[velero] fix: consistent node-agent podLabels #623
Conversation
The node-agent was using ".Values.podLabels" in the DaemonSet while using ".Values.nodeAgent.labels" in the podMonitor. This led to an inconsistency and made it impossible to have different labels for the node-agent compared to the Velero deployment. Signed-off-by: foyerunix <foyerunix@foyer.lu>
| @@ -557,6 +557,10 @@ nodeAgent: | |||
| # labels to set for the node-agent daemonset. Optional. | |||
| labels: {} | |||
|
|
|||
| # Additional pod labels for the node-agent daemonset. Optional | |||
| # ref: https://kubernetes.io/docs/concepts/overview/working-with-objects/labels/ | |||
| podLabels: {} | |||
There was a problem hiding this comment.
Isn't there is the label for the node-agent?
helm-charts/charts/velero/values.yaml
Lines 557 to 558 in f0f07de
There was a problem hiding this comment.
Hello,
Yes this is where the labels for the node-agent DaemonSet are defined. Meanwhile the PodMonitor will not match the labels of the DaemonSet but the labels of the pods created by the DaemonSet.
The labels of the pods of the node-agent DaemonSet are defined as ".Values.podLabels", hence they will be the same as the labels of the Velero Deployment. But the PodMonitor will try to match the labels defined in ".Values.nodeAgent.labels".
So with the current chart it is impossible to have different labels between the Velero Deployment and the Velero node-agent.
Best Regards.
Hello,
The node-agent was using ".Values.podLabels" in the DaemonSet while using ".Values.nodeAgent.labels" in the podMonitor.
This led to an inconsistency and made it impossible to have different labels for the node-agent compared to the Velero deployment.
Best Regards.