Node02 unable to join cluster after upgrade from 7.13.0 to 7.13.7

[Wim Abts]

Hi, 

We had a working Confluence TST cluster with 2 nodes on version 7.13.0.

After updating both nodes to version 7.13.7 we have the following behavior:

• node 1 starts just fine
• once node 1 is completely started we start node 2
• In the logs we see error messages about the authentication to the cluster and asking to verify the cluster name and secret (both are ok and the same on both nodes)
• Node 2 than takes the cluster and node 1 get's an error 'Could not open Hibernate Session for transaction; nested exception is com.hazelcast.core.HazelcastInstanceNotActiveException: Hazelcast instance is not active!'

When we then restart node 1, we have the same behavior (node 1 is ok but node 2 crashes with the same error).

So there seems to be a problem with the cluster after the update.

On both machines port 5801 is open on the firewall.

Errors: 

confluencetst02 | 2022-06-15 08:43:04,086 INFO [Catalina-utility-1] [atlassian.confluence.cluster.DefaultClusterConfigurationHelper] lambda$populateExistingClusterSetupConfig$5 Populating setup configuration if running with Cluster mode...

confluencetst02 | 2022-06-15 08:43:04,842 INFO [Catalina-utility-1] [confluence.cluster.hazelcast.HazelcastClusterManager] configure Configuring Hazelcast with instanceName [confluence], join configuration TCP/IP member addresses: 153.89.154.102|153.89.154.100, network interfaces [172.19.0.3] and local port 5801

confluencetst02 | 2022-06-15 08:43:04,842 INFO [Catalina-utility-1] [confluence.cluster.hazelcast.HazelcastClusterManager] startCluster Starting the cluster.

confluencetst02 | 2022-06-15 08:43:08,219 INFO [hz.confluence.event-2] [confluence.cluster.hazelcast.LoggingClusterMembershipListener] memberAdded [153.89.154.100]:5801 joined the cluster

confluencetst02 | 2022-06-15 08:43:08,221 INFO [hz.confluence.event-2] [confluence.cluster.hazelcast.LoggingClusterMembershipListener] logClusterMembers Cluster now has 1 members: [[153.89.154.100]:5801]

confluencetst02 | 2022-06-15 08:43:09,483 WARN [hz.confluence.cached.thread-2] [hazelcast.interceptor.authenticator.DefaultClusterJoinManager] checkNodeAuthenticationEnabled CONNECT(172.19.0.3:55333 -> 153.89.154.102:5801): Node authentication failed: Cluster authentication failed. Please make sure all members share the same value for 'confluence.cluster.name' and 'confluence.cluster.authentication.secret' in confluence.cfg.xml.

confluencetst02 | 2022-06-15 08:43:10,330 INFO [Catalina-utility-1] [confluence.cluster.hazelcast.HazelcastClusterManager] startCluster Confluence cluster node identifier is [91d9cc86]

confluencetst02 | 2022-06-15 08:43:10,330 INFO [Catalina-utility-1] [confluence.cluster.hazelcast.HazelcastClusterManager] startCluster Confluence cluster node name is [confluencetst02]

Any ideas?

[D_Ante B]

The WARN that you are seeing is in reference to your confluence.cfg.xml in your home directory. It looks like the following two properties:

<property name="confluence.cluster.authentication.enabled">true</property>

<property name="confluence.cluster.authentication.secret">some_secret</property> 

Verify that the secret is the same on both your servers. After making them the same for our servers, all nodes were able to join the cluster