Closed arogic closed 1 year ago
I got around the issue by setting node taints on my AKS Windows Node Pool.
"node_taints": [ "app.kubernetes.io/name=solr:NoSchedule" ]
However it would be great, if NodeSelector would be carried over from the values file to the job.
Description
We have an AKS Kubernetes Cluster with Linux and Windows Node Pools. We deploy the zookeeper-operator helm chart through the solr-operator helm chart. When we delete the solr-operator using helm, the "solr-operator-zookeeper-operator-pre-delete" job gets scheduled on the Windows NodePool which obviously fails. Is there a way to pass NodeSelector values over for the "solr-operator-zookeeper-operator-pre-delete" job to be scheduled on the Linux Node Pool?
Importance
Location
Suggestions for an improvement
Option in the helm chart values file to define nodeSelector for the "solr-operator-zookeeper-operator-pre-delete" job.
Thanks in advance.