Philipp Sontag

Hey @lack, is there any chance you would be willing to update this branch and move the PR to the new repository at https://github.com/nolar/kopf? If you don't have time at...

> When we call retire_workers the scheduler handles the graceful shutdown and the worker process exits. Ideally this should allow the Kubernetes Pod to go into a Completed phase. I've...

> The plan is for the API call to return immediately. So the way we are nuking them right now is premature and we need to fix that. Ideally, we...

I am not familiar with your test setup but I assume you have not installed the peering objects in your ci cluster. The logs mention kopf._cogs.clients.watching:watching.py:122 Pausing the watch-stream for...

Hey @nolar did have a chance to look into this already?

As far as I know this currently the expected and documented behavior. See https://kopf.readthedocs.io/en/stable/results/#results-delivery

Then I guess the example is wrong. It is currently not possible see https://github.com/nolar/kopf/issues/319#issuecomment-912771812 for an easy workaround though. There is also the more generic feature request #827 but I...

Hey @nolar can you confirm that this is the expected behaviour? If the problem is unclear or you need more information please let me know.

Hey @nolar would you accept a PR that includes the operation in the handler id?

I have seen that there were some promising changes merged for beta.4. I think the [mulitenancy docs](https://github.com/Azure/azure-service-operator/blob/main/docs/v1/design/multitenancy.md) mention the new approach under option 2 but there is no further documentation...