User and organization data for Docker EE is stored in a RethinkDB database which is replicated across all manager nodes in the UCP cluster.
UCP constantly monitors the health of the replicas and alerts when one or more of the replicas are unhealthy.
Rethinkdb cluster unhealthy:
m replicas are healthy.
There are currently
m RethinkDB replicas in the system, but only
n of them are healthy.
- You can look at the logs by running
docker logs ucp-controller 2>&1 | grep "rethinkDB table status"
- You can also try to troubleshoot by using the instructions posted here.
- If you cannot find what is wrong, collect the support dump and file a support ticket.
Since this problem is in the distributed database of UCP, you should not add or remove nodes from your cluster until the problem is solved. Otherwise, it can lead to data loss.