Neil Prasad created FLINK-35340:
-----------------------------------
Summary: Flink Kubernetes Operator CRDs not applying correctly
Key: FLINK-35340
URL: https://issues.apache.org/jira/browse/FLINK-35340
Project: Flink
Issue Type: Bug
Components: Kubernetes Operator
Affects Versions: kubernetes-operator-1.8.0, kubernetes-operator-1.7.0
Reporter: Neil Prasad
>From version 1.7.0 onwards, CRDs include an additional parameter for
>additionalPrinterColumns, "priority":
{code:java}
- additionalPrinterColumns:
- description: Last observed state of the job.
jsonPath: .status.jobStatus.state
name: Job Status
priority: 0
type: string
- description: "Lifecycle state of the Flink resource (including being
rolled\
\ back, failed etc.)."
jsonPath: .status.lifecycleState
name: Lifecycle State
priority: 0
type: string {code}
When applying the CRDs initially or as an upgrade the CRD doesn't apply
correctly and omits the newly added "priority" field. This is both tested on a
fresh install via Minikube as well as an existing GKE deployment.
For what it's worth, the versions test on is k8s v1.29.3. When applying these
CRDs, there is no error message that comes up and nothing appears in the api
server logs.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
