Apache crashing every hour - can't find the cause

[iso99]

I can't find the reason why Apache keeps on crashing. I have 2 http monitors and both are reporting downtimes when this happen. Fortunately it's only short, it goes back on after 3-5 minutes. On the downside, it happens at least once every hour. It's been almost two weeks since it started.

During downtime, server load is normal (~1) and lots of free memory (I have 16gb on board). No significant error messages as well on /var/log/messages. Although I see notices of Graceful Reboot notices and lines containing the Apache signature saying it's running normally again.

Other than that I have no clue.

I'm using Cloudlinux.

Apache settings:

ServerLimit 2000
MaxClients 2000
MaxRequestsPerChild 10000
KeepAlive On
KeepAliveTimeout 15
MaxKeepAliveRequests 150

I suspected RLimitMem and RLimitCPU as the source so I commented it, but the crashing still persists.

I need advice what to do next.

Thanks a lot!

 

[Shavaun]

If the log is only showing that it is performing a graceful restart, then it might be restarting due to the logging.

Try enabling piped logging to see if it fixes it. The option is located at: WHM » Service Configuration » Apache Configuration » Piped Log Configuration


You can find our documentation on this option here: https://documentation.cpanel.net/display/ALD/Piped+Log+Configuration

 

[iso99]

I suspected that too after browsing the forums. It seems it did the trick. I will monitor for 24 hours to check.