-
Bug
-
Resolution: Unresolved
-
Undefined
-
4.19
-
Incidents & Support
-
False
-
-
False
-
None
-
-
None
Description of problem:
In virtual machine details and metrics CPU usage graph displays 'no data available' even as a cluster-admin, memory and disk is displayed without issues. When clicking on 'no data available' hyperlink the metrics are visible in Prometheus for all VMs
Version-Release number of selected component (if applicable):
OCP-V 4.19.12
How reproducible:
This affect 2 customer clusters and all VMs
Actual results:
All VMs shows 'no data available'
Expected results:
CPU usage is reported
Additional info:
2 different clusters have the same issue - all the VMs shows No data available for the CPU, just for CPU in Overview -> Utilization -> CPU and also in VM -> Metrics -> CPU
No change when we switch from 5min to different time, if you click on the no data available it will redirect you to the Prometheus metrics that works fine . Also not related to one browser, Edge/Firefox the same
I tried console debug but that shows data from the prometheus
----------
11:24:20.997 XHRGET
https://console-openshift-console.apps.xxx-xxx.xxx.xxx/api/prometheus-tenancy/api/v1/query_range?start=1763561928.697&end=1763562228.697&step=5&namespace=virtual-machines&query=sum(rate(kubevirt_vmi_cpu_usage_seconds_total{name='rhel9-blush-crocodile-66',namespace='virtual-machines'}[5m]))+BY+(name,+namespace)
[HTTP/1.1 200 OK 34ms]
response JSON
status "success"
data { resultType: "matrix", result: (1)[…], analysis: {} }
resultType "matrix"
result [ {…} ]
0 { metric: {…}, values: (61)[…] }
metric { name: "rhel9-blush-crocodile-66", namespace: "default" }
values (61)[ (2)[…], (2)[…], (2)[…], (2)[…], (2)[…], (2)[…], (2)[…], (2)[…], (2)[…], (2)[…], … ]
0 [ 1763587585.437, "0.08154084071418519" ]
1 [ 1763587590.437, "0.08154084071418519" ]
2 [ 1763587595.437, "0.08048148148148156" ]
3 [ 1763587600.437, "0.08048148148148156" ]
4 [ 1763587605.437, "0.08048148148148156" ]
5 [ 1763587610.437, "0.08048148148148156" ]
6 [ 1763587615.437, "0.08118518518518546" ]
7 [ 1763587620.437, "0.08118518518518546" ]
8 [ 1763587625.437, "0.08118518518518546" ]
9 [ 1763587630.437, "0.08118518518518546" ]
10 [ 1763587635.437, "0.08118518518518546" ]
----------
This happens for all the users even for kubeadmin so it should not be something with cluster rights for specific users, i tried the same version in reproducer 4.19.12 but CPU shows metrics for all VMs
