httpd-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Skylos <>
Subject [users@httpd] Runaway thread, troubleshooting tools needed
Date Wed, 13 Apr 2005 19:51:36 GMT

I'm pretty new to the concept of threads and debugging them - I
currently am running...

Server version: Apache/2.0.53
Server built:   Mar  6 2005 21:52:49

I am utilizing the worker mpm to serve a moderate volume website with
two servers.  I extensively use mod_perl, but I realize this is not a
mod_perl list.  I purely wish to learn ways of finding out what
stage/process is going on when this thread/process locks up.  This
should be general to apache, right?

My symptoms are that a server will start utilizing 100% of one of the
4 cpus.  (this happens on both)  This usually happens within 18 hours
of a server startup, but seems entirel.y random.  If I do an
'apachectl restart', the problem will almost always go away.  However,
even on shutdown, there will be a several httpd processes left that
will not die without a kill -9 to them specifically.

When I display top, it shows the cpu usage, but it doesn't show which
process is using it - sorting by cpu usage shows no process using more
than about 0.3% of the cpu.

I need help.  Can you suggest commands that would allow me to track
the threads, see what might be happening in the request stream when
this occurs?



The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:> for more info.
To unsubscribe, e-mail:
   "   from the digest:
For additional commands, e-mail:

View raw message