From: kdea@alpine-la.com
Date: Thu Mar 24 2005 - 17:07:03 EST
Hello Managers,
I want to thank Andy Wagg, who gave me his response, which was what we were
afraid to hear. Also, the engineers do need the core file so setting the
core size to 0 was not possible either. The only recourse was to reset the
machine, which was what I ended up doing.
Kevin
--Andy Wagg's response--
As far as I know your only option is to wait for it. In the past when I
have had this problem we've waited around 30 minutes for a 2Gb core to
be written. The only soultion for this in the future would be to limit
the size of your core file or prevent it from being generated at all.
This can be done by setting the coresize in the process to 0 or a size
you are willing to wait for. Obviously, if you require the full core to
determine the problem this will limit you somewhat.
---Original Problem---
One of our engineers asked me to kill a process that was running, and
hogging up 100% of the CPU. It was for some custom program that they wrote
to write to a database of files. The process would not die, even with
repeated attempts with 'kill -9'. Upon investigation, we've discovered
that it may be writing to a core file. The core dump is getting bigger and
bigger, but I don't know how long this will go on. We do have 32 GB of
memory on the machine, and this process may be using a good deal of it. Is
there a way I can stop all of this, the core dump and the parent process?
The machine is an ES40 running Tru64 5.1a. Restarting is not an option.
-- Kevin Dea UNIX System Administrator Alpine Electronics Research of America
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:50:17 EDT