nworksVIC.exe taking up all the CPU 100%

[waxil]

Also vpxd.exe seems to be sharing the CPU, but between these two it's constantly at 100%

Forefront and OV have been stopped and the same symptoms are still showing, albeit with occasional drop offs, but on the whole it's still being maxxed out.

Has anyone else seen \ resolved this issue?

[Gostev]

Chris, I am not nworks guy, but AFAIK collectors are very resource intensive, which is why we recommend installing them on dedicated VMs. So the high load is normal. But this could be a sign of overload too, so you should make sure that you have enough deployed collectors to cover the environment you have.

[waxil]

Cheers Anton,

It has only recently started doing this as it's breaching it's own OV thresholds!
But nothing has changed within the environment..

I'm fairly certain that the VM it is running on has enough capacity also.

As I didn't install this collector, I'm not 100% how it should work or be configured.. Maybe we need an additional collector VM?

Chris.

[Gostev]

There are a few sizing calculators avaialble on this page under Documentation > Resources section:
http://www.veeam.com/vmware-esx-monitor ... tions.html

Particularly, deployment planning calculator:
http://www.veeam.com/support/nworks_deployment.html

You may want to check these out.

But if this did not happen before, and just started happening all of a sudden, you may also want to open a support case and have our technical staff look at the logs...

Thanks!

[waxil]

Just wanted to return to my post to explain what eventually fixed this issue.

It turns out that the VIC was running on a VCenter Server, which only had 1 x CPU core and 2gb ram..

Having logged a call with excellent support people at Veeam, they analysed my collector logs and noted that the virtual machine spec may not be up to the job.

So, having increased the virtual hardware spec on the VC VM to 2x CPU Cores and 3gb RAM (after discovering that you need the VC to do this unless you know which ESX host the VC VM is running on!!!! ) all is now much better and the VM is no longer breaching OpenView CPU thresholds.

It is still a very hard working system and even now with 2 Cores (vCPU's), it runs both constantly at over 50% utilization, but looks much healthier than it's previous constant 100%

So many thanks to the support peeps who assisted me in getting to the bottom of this issue and hopefully this forum post may assist others in bottoming out similar issues.


For info, I was unable to figure out the sizing calculators mentioned in a previous post by Gostev.. very confusing - some tech notes may be a good idea for whoever put these together as I couldn't work out how to get it to produce any useful figures or even where I was supposed to be entering our site specific information and numerics.

Many Thanks again,

Chris.

[Irina Simkina]

Hi Chris!

Thank you very much for the feedback! It's pleasent to hear that your problem was resolved.
The deployment calculator is mostly described in the Deployment guide. But not other sizing calculators indeed
We will add some description to make them more clear and usefull.

If you have any questions concerning the sizing please feel free to contact us.
Irina