Multiple nph-zms processes

vandl · Post by **vandl** » Wed Jun 28, 2006 4:39 am

I read this in the README:
If you find you have several zms processes running when nothing is being viewed then it is likely you need a patch for apache (see the Troubleshooting section).

The troubleshooting section did not mention anything about a patch for apache. I also cannot find anything addressing this in the wiki or in these forums.

I am running an rpm apache2-2.0.47-6.12.92mdk.i586.rpm (Mandrake 9.2 w/patch) And I built ZoneMinder v1.22.2 from the source.

Almost everything else works fine (thanks for the great software), I am just curious how to stop all these zms processes from getting wedged, as it slows my entire server to a crawl after a while.

They take CPU and memory even when the browser that connected to them has long since been closed.

- tw

Post by **cordel** » Wed Jun 28, 2006 6:45 am

Update your apache package to 2.0.48. That should take care of your problem.

Regards,
Corey

vandl · Post by **vandl** » Wed Jun 28, 2006 6:14 pm

Thanks -- unfortunantly, I'm having a bit of a hassle finding a 2.0.48 rpm version of apache for Mandrake 9.x that is not compiled as a threaded MPM (threaded MPMs seem to be conflicting with PHP) I should just build apache (or a whole machine) from source, or upgrade to a newer distro. It's been a few years...

FWIW, I wrote a little script that I put in my crontab to run once per minute that kills all but the last 3 nph-zsm processes based on PID. This keeps my server from getting bogged down for now until I get apache upgraded.

ps ax | grep nph-zms | tac | awk 'NR>4{ print $1 | "xargs kill -9" }'

PeterHoward · Post by **PeterHoward** » Thu Jul 20, 2006 1:26 am

cordel wrote:Update your apache package to 2.0.48. That should take care of your problem.

I'm seeing this behaviour using apache 2.0.55 (specifically 2.0.55-4ubuntu2). I built ZM 1.22.2 from source as well.

Gotta love a load average of 16.5

It eventually cleans up by itself (taking about 10 minutes).

PJH

PeterHoward · Post by **PeterHoward** » Thu Jul 20, 2006 11:48 pm

Ah, the "sorry, no idea" silence . . .

I went to the old FAQ, and both links (to the patched RPM and the forum discussion) don't work.

Can anyone point me to the original apache bug info so I can track down that it's really in Ubuntu's release?

PJH

eddyhkim · Post by **eddyhkim** » Mon Jul 24, 2006 2:30 am

i'm running ubuntu 6.06 with apache 2.0.55 on 3 machines. I saw a bit of a problem with high loads until i upped the shared memory, but i haven't been seeing any particular problem with nph-zms processes hanging around after the client closes. these machines are intel boxes (1.5 ghz and up), 1 gig memory with nothing else difficult going on in them.

no idea really whats going on but just a data point for saying there aren't any problems here in that regard.

PeterHoward · Post by **PeterHoward** » Mon Jul 24, 2006 11:05 pm

Thanks for that. It sounds like my HW is similar (Celeron 2GHz, 1G memory)

Did you build ffom source?
What did you set the shared memory to?

Thanks again.