[Perfparse-users] Nagios hangs with perfparse enabled

[Richard D. <de...@ya...>]

I have been running Nagios 1.2 on a older host and
decided to upgrade the existing version to a new
server with more power (load was 5+ on older system). 
The changes were also:

1.  RHES3 => RHES4
2.  Nagios pluging 1.3.x to 1.4
3.  Enabled MySQL 4.1.x

The initial migration of the Nagios 1.2 went as
planned and everything in Nagios worked as expected.  
I then went to add perfparse using the older default
cron job perfparse.sh method to start and learn how it
worked.  Initially, everything was working and MySQL
was updating.  

So after doing a few QA days, I put the new system
into full production.  Everthing started out okay, but
I found the next morning that Nagios itself was hung. 
I got Nagios restarted and monitored.  Nagios hung
again after several hours.  The time factor varied so
it was from a few to over 12 hours, but when Nagios
hung,the processes would all be in a write state
(strace check).  I needed to killall to remove the
hung Nagios processes.  I did this a few more times to
verify, then backed out perfparse (commented out
nagios.cfg section and disabled by setting
process_performance_data=0).  I commented out the cron
job which was running every 10 minutes.  I reboot the
system to be sure and Nagios has been up and
processing (load now 0.5) without any hanging.  

The question is, has anyone had this experience and
what can I do about it.  I want to capture the data
with perfparse and will use an alternative method as
suggested.  Again, any advice on setting up perfparse
to avoid this hanging.  If you need more data, I can
get that too.

Thanks,

Richard


		
____________________________________________________
Start your day with Yahoo! - make it your home page 
http://www.yahoo.com/r/hs