Support Questions
Find answers, ask questions, and share your expertise

Unable to process alert for an invalid cluster named

Explorer

i am getting following logs in my ambari-server logs . I check the database and there is no entry of this cluster name "oldclustername" in any tables(i grepped on my dump file and also checked manually in all tables). I named my cluster as "oldclustername" but then renamed it to "newclustername" but logs are still showing these messages. Only table I see suspicious entries is ambari.request which has cluster id=-1 and 2, where cluster_id correctly maps to my newclustername but -1 looks like a wrong entry. Any suggestions where should I look?

23 Jun 2016 23:38:32,233 ERROR [alert-event-bus-2] AlertReceivedListener:329 - Unable to process alert yarn_nodemanager_health for an invalid cluster named oldclustername
23 Jun 2016 23:38:32,361 ERROR [alert-event-bus-2] AlertReceivedListener:329 - Unable to process alert ams_metrics_monitor_process for an invalid cluster named oldclustername
23 Jun 2016 23:38:32,362  WARN [alert-event-bus-2] AlertReceivedListener:248 - Cluster lookup failed for clusterName=oldclustername
23 Jun 2016 23:38:32,362  WARN [alert-event-bus-2] AlertReceivedListener:134 - Received an alert for ambari_agent_disk_usage which is a definition that does not exist anymore
23 Jun 2016 23:38:32,362 ERROR [alert-event-bus-2] AlertReceivedListener:329 - Unable to process alert kafka_broker_process for an invalid cluster named oldclustername
23 Jun 2016 23:38:32,942 ERROR [alert-event-bus-2] AlertReceivedListener:329 - Unable to process alert datanode_webui for an invalid cluster named oldclustername
23 Jun 2016 23:38:32,943 ERROR [alert-event-bus-2] AlertReceivedListener:329 - Unable to process alert datanode_process for an invalid cluster named oldclustername
23 Jun 2016 23:38:32,943  WARN [alert-event-bus-2] AlertReceivedListener:248 - Cluster lookup failed for clusterName=oldclustername
23 Jun 2016 23:38:32,943  WARN [alert-event-bus-2] AlertReceivedListener:134 - Received an alert for ambari_agent_disk_usage which is a definition that does not exist anymore
23 Jun 2016 23:38:32,943 ERROR [alert-event-bus-2] AlertReceivedListener:329 - Unable to process alert yarn_nodemanager_webui for an invalid cluster named oldclustername
23 Jun 2016 23:38:32,943 ERROR [alert-event-bus-2] AlertReceivedListener:329 - Unable to process alert ams_metrics_monitor_process for an invalid cluster named oldclustername
23 Jun 2016 23:38:32,943 ERROR [alert-event-bus-2] AlertReceivedListener:329 - Unable to process alert yarn_nodemanager_health for an invalid cluster named oldclustername
1 ACCEPTED SOLUTION

Accepted Solutions

Super Collaborator

There is a bug on older Ambari versions where if you rename a cluster (or delete a cluster), you'll get these errors in your logs. It's because the agents have not un-scheduled the old cluster's alerts. You can just restart the agents to have these errors go away.

View solution in original post

3 REPLIES 3

Super Collaborator

There is a bug on older Ambari versions where if you rename a cluster (or delete a cluster), you'll get these errors in your logs. It's because the agents have not un-scheduled the old cluster's alerts. You can just restart the agents to have these errors go away.

View solution in original post

Explorer

I restarted the agents but looks like the errors are still present. I see the old cluster name in a execution_commands table. I have identified the task_id which are having the command haivng my old clustername. Should I delete them from the database?

Did you solved the issue? We have the same erros in our 2.4.2 cluster.

ERROR [alert-event-bus-2] AlertReceivedListener:341 - Unable to process alert ams_metrics_monitor_process for an invalid cluster named XXXX.

WARN [qtp-ambari-client-68] ViewRegistry:855 - Could not find the cluster identified by XXXX.