The cluster page of worker nodes fails to load and shows the message "The latest cluster data is not available, most likely primary node is not accessible" while the nodes in the cluster page loads fine in the primary node
The errors in the runtime.log will match the below snippet
Log location:/var/log/vmware/loginsight/runtime.log
[1234-12-12 11:11:11.] ["main"/##.##.##.## INFO] [com.vmware.loginsight.daemon.DaemonClientStubManager] [Initializing daemon commands client pools.]
[1234-12-12 11:11:11.] ["main"/##.##.##.## INFO] [com.vmware.loginsight.daemon.LogInsightDaemon] [Exception during setup environment] [Exception during setup environment]
[1234-12-12 11:11:11.] ["main"/##.##.##.## FATAL] [com.vmware.loginsight.daemon.LogInsightDaemon] [Error starting services]
com.vmware.loginsight.config.MasterFQDNUnresolvableException: <hostname> cannot be resolved and there is not master IP buffer
[1234-12-12 11:11:11.] ["BucketManager-thread-1"/##.##.##.## ERROR] [com.vmware.loginsight.commons.rpc.clientconnpool.ClientPoolsHolder] [Need IP to create Thrift client. Got '<hostname>']
Aria operations for logs 8.x
The inter node communication client pools are constructed incorrectly when aria operations for logs is not able to correctly resolve provided master FQDN to an IP address
Note: Take cold snapshots of all the Aria operations for logs nodes in the cluster
cp /storage/core/loginsight/config/loginsight-config.xml#<highest_number> /storage/core/loginsight/config/loginsight-config.xml#<highest_number>.bak
vi /storage/core/loginsight/config/loginsight-config.xml#<highest_number>
service loginsight restart
This change is internal to Aria operations for logs hence it wont affect DNS entries. Aria operations for logs will still be accessible via FQDN.