I have a Object Store Cluster deployed in 11.5.
For reasons I had to unregister the Object Store. When I registered the Object stores again, they failed to validate in the server validation endpoint (admin/data/items/cloudStores/...). No issues in describedatastore.
How to fix the cluster if it fails to establish?
The health checks are inconsistent across machines and there is no documentation explaining the parameters:
DeploymentMode: cluster
finalized: true
| Validation | ObjectStore1 | ObjectStore2 | ObjectStore3 |
| machine.overalhealth | Transitional | Transitional | Transitional |
| role | cluster_member | cluster_member | cluster_member |
| healthcheck.enable | false | false | false |
| status | Starting | Starting | Starting |
| isSCMhealthy | true | true | true |
| isOMHealthy | true | false | false |
| isDataNodeHealthy | false | false | false |
| isS3GHealthy | false | true | true |
| (isSCMPrimordialNode) | true (managerNumber 1) | (managerNumber 2) | (managerNumber 3) |
datastore.overalhealth: Unhealthy
1) What is SCM, OM, DataNode, S3G? Maybe StorageClusterManager, Object Manager, DataNode (for replication?), S3-Gateway internal?
2) How to fix the cluster? What is wrong?
My personal feeling is that the cluster is not syncing correctly. Only option I see is to delete the datastore config|data folder and start from scratch, but this involves many steps to prevent data loss, I´d like to avoid this. I would prefer a way to re-establish the cluster as-is.
Similar but other issue: Clean Install Object Store Cluster 11.5 Unhealthy - Esri Community
