I have created a Confluence 7.11.1 cluster in AWS EKS. I have followed all the steps in https://confluence.atlassian.com/doc/set-up-a-confluence-data-center-cluster-982322030.html.
When I deploy the first pod (in a StatefulSet), there are no issues and I can see the Node populate in the UI under 'Clustering'. When I have the second pod deployed, the first Node disappears and the second node in only visible.
But if I remove the second node and only leave the first node, it will fail with one of the following:
- Clustered Confluence: Database is being updated by an instance which is not part of the current cluster. You should check network connections between cluster nodes, especially multicast traffic.
- 500 error: Hazelcast instance is not active!
To fix the issue I destroy all pods and only have one pod running.
I have created a Jira 8.14.0 cluster in AWS EKS without an issue using the default discovery, so I assume this issue is to do with Hazelcast. I have used Multicast as I am unable to use the other methods due to:
