Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/16814 )
Change subject: [tserver] KUDU-2612: participant op RPC endpoint ...................................................................... Patch Set 1: (5 comments) http://gerrit.cloudera.org:8080/#/c/16814/1//COMMIT_MSG Commit Message: http://gerrit.cloudera.org:8080/#/c/16814/1//COMMIT_MSG@9 PS1, Line 9: This adds an RPC endpoint to the tablet servers that allows proxies to : interact with transaction participants. Could you add more colors on this, adding an example of a scenario when and how this is going to be used. Adding a reference to a particular piece (or step) in the txn design doc might be a good thing to add, I guess. http://gerrit.cloudera.org:8080/#/c/16814/1/src/kudu/integration-tests/txn_participant-itest.cc File src/kudu/integration-tests/txn_participant-itest.cc: http://gerrit.cloudera.org:8080/#/c/16814/1/src/kudu/integration-tests/txn_participant-itest.cc@521 PS1, Line 521: not RUNNING BTW, what is the expected behavior what a tablet is bootstrapping? Is it the same as in case of tombstoned tablet or it should be something else? http://gerrit.cloudera.org:8080/#/c/16814/1/src/kudu/integration-tests/txn_participant-itest.cc@529 PS1, Line 529: dummy-tablet-id BTW, would we allow status tablet to participate in a transaction? Not sure it would contradict to anything, but from the other side I'm not sure we should allow that. Maybe you have already thought about that and know the answer :) http://gerrit.cloudera.org:8080/#/c/16814/1/src/kudu/tserver/tserver_admin.proto File src/kudu/tserver/tserver_admin.proto: http://gerrit.cloudera.org:8080/#/c/16814/1/src/kudu/tserver/tserver_admin.proto@113 PS1, Line 113: message ParticipantRequestPB { : optional bytes tablet_id = 1; : optional ParticipantOpPB op = 2; : } Just want to double-check it's safe to swap the fields. I guess it's so because this message hasn't been sent over wire yet, even if the changelist that introduced ParticipantRequestPB is already in released bits: $ git tag -l --contains 14d4117afa 1.13.0 1.13.0-RC1 1.13.0-RC2 Is it correct? http://gerrit.cloudera.org:8080/#/c/16814/1/src/kudu/tserver/tserver_admin.proto@256 PS1, Line 256: Have a tablet participate in a transaction It would nice if the comment were a bit more clearer w.r.t. what is the semantics of this method from the callee side, i.e. what 'have' here means? -- To view, visit http://gerrit.cloudera.org:8080/16814 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: Ic48895438ce67e453d235934ac560efe8415921b Gerrit-Change-Number: 16814 Gerrit-PatchSet: 1 Gerrit-Owner: Andrew Wong <aw...@cloudera.com> Gerrit-Reviewer: Alexey Serbin <aser...@cloudera.com> Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com> Gerrit-Reviewer: Hao Hao <hao....@cloudera.com> Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Comment-Date: Mon, 07 Dec 2020 06:47:40 +0000 Gerrit-HasComments: Yes
