This discussion has been locked.

You can no longer post new replies to this discussion. If you have a question you can start a new discussion

High CPU Load

Hi,

I have installed a Sophos UTM 9.005-16 (64bits) on a VMWare host.
The only purpose of this Sophos UTM is to manage my Websites.

From time to time I experience a huge load without any reason, I just reboot the Sophos Virtual Machine.

I have some trace to show you :

For Now I use only :

When I manage to get a top, this is what I saw :

The CPU spike to 99% and my Websites was not showing up anymore, so I reboot the VM.

For the CPU (in the VM) it's :

processor : 4
vendor_id : GenuineIntel
cpu family : 6
model : 45
model name : Intel(R) Xeon(R) CPU E5-1620 0 @ 3.60GHz

Can you help me to find what happened ?

Thank A lot.

This thread was automatically locked due to age.

0 BarryG over 13 years ago

Hi, pngquant and some of your other processes listed are part of the reporting system.

Do you have enough disk I/O bandwidth for the system?

You could limit the reporting history if needed.

Barry
- 0 Gaeld over 13 years ago in reply to BarryG
  
  Hi BarryG,
  
  Thanks for your reply !
  
  The Esxi Host is on SSD disk and the history is limited to 3 days (set in Logging & Reporting > Log Settings > Local Logging > Automatic log file deletion) and define at max 1 week in Logging & Reporting > Reporting Settings > Settings.
  
  So I don't think that's a I/O problem but I soon as the problem occurs, I wiil look at the I/O and produce more logs.
  
  By the way, I already rebuild the Postgres database to check if this is the problem.
0 BAlfson over 13 years ago

By the way, I already rebuild the Postgres database to check if this is the problem.

Good work at finding that on your own here!

Do you know what happened at about 01:30 to cause the increase from 3% to 15% CPU? Do you see anything on the Hardware graph that seems to correspond with your CPU graph?

Cheers Bob

Sophos UTM Community Moderator
Sophos Certified Architect - UTM
Sophos Certified Engineer - XG
Gold Solution Partner since 2005

MediaSoft, Inc. USA
0 Gaeld over 13 years ago

Hi BAlfson,

No I don't know what happened at 01:30. There is no significant increase of connection, memory usage is flat (21-24% at all time).

I have try to check in /etc/crontab if it something there to show me any process starting at 01:30 without success.

Maybe the postgres rebuild will solve this issue.

That's very weird because I don't have a lot of services activated and not too much connections.

EDIT : Sorry, the only process crontabed at 01:30 is the log cleaner :
30 1 * * * root /usr/local/bin/logcleaner.plx -e
- 0 Gaeld over 13 years ago in reply to Gaeld
  
  Hi All,
  
  Just to let you know that so far, I did'nt experienced any trouble.
  
  Maybe the Postgres Rebuild was a good hint.
  
  thanks for your help anyway !!!!!
  
  Best regards,
  
  Gaël.