Hello, I'm wondering what the proper strategy is for performing rolling restarts when running Elasticsearch in K8S with ECK?
Before we ran in k8s, we followed this guide: Full-cluster restart and rolling restart | Elasticsearch Guide [8.17] | Elastic
After switching to ECK, I was under the impression (maybe mistakenly) that this was no longer necessary. I thought that doing a 'rollout restart' on the stateful set(s) would cause the operator do that for you? Or change some other transient settings to help with a quicker restart?
We've had some very long restarts, typically because shards start to move around once a node comes back up.
I found this guide on github: Better Documentation for Performing Manual Rolling Restarts (of underlying hosts) of Persistent Clusters · Issue #5305 · elastic/cloud-on-k8s · GitHub. But I don't know if it's overkill.
I'm running version 2.3 of the operator, so I realize I could be missing newer features. I plan to update soon. But either way, I would love to see some documentation on how you are supposed to perform rolling restarts in ECK.
Thank you!