CPU Load Spike daily

[belvdr]

Update: as of 9am CST I have been at double digit load for 25 minutes and counting. I'm going to kill the nagios process and le tit sit for a few minutes just so I can use the server.

Just curious and on whim, what are your server details? I'm curious if it's a VM hardware thing.

[emislivec]

So in short, the new scheduling is not good at all*except for the load easing).

I have a suspicion that commit will get reverted . . . .

...yeah, I have a suspicion you're right...
That commit applies a random adjustment to certain checks. It obviously needs to be more constrained, and probably not random...

More on this soon as Eric[1]'s work progresses. . . .

I'm looking into the static scheduling algorithm Core runs on startup when not using the retained schedule. We probably need to run a full, proper reschedule more often, which is simpler than it sounds. Checks with different/longer intervals, retries, dependent checks, host/service additions/removals, time changes, etc., make things 'interesting'. Efficiently generating a smooth schedule that doesn't bunch checks at some point is pretty much impossible without violating the check_interval somewhat. However, the steady state should be much smoother for simpler schedules.

So yeah, back to the code.

[Smark]

Hello everyone,

I've read through this post but I want to be sure what I'm seeing on my system is the same issue we're discussing in this thread... Please see the attachment which shows the load average of the Nagios server over about 24hrs.

I did nothing differently but did notice the change in load pattern shortly after an update. While there were upticks in usage, the spikes were smaller and closer together. For the time being I've just been throwing CPUs at it. Currently I'm at 16 vCPUs at 6GB RAM.

[tmcdonald]

Your symptoms do seem to line up well with what we are seeing elsewhere.

[Smark]

Your symptoms do seem to line up well with what we are seeing elsewhere.

Thanks. Please let me know how I can help by submitting logs or configuration information. While troubleshooting another issue (turned out to be related to colons in service names) I de-provisioned my RAMDisk and noticed the load spikes went from about 12-15 to ~20ish.

[BanditBBS]

Update: as of 9am CST I have been at double digit load for 25 minutes and counting. I'm going to kill the nagios process and le tit sit for a few minutes just so I can use the server.

Just curious and on whim, what are your server details? I'm curious if it's a VM hardware thing.

I am going to test if it is a VM host issue hopefully later today. Getting another host configured now! I saw this at my house testing 2014 and I moved the server to a new VM host and the issue never appeared again. The weird thing is, both hosts were identical HP DL360G5 with 64GB ram and 2 quad core XEON.

The server at work we are running on is:
Dell M605, ESXi 5.x, 64 GB, two 6 core AMD 2.40Ghz

[belvdr]

Update: as of 9am CST I have been at double digit load for 25 minutes and counting. I'm going to kill the nagios process and le tit sit for a few minutes just so I can use the server.

Just curious and on whim, what are your server details? I'm curious if it's a VM hardware thing.

I am going to test if it is a VM host issue hopefully later today. Getting another host configured now! I saw this at my house testing 2014 and I moved the server to a new VM host and the issue never appeared again. The weird thing is, both hosts were identical HP DL360G5 with 64GB ram and 2 quad core XEON.

The server at work we are running on is:
Dell M605, ESXi 5.x, 64 GB, two 6 core AMD 2.40Ghz

Are your virtual hardware and tools up to date?

We're on Hyper-V 2012. My VM specs are:

4 vCPU
12 GB RAM

[BanditBBS]

Are your virtual hardware and tools up to date?

We're on Hyper-V 2012. My VM specs are:

4 vCPU
12 GB RAM

Working on getting myself access to Vcenter so I can look at this myself.

[belvdr]

Working on getting myself access to Vcenter so I can look at this myself.

Great! If I can be of any help, let me know. Here's the VMware KB on virtual hardware versions for various products:

http://kb.vmware.com/selfservice/micros ... Id=1003746

I haven't experienced any issue by upgrading virtual hardware and tools before, so it's a fairly safe option. Be sure to update tools first, then hardware.

[belvdr]

Were you able to find anything interesting in the VM hardware?