Michael Morello

Although we still need to write proper documentation, here are the steps to connect to a hosted deployment in https://cloud.elastic.co/deployments from an ECK managed Elasticsearch cluster: > [!WARNING] > The...

I think this one is not that technically involved. The question is more how customizable do we want to make the internal passwords? Is it just a fixed length? Should...

I guess you mean here: https://github.com/elastic/cloud-on-k8s/blob/54ec0e397caf2a5439a654d634104619a740bc07/deploy/eck-operator/templates/operator-network-policy.yaml#L23-L27?

Could you please provide the operator logs or the `Elasticsearch` status subresource when the upgrade was not progressing? From your comment it is not clear if the upgrade was blocked...

Thanks. This predicate is indeed designed to handle that situation: a node can be restarted even if the cluster state is yellow as long as the node version does not...

Is it somewhat a duplicate of https://github.com/elastic/cloud-on-k8s/issues/8429?

Without questioning the need for a fix I think this only happens if there is only one master node left to upgrade, and that Pod is evicted/deleted. For example deleting...

> Was Elasticsearch working as expected during this test? Cluster was green with the following indices: ``` health status index uuid pri rep docs.count docs.deleted store.size pri.store.size sth green open...

I'll try to reproduce it in my dev env, with log level increased.

FWIW the last successful run was on Saturday `2024-04-27T23:06:08.133397756Z` , first failure is reported on Monday `2024-04-28`. There is no commit in ECK during that time frame 😕