Shoothzj opened a new issue #7544: URL: https://github.com/apache/pulsar/issues/7544
We use the ability of auto update partition. During our performance test, there are some unstable requests which latency > 5s! (we set send timeout to 5 seconds too).Call back is failure. I check the server log and the zookeeper log, there is fine. (May be broker zookeeper timeout is 30, doesn't print log). Also check the client, server, zookeeper garbage collect log. They < 1s. Then I think may be partition update request is too frequecy ? The metadata request and msg request share a eventloop. My client process(two process instances) has 50 pulsar client, whcih owns average the 60 topic. May be I should adjust auto update requests period to 3 or 5 minute? ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected]
