Setting up HPA for Tekton Services

[naivedeveloper95]

I wanted to setup HPA for Tekton and it's related services (including tekon-pipelines, workloads and resolvers), however, in the repo - I only see reference for webhook-hpa.

https://github.com/tektoncd/pipeline/blob/main/config/webhook-hpa.yaml

Questions -

• Is there any reason for not having hpa for others?
• If we add multiple replica for tekton-event-controller, will end up redundant events?
• Any recommendation on best practices on HPA for tekton?
• For tekton-events-controller specifically: How does it handle event deduplication across replicas?
• Is there built-in coordination between multiple event controller instances?
• Are there any known issues or considerations when scaling tekton-controller beyond a single replica?
• What's the recommended approach for handling leader election in multi-replica tekton-controller deployments
• Are there any shared resources or locks that could cause conflicts with multiple controller replicas?

cc @tektoncd/operator-maintainers

[vdemeester]

cc @tektoncd/operator-maintainers as this is something we can enable from the operator side of things

[naivedeveloper95]

Sure. Let me do that. Thanks

[waveywaves]

Should we move this to https://github.com/tektoncd/operator

[vdemeester]

oh, I thought I did, yes indeed.