planetscale / planetscale/vitess-operator
[Bug] Using "latest" image tag breaks versioned deployments of operator
Nobody has claimed this yet.
- Dominant language
- Go
- Stars
- 372
- Forks
- 102
- Avg merge
- 3d 5h
- Merged PRs (30d)
- 15
Description
The deployment manifest of the vitess-operator uses the latest tag (https://github.com/planetscale/vitess-operator/blob/1001a86b8a9787180a1cd47a0d88a121da712d61/deploy/operator.yaml#L19C1-L19C50) which breaks this deployment in case the k8s configuration at a tagged branch is incompatible with the latest image.
I found this when tried to deploy the manifests of version v2.14.0 of the operator:
kubectl apply -k "github.com/planetscale/vitess-operator/deploy?ref=v2.14.0"
The latest version of the image requires access to the horizontalpodautoscalers, but the role definition at the v2.14.0 tag (https://github.com/planetscale/vitess-operator/blob/v2.14.0/deploy/role.yaml) does not yet include permissions for those resources as it does in the v2.15.0 tag: (https://github.com/planetscale/vitess-operator/blob/v2.15.0-rc1/deploy/role.yaml)
rules:
...
...
- apiGroups:
- autoscaling
resources:
- horizontalpodautoscalers
verbs:
- '*'
I would suggest matching the tag either directly in the deploy/operator.yaml or by adding a section like
images:
- name: planetscale/vitess-operator
newTag: <version>
in deploy/kustomization.yaml
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Compare deploy/operator.yaml and deploy/kustomization.yaml at the versioned tags, starting with the v2.14.0 manifest and its image reference. Check how the image tag is selected and verify that applying the versioned Kustomize deployment uses a compatible operator image and permissions. Done means a tagged deployment no longer pulls latest and can be applied with the documented kubectl apply -k command.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- kubernetes
- Domain
- devops, infrastructure
- Issue type
- Bug
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Activity status
- Stale
- Clarity
- Clearly specified
- Newbie friendliness
- 35/100