You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@kudu.apache.org by "Andrew Wong (Code Review)" <ge...@cloudera.org> on 2020/01/04 04:39:00 UTC
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Andrew Wong has uploaded this change for review. ( http://gerrit.cloudera.org:8080/14977
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
KUDU-3011 p2: mechanism to quiesce leadership
This plumbs a functor from the TabletServer down to the RaftConsensus
instance of each TabletReplica that indicates whether the server is in a
quiescing state.
When in this state, any election started by a RaftConsensus instance on
the server will fail with an error.
Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
---
M src/kudu/consensus/raft_consensus.cc
M src/kudu/consensus/raft_consensus.h
M src/kudu/consensus/raft_consensus_quorum-test.cc
M src/kudu/integration-tests/CMakeLists.txt
A src/kudu/integration-tests/tablet_server_quiescing-itest.cc
M src/kudu/master/sys_catalog.cc
M src/kudu/tablet/tablet_replica-test.cc
M src/kudu/tserver/tablet_copy_source_session-test.cc
M src/kudu/tserver/tablet_server.cc
M src/kudu/tserver/tablet_server.h
M src/kudu/tserver/ts_tablet_manager.cc
M src/kudu/tserver/ts_tablet_manager.h
12 files changed, 206 insertions(+), 11 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/77/14977/1
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: newchange
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 1
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Adar Dembo (Code Review)" <ge...@cloudera.org>.
Adar Dembo has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 4: Code-Review+2
(1 comment)
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc
File src/kudu/consensus/raft_consensus.cc:
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc@382
PS4, Line 382: WARN_NOT_OK(StartElection(NORMAL_ELECTION, INITIAL_SINGLE_NODE_ELECTION),
: "Couldn't start leader election");
So for singleton replicas, when should we expect this to fail? What's the outcome? A forever follower replica?
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 4
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Mon, 06 Jan 2020 21:10:17 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Alexey Serbin (Code Review)" <ge...@cloudera.org>.
Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 10: Code-Review+2
(2 comments)
http://gerrit.cloudera.org:8080/#/c/14977/7//COMMIT_MSG
Commit Message:
PS7:
> Yeah, I would rather address that in a later patch, if you don't mind -- I
SGTM.
http://gerrit.cloudera.org:8080/#/c/14977/7//COMMIT_MSG@13
PS7, Line 13: election started
> It's a good question, and I actually originally started going down that rou
SGTM. Probably, the better optimization is to avoid tablet copying at the master side when target tablet server is quiescent.
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 10
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Tue, 07 Jan 2020 07:09:17 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Andrew Wong has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 3:
(2 comments)
http://gerrit.cloudera.org:8080/#/c/14977/2/src/kudu/consensus/raft_consensus.cc
File src/kudu/consensus/raft_consensus.cc:
http://gerrit.cloudera.org:8080/#/c/14977/2/src/kudu/consensus/raft_consensus.cc@444
PS2, Line 444: if (server_ctx_.quiescing && server_ctx_.quiescing->load()) {
> Will it be possible to quiesce a tserver as it is bootstrapping replicas? I
You're right for the singe replica case, but it seems like failure to start an election is usually inconsequential via warning logs. I'll follow suit in the single replica case.
http://gerrit.cloudera.org:8080/#/c/14977/2/src/kudu/tserver/tablet_server.h
File src/kudu/tserver/tablet_server.h:
http://gerrit.cloudera.org:8080/#/c/14977/2/src/kudu/tserver/tablet_server.h@107
PS2, Line 107: // The options passed at construction time.
: const TabletServerOptions opts_;
:
> Make it an std::atomic and forgo the lock?
Done
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 3
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Mon, 06 Jan 2020 07:50:41 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Andrew Wong has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 7:
(3 comments)
The TSAN failure of TestNumLeadersMetric wasn't related to this patch; I'll get to the bottom of it.
http://gerrit.cloudera.org:8080/#/c/14977/7//COMMIT_MSG
Commit Message:
http://gerrit.cloudera.org:8080/#/c/14977/7//COMMIT_MSG@13
PS7, Line 13: election started
> What about tablet copying? Should the system behave differently in case if
It's a good question, and I actually originally started going down that route. It'd probably be a good optimization, but I don't think it is a requirement. Tablet copies are "crash safe" in that shutting down while one is happening will be fine -- the copy will just need to be restarted.
