Abhishek Dasgupta

Closing this issue as it is addressed after compaction enablement. Please reopen if needed

/close

This issue is old issue but we need to validate the requirement of the test cases in the current context and write test cases for the scenarios which are now...

We don't have proper test cases for ETCD events that uses empty storage provider. So, codes written for ETCD main, which has proper storage provider set, may not work with...

We need to observe the metrics mentioned in this issue in our usage to come up with the right values per cloud provider

@DelinaDeng It shouldn't be related to this issue. The issue you mentioned seems due to pods not running the container at all due to unavailability of resources or scheduling problem....

As per an out of band discussion with @shreyas-s-rao , we decided to consider `jobs_total` with failed label as metrics to raise alert upon crossing a threshold both for shoots...

Any progress to this? Did we find which ETCD version to use?

Hi Kesavan, I created a branch https://github.com/abdasgupta/etcd-druid/tree/perf-test . I made some changes in https://github.com/abdasgupta/etcd-druid/blob/b87c104637a7719bf6ce9cdbf72ae47e49d18148/pkg/component/etcd/configmap/configmap.go#L161-L168 to include the flags you requested. Normally, we provide flag values through ETCD CR. But as...

Please follow: https://github.com/kubernetes-sigs/controller-runtime/blob/master/designs/use-selectors-at-cache.md