Clustered instance does not start with Unicast listening address (not allowed to create a new cluster)
This article applies to Confluence clustered 5.4 or earlier.
Clustered instance is configured with Unicast via WKA (well known addresses: How to Configure a Cluster Without Multicast Traffic )
The following appears in the
2013-04-12 16:37:05,685 ERROR [Logger@1492390803 3.3.1/389] [Coherence] log 2013-04-12 16:37:05.685 Oracle Coherence GE 3.3.1/389 <Error> (thread=Cluster, member=n/a): Node /fe80:0:0:0:cdae:4065:a6ca:dc4d%14:8088 is not allowed to create a new cluster; WKA list: [/172.16.1.70:8088, /172.16.1.70:8089] 2013-04-12 16:37:05,705 ERROR [main] [atlassian.confluence.setup.ConfluenceConfigurationListener] contextInitialized An error was encountered while bootstrapping Confluence (see below): Failed to start Service "Cluster" (ServiceState=SERVICE_STOPPED, STATE_ANNOUNCE) java.lang.RuntimeException: Failed to start Service "Cluster" (ServiceState=SERVICE_STOPPED, STATE_ANNOUNCE)
Same cause as this: Configuration of Confluence Cluster Fails with 'Cannot assign requested address' :
There's a known issue with the JVM and IPv6, particularly on Linux. Refer to Data Center Troubleshooting for more information.
ResolutionThe workaround is to tell the JVM to prefer IPv4, by adding -Djava.net.preferIPv4Stack=true to JAVA_OPTS (This can be set in
CONFLUENCE_INSTALL/bin/setenv.sh). See Configuring System Properties for specifics on how to set this property.