Issue with Jira Data Center Cluster Index Replication

Pola_ Venkat

We are encountering an issue with Cluster index replication in our Jira Data Center environment, getting a message “The replication for cluster node 'jira-uat-node2' is delayed by 25,221 seconds”. Has anyone else experienced a similar problem? If so, we would greatly appreciate your assistance with resolving it.

Environment Details:

• We recently migrated from Jira Server (version 8.20.9) to Jira Data Center (version 9.12.11).
• Our platform migration involved moving from Windows Server 2016 (with SQL Server DB v13) to Linux (RHEL 7, with SQL Server DB v2016).
• We have set up clustering with Docker containers (Node 1 and Node 2) for load balancing.

Current Issue:

It seems the cluster synchronization issue might be related to firewall port configurations. We have configured port 40001, but no traffic is detected (inbound or outbound). However, we do observe traffic on port 2377.

Has anyone encountered this issue or have any suggestions for resolving it?

 

 

 

 

 

Radek Dostál

Not sure about 2377, that shouldn't be part of Jira (maybe you might find out what process is using it). But https://confluence.atlassian.com/kb/ports-used-by-atlassian-applications-960136309.html also mentions 40011 for Ehcache object, so maybe that one is missing from whitelisting?

You might also find more info in the app logs on either of the nodes, if they can't talk, they likely will throw something of interest in there, but since you only mentioned 40001 I'd try adding 40011 if that does anything.