[Documentation] TTL_SECONDS_AFTER_WORKFLOW_FINISH

[Bobgy]

pipelines/backend/Dockerfile.persistenceagent

Line 21 in 6189681

ENV TTL_SECONDS_AFTER_WORKFLOW_FINISH 604800

the env var works for deployment ml-pipeline-persistenceagent
It deletes workflow resources after it is completed and reaches a TTL

[Bobgy]

There should be documentation how this should be configured and in which situations.

[Bobgy]

References:

• recommendations for kubernetes upper limit for CRs https://github.com/kubernetes/enhancements/blob/master/keps/sig-api-machinery/20180415-crds-to-ga.md#scale-targets-for-ga

[jingzhang36]

Let me try to see if we can have multiple controlling factors here. For example, if workflow controller detects either a certain number of workflows have been reached, or the workflow is older than ttl, the controller can gc the oldest workflow. In other words, ttl is the maximum time of a workflow, which can gc'ed earlier if the other constraint (workflow count upper limit) is reached.

[Bobgy]

/lifecycle frozen

[juliusvonkohout]

/close

Because it seems to be working

[google-oss-prow]

@juliusvonkohout: Closing this issue.

In response to this:

/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.