d-c-manning commented on PR #2061: URL: https://github.com/apache/phoenix/pull/2061#issuecomment-2680790847
Makes sense to me. But then if this is a heavily used code path, isn't a metric a better choice than a log? It sounds like you would want a threshold that is <1 ms, but that threshold can be exceeded on a JVM pause, or on any RPC call. It will be noisy, but it sounds like if it's much greater than 1ms, then it won't be useful to you. I understand there was some regression that had more metadata RPC calls than expected... but that it will still be normal for the occasional plan to require a metadata RPC. And if so, that wouldn't necessarily be a problem that deserves a log? Would the metric be sufficient then? Maybe I misunderstand the value we get from the log vs. the metric. CC @tkhurana -- 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: issues-unsubscr...@phoenix.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
