Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/15323 )
Change subject: ksck: display quiecing-related info ...................................................................... Patch Set 2: Code-Review+1 (2 comments) http://gerrit.cloudera.org:8080/#/c/15323/1/src/kudu/integration-tests/tablet_server_quiescing-itest.cc File src/kudu/integration-tests/tablet_server_quiescing-itest.cc: http://gerrit.cloudera.org:8080/#/c/15323/1/src/kudu/integration-tests/tablet_server_quiescing-itest.cc@433 PS1, Line 433: ; > I'm a bit conflicted about this. It's not information that you'd really thi I think there is some value in seeing number of leader replicas and number of active scanners in ksck output, and I think it's nice to have that by default: it's a nice extension and probably we shouldn't be constrained here by backwards compatibility of the ksck output (if any). For the 'quiescing' column I'm not quite sure about including it by default into ksck. From the other side, given its rather dynamic nature (it appears only when there is at least one quiescing server), it looks a bit tricky to me to reason about. It would be great to understand the use-case here. Up to what extent do we want to use ksck to signal about on-going quiescing if it's already covered by a dedicated kudu CLI sub-command? Is this just to let the operator know that a cluster in not-so-regular mode of operation? If so, then maybe it should be added just as a note/warning somewhere instead of outputting the quiescing status of every server? Anyway, I don't feel strong about this. It would be great to get more feedback on this from other people. http://gerrit.cloudera.org:8080/#/c/15323/1/src/kudu/tools/ksck_remote.cc File src/kudu/tools/ksck_remote.cc: http://gerrit.cloudera.org:8080/#/c/15323/1/src/kudu/tools/ksck_remote.cc@81 PS1, Line 81: true > Re: backwards compatibility, I don't think it's unreasonable to just log a Yep, that makes sense to me. One question: once quiescing status is observed in ksck as in this patch, do we expect people to run quiesce-specific sub-command anyways or this output from ksck will be enough to collect all the necessary information? -- To view, visit http://gerrit.cloudera.org:8080/15323 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: Ibdc650eb3ee30e8993330f2cbd389076ea2bad49 Gerrit-Change-Number: 15323 Gerrit-PatchSet: 2 Gerrit-Owner: Andrew Wong <[email protected]> Gerrit-Reviewer: Adar Dembo <[email protected]> Gerrit-Reviewer: Alexey Serbin <[email protected]> Gerrit-Reviewer: Andrew Wong <[email protected]> Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Reviewer: Tidy Bot (241) Gerrit-Comment-Date: Tue, 03 Mar 2020 00:27:54 +0000 Gerrit-HasComments: Yes
