Nagios Support Forum

Our Nagios core 3.5.1 server creates cpu load spike every 7 hours. It does it like clockwork. This creates check delays which results in erroneous data and potential alerts during the load spikes.
In common example as Nagios server slows down, our icmp/ping checks return excessive round trip times, while network is actually working just fine.
While we suspect disk i/o is culprit for the load issue, we are not able to understand why it happens at such regular intervals.
Is there a some configuration parameter(s) we can adjust that might affect cpu load?

cptcrush wrote:Is there a some configuration parameter(s) we can adjust that might affect cpu load?

Nothing that will fix this. It was first reported in core 4, though that is not to say that 3.5.x was not effected as well. We did make a number of changes to the scheduler in core 4.0.8 in order to try and compensate for this behavior.

We are also seeing this problem with 4.0.8

We upgrade our server from CentOS 5 running Nagios 3.5 to CentOS 7 running Nagios Core 4.0.8. We didn't have any load issues on the old server but on the new one we see the load spike (using the same config file).

We have 204 host and 1728 checks. CPU usage and disk wait times all look normal.

Yep, just a heads up that we are looking into this, there are a multitude of threads open on this issue. We're going to lock this one for now as the issue is being tracked, feel free to ask for it to be re-opened should you wish to share additional information. Thank you!

Hey, did you know that there are CPU load spikes caused by Nagios every seven hours?