Hello,
I have been testing vROps for about 5 weeks, but I am still not able to understand the stress percentage. From all I have managed to read, stress should be the time and area covered in the stress zone?
All my stress calculations seems the same, it is so way off from what I would expect by looking at the graph. I can clearly see that it is not spending 51% of its time in the stress zone.
I though maybe it got its high stress value from the ESXi host it is running on? I see that the host have had some stress, but here as well I clearly see that is has not spent 99% of its time in the stress zone.
I am using vROps 6.0.1 from a fresh install. Does someone have a clue on what is going on/what is wrong?
Thanks for any help/insight
I contacted support and got an answer
That score is not regarding the average percent stress over the 30 days, but rather the value of the highest peak of stress during that window.
This official VMware feature video explains the Stress calculation to great detail. I recommend you have a look through, and then let me know if you have any other questions:
http://featurewalkthrough.vmware.com/#!/vsphere-with-operations-management-6/performance-and-health-...
(Circa 4:20 into the video that value is explained)
That explains it. I wish they could have marked the graph for which period/peak they have used for stress calculation. I think that would have made it easier to understand.
The period/peak is default set for 60 minutes, but it can be changed in the policy. Changing it does not instantaneously update the stress score. I assume it will have to wait for a scheduled recalculation.
Same issue here. I don't understand how stress is calculated. As you can see, CPU wasn't in stress zone for 160.44% of it's time. Stress is calculated, by default, on a 60 minutes period, but it wasn't in the stress zone in the last hour.
I contacted support and got an answer
That score is not regarding the average percent stress over the 30 days, but rather the value of the highest peak of stress during that window.
This official VMware feature video explains the Stress calculation to great detail. I recommend you have a look through, and then let me know if you have any other questions:
http://featurewalkthrough.vmware.com/#!/vsphere-with-operations-management-6/performance-and-health-...
(Circa 4:20 into the video that value is explained)
That explains it. I wish they could have marked the graph for which period/peak they have used for stress calculation. I think that would have made it easier to understand.
The period/peak is default set for 60 minutes, but it can be changed in the policy. Changing it does not instantaneously update the stress score. I assume it will have to wait for a scheduled recalculation.
OK, so when we get a spike, we have to wait 30 days before the stress badge comes back to green (spike will disappear from graph)?