neils-dev commented on PR #4140: URL: https://github.com/apache/ozone/pull/4140#issuecomment-1404551502
Thanks @xBis7. I took a look at the prom `omha_metrics_ozone_manager_ha_leader_state` metric when it is tracked and charted on om leadership transition. I rendered the prometheus endpoint with the prometheus web app and performed failover with _**2 om nodes**_. On failover currently with the extra tag for "_state_" we get extra traces, in this case 4, one for each state change and gauge change, see - https://github.com/neils-dev/play/blob/main/images/failover_extra_traces.png. It would be much cleaner to keep the tags to a min and just use the gauge to reflect the leader and changes to the leader. This can be seen when failover is rendered on prometheus with simplified tags, two traces this time, see - https://github.com/neils-dev/play/blob/main/images/failovers_just_gauge.png. Failover is the criss-cross when rendered. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
