[Zope] Troubleshooting Zope 100% CPU utilization

Paul Winkler pw_lists at slinkp.com
Wed Mar 9 17:47:46 EST 2005


On Wed, Mar 09, 2005 at 04:22:39PM -0500, Matthew X. Economou wrote:
> After running properly for about an hour or two, my Zope instance stops
> responding to HTTP requests and begins consuming all available CPU time.
> How should I go about troubleshooting this problem?
(snip)
> What else should I check?

Two more suggestions:

* try enabling the trace aka debug log in etc/zope.conf.
(But this won't help until after you've restarted zope.
I generally turn on trace logging for all my zopes, it's helped
pinpoint problems on a couple of occasions.)
There's a script for analzying it, this comes with zope
in bin/requestprofiler.py.

* google for "debug spinning zope".  This too has helped
me find a method that was hanging (although in my case it
wasn't CPU, it was waiting forever on a connection that
was blocked by incorrect firewall config). If a method never
returns, all it takes is for all your zope worker threads to
hit that method and bye-bye zope.

-- 

Paul Winkler
http://www.slinkp.com


More information about the Zope mailing list