CPU Load spike every 7 hours

[slansing]

Could everyone who is experiencing this increase the debug level in nagios.cfg to 12, and extend the max_debug_log_size to 10mb, and restart nagios then send them over after some time in a bad state? Ideally I would like to see some time where the system is usable and some portion of when it becomes bogged down in those debugs.

[WillemDH]

Good news. Since the 1.4 update the load has been normalised massively. No more CPU spikes detected. You can clearly see the difference since the update in the screenshot:
This thread can be closed for me.. (I don't speak for all the others responding in this thread)

Thanks!

Willem

[tmcdonald]

We're going to keep this thread open for a bit until we can hear back from the others.

[BanditBBS]

I'll chime in and say I am no longer concerned with load spikes and/or slow apply configs. 1.4 seems like a very stable release.

[abrist]

Nice. If there is anyone else who was experiencing these load spikes, please chime in. I do not want to lock this thread if there are continued issues with load spikes and core 4.

[SDK]

The spikes are also gone in our environment.

I noticed since the update that the check intervals are not precise anymore:

For example:

1 minute interval

1min.png

Sometimes they have a variance and after the check they get scheduled exactly again:

5min.png

Is this per design or is something wrong going on here.

Kind regards

[slansing]

Yes, those check frequencies can very a bit from time to time based on the scheduling queue, it is normal.