Created attachment 1118385 [details] evm.log attached. Description of problem: Metrics does not show information for pods/containers when selecting Utilization information. Version-Release number of selected component (if applicable): 5.5.2.2 How reproducible: Always Steps to Reproduce: 1.Setup CFME to enable Metrics 2.Select a Pod in CFME : Containers --> Pods 3.Select Monitoring --> Utilization Actual results: Error is shown: No Utilization Data Available Expected results: Utilization data + graphs should be seen. Additional info: EVM log attached.
The error seems related to the fact that there are no information on the nodes cores. Please go into the nodes summary page and check that they all have capacity information (cpu and memory). Sadly this information is not available in the logs. Feel free to ping me on IRC and we'll look at it together.
Created attachment 1118393 [details] screenshot.png Looking in Einat's environment I found one node that was defined but never reported any information (see screenshot). Since we have fixed bug 1288549 we are now able to deal with nodes that never reported any information but now they're breaking metrics collection.
Best way to address this is to catch the validation error and return empty metrics for the relevant object.
I think that this is easy to reproduce. It's just a matter to add a new dummy node in OpenShift (that doesn't exists) and see what breaks in ManageIQ (inventory refresh and metrics collection). Yaacov can you give it a try? Thanks.
Submited upstream: https://github.com/ManageIQ/manageiq/pull/8213
Yaacov shouldn't this be moved to POST?
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2016:1348