capistrant commented on PR #19139: URL: https://github.com/apache/druid/pull/19139#issuecomment-4083717240
> marking this as draft while I evaluate a competing approach that uses pod phase instead of readiness I think that I prefer readiness to pod phase. being ready means that you are ready to announce yourself to the cluster. If a readiness probe is properly configured using our recommendations, if the probe fails enough to cross threshold of being not ready, the service is no longer ready to be announced and should stop getting traffic. also in protecting against the worst case of a pod having announce labels but the container having been killed without proper shutdown + not being rescheduled, readiness will properly handle undiscovering such a service nearly immediately upon crash. -- 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]
