問題

When attempting to start Confluence Data Center, you receive the following message in atlassian-confluence.log:

2017-05-31 21:04:28,952 ERROR [localhost-startStop-1] [atlassian.config.bootstrap.DefaultAtlassianBootstrapManager] init Home is not configured properly: 
com.atlassian.config.ConfigurationException: Exception bootstrapping cluster:the cluster is running build XXXX of Confluence, while this node is running YYYY

The values for XXXX and YYYY refer to separate Confluence Build Numbers, and will vary with your environment.

This then leads to Confluence does not start due to Spring Application context has not been set; as the configuration has not been completed. Despite the presence of the Spring Application Context error, this is not a permissions based problem - the root cause is the original configuration problem.

原因

A second node is using the same multicast IP address and is running a different version of Confluence.

診断

There are two possible scenarios where this error might occur:

• A separate instance of a different version, that simply has the same multicast information as your original server
• A different version of the same instance, connecting to the same database

If your two instances are separate (i.e, they have a different database) then simply adjusting the multicast address to a different value will resolve the problem.

If there are multiple versions in the cluster (and they connect to the same database), look to see which is larger:

• If the cluster is running the smaller build number, check that node has failed to start up (as described in Confluence will not start up because the build number in the Home Directory doesn't match the build number in the Database, after upgrade)
• If the node is running the smaller build number, it suggests that the database has been upgraded by another node, and you should install the same version of Confluence

Contact Atlassian Support for assistance with troubleshooting this error.

回避策

Shutting down all other instances of Confluence (and then re-starting the original instance) should resolve the problem

ソリューション

Verify the cluster settings that are present in confluence.cfg.xml are valid for nodes in that cluster, and all nodes are starting on the same version of Confluence. Ensure other instances of Confluence do not have the same cluster information