Versions (relevant - OpenSearch/Dashboard/Server OS/Browser):
2.15.0
Describe the issue:
I’ve been testing out the remote-backed storage implementation as described here and want to understand the expected behaviour.
I have a cluster with remote-backed storage fully enabled, including state. If the cluster has a catastrophic failure, I would expect that it could be rebuilt from the remote state as described in the documentation.
When testing the feature, I noted that when the cluster was re-created after the failure whilst it did sync the .opendistro_security index - it remained in an uninitialized state. Are their additional steps required before the cluster is made functional?
The output suggested running the securityadmin.sh tool, but I didn’t expect this to be required, and would lose my existing security configuration.
Am I missing something here?