Andrew Wong has posted comments on this change. ( http://gerrit.cloudera.org:8080/17211 )
Change subject: [test] KUDU-3266 Fix flakiness in dynamic_multi_master test ...................................................................... Patch Set 2: Code-Review+2 (1 comment) http://gerrit.cloudera.org:8080/#/c/17211/2/src/kudu/master/dynamic_multi_master-test.cc File src/kudu/master/dynamic_multi_master-test.cc: http://gerrit.cloudera.org:8080/#/c/17211/2/src/kudu/master/dynamic_multi_master-test.cc@607 PS2, Line 607: // We can run into table not found error in cases where the : // previously paused master that's leader of prior term resumes : // and the up to date follower doesn't become leader and the resumed : // master from previous term isn't up to date. See KUDU-3266 for details. : ASSERT_EVENTUALLY([&] { : NO_FATALS(cv.CheckRowCount(table_name, ClusterVerifier::EXACTLY, 0)); : }); It might be interesting to explore different ways to ensure consistency at least within the same client for master operations, i.e. "read-your-writes" for master operations. For tablet servers we do this when the client specifies RYW mode by propagating timestamps to the client, having clients supply the highest seen timestamp prior to each scan request, and waiting on the tablet servers until the timestamp is considered safe to scan (similar to a snapshot scan, but with a Kudu-specified timestamp). This seems like a reasonable fix for now though. -- To view, visit http://gerrit.cloudera.org:8080/17211 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: Ifac1d95707064b6ac2624d3f52336d6c39afd3c8 Gerrit-Change-Number: 17211 Gerrit-PatchSet: 2 Gerrit-Owner: Bankim Bhavsar <[email protected]> Gerrit-Reviewer: Andrew Wong <[email protected]> Gerrit-Reviewer: Bankim Bhavsar <[email protected]> Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Comment-Date: Mon, 22 Mar 2021 23:54:28 +0000 Gerrit-HasComments: Yes
