Abhishek Rawat has posted comments on this change. ( http://gerrit.cloudera.org:8080/15378 )
Change subject: IMPALA-9466: impala-shell client retry for hs2-http protocol ...................................................................... Patch Set 6: (1 comment) http://gerrit.cloudera.org:8080/#/c/15378/6/shell/impala_client.py File shell/impala_client.py: http://gerrit.cloudera.org:8080/#/c/15378/6/shell/impala_client.py@933 PS6, Line 933: _do_hs2_rpc > yeah, it would be good to understand why we still see failed rpcs then. its I calculated the time for rpcs and in all cases its much lesser than 240s. It probably does depend on lot of other factors, but in general it should be much less than 240s. And so most cases where nginx reloads worker processes, there are no connection drops. I have however seen nginx still drop connections (and not wait for graceful timeout of 240s), if there are multiple reloads in a very short period of time which is a possible scenario and something we've seen. I have a test which reproduces this scenario. I think, in general the assumption that there will always be a stable connection isn’t really true for cloud deployments. -- To view, visit http://gerrit.cloudera.org:8080/15378 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: Impala-ASF Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I0da9e9e8d34a340eaf763397cc095ff6260d65d5 Gerrit-Change-Number: 15378 Gerrit-PatchSet: 6 Gerrit-Owner: Abhishek Rawat <[email protected]> Gerrit-Reviewer: Abhishek Rawat <[email protected]> Gerrit-Reviewer: David Knupp <[email protected]> Gerrit-Reviewer: Impala Public Jenkins <[email protected]> Gerrit-Reviewer: Sahil Takiar <[email protected]> Gerrit-Comment-Date: Tue, 17 Mar 2020 18:30:44 +0000 Gerrit-HasComments: Yes
