close () 2ff3b118
> .closef () 2ff3b168
> .closefd () 2ff3b1b8
> .fs_exit () 2ff3b218
> .kexitx () 2ff3b268
> .kexit () 2ff3b2e8
> .sig_deliver () 2ff3b358
> IAR not in kernel segment.
> ==========================================================================
> Does anyone have a clue as to what the problem may be?
> TIA,
> Tony
> --
1. ksh: trap '...' exit int ... or just trap '...' exit?
trap '...' exit int ...
int will cause the script to exit, so '...' get run twice:
#!/bin/ksh
trap 'print a' exit int
sleep 10
^Ca
a
But the natural exit will always causes '...' to run,
which is often not desired.
So should we always write something like
trap 'print a' exit int
...
trap - exit
exit 0
?
--
Michael Wang
http://www.unixlabplus.com/
2. New Server Setup + content filter
3. How do I get 'w' to say that i am doing something else
5. What am I doing wrong with 'chat'?
7. why ./configure use '{ (exit 1); exit 1; }' ?
8. HELP - It will now work - you you know why?
9. What is this supposed to return: "trap 'exit ' 0; exit"?
10. what is 'protection fault' and 'system calls' ?
11. why would 'cp' or 'ls -al' command hangs on Solaris 2.7 ?
12. 'write' won't write; exits immediately.
13. First boot after install hangs after ''Starting PCMCIA services''.