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc
File src/kudu/consensus/raft_consensus.cc:
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc@382
PS4, Line 382: WARN_NOT_OK(StartElection(NORMAL_ELECTION, INITIAL_SINGLE_NODE_ELECTION),
: "Couldn't start leader election");
> Whoops, it seems I started some comment and left it incomplete. I was abou
>Thinking ahead to KUDU-2975, are there any WAL write subcalls that we should be concerned with?
Yes, but not here.
>Whoops, it seems I started some comment and left it incomplete. I was about to ask whether the single-replica scenario case should be covered by a test, i.e. making sure that a single replica tablet starts an election and elects its single replica a leader once quiescent period is over.
Yep, TServerQuiescingParamITest.TestNoElectionsForNewReplicas is parameterized by the number of replicas (1 and 3).
http://gerrit.cloudera.org:8080/#/c/14977/7/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/14977/7/src/kudu/integration-tests/tablet_server_quiescing-itest.cc@64
PS7, Line 64: hitting
> drop ?
Done
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 7
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Tue, 07 Jan 2020 00:03:07 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Adar Dembo (Code Review)" <ge...@cloudera.org>.
Adar Dembo has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 7: Code-Review+2
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 7
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Mon, 06 Jan 2020 22:13:44 +0000
Gerrit-HasComments: No
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Adar Dembo (Code Review)" <ge...@cloudera.org>.
Adar Dembo has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 7:
(1 comment)
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc
File src/kudu/consensus/raft_consensus.cc:
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc@382
PS4, Line 382: WARN_NOT_OK(StartElection(NORMAL_ELECTION, INITIAL_SINGLE_NODE_ELECTION),
: "Couldn't start leader election");
> Failing to write cmeta or failing to write a WAL entry would lead to a cras
Thinking ahead to KUDU-2975, are there any WAL write subcalls that we should be concerned with?
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 7
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Mon, 06 Jan 2020 22:13:07 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Adar Dembo (Code Review)" <ge...@cloudera.org>.
Adar Dembo has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 6: Code-Review+2
(1 comment)
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc
File src/kudu/consensus/raft_consensus.cc:
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc@382
PS4, Line 382: WARN_NOT_OK(StartElection(NORMAL_ELECTION, INITIAL_SINGLE_NODE_ELECTION),
: "Couldn't start leader election");
> If the replica is being shut down, StartElection() can fail. There also see
I was worried about IO-related failures e.g. updating cmeta or writing a WAL entry. Would they all lead to a failed replica?
Basically just want to make sure we never wind up in a state where the replica is a follower, isn't failed, and nothing happens after that.
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 6
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Mon, 06 Jan 2020 21:50:24 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Alexey Serbin (Code Review)" <ge...@cloudera.org>.
Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 7:
(2 comments)
http://gerrit.cloudera.org:8080/#/c/14977/7//COMMIT_MSG
Commit Message:
http://gerrit.cloudera.org:8080/#/c/14977/7//COMMIT_MSG@13
PS7, Line 13: election started
What about tablet copying? Should the system behave differently in case if a tablet servers is in the quiescent state?
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc
File src/kudu/consensus/raft_consensus.cc:
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc@382
PS4, Line 382: WARN_NOT_OK(StartElection(NORMAL_ELECTION, INITIAL_SINGLE_NODE_ELECTION),
: "Couldn't start leader election");
> Thinking ahead to KUDU-2975, are there any WAL write subcalls that we shoul
Should
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 7
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Mon, 06 Jan 2020 23:07:01 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Hello Tidy Bot, Kudu Jenkins, Adar Dembo,
I'd like you to reexamine a change. Please visit
http://gerrit.cloudera.org:8080/14977
to look at the new patch set (#5).
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
KUDU-3011 p2: mechanism to quiesce leadership
This plumbs a functor from the TabletServer down to the RaftConsensus
instance of each TabletReplica that indicates whether the server is in a
quiescing state.
When in this state, any election started by a RaftConsensus instance on
the server will fail with an error.
Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
---
M src/kudu/consensus/raft_consensus.cc
M src/kudu/consensus/raft_consensus.h
M src/kudu/consensus/raft_consensus_quorum-test.cc
M src/kudu/integration-tests/CMakeLists.txt
A src/kudu/integration-tests/tablet_server_quiescing-itest.cc
M src/kudu/master/sys_catalog.cc
M src/kudu/tablet/tablet_replica-test.cc
M src/kudu/tserver/tablet_copy_source_session-test.cc
M src/kudu/tserver/tablet_server.cc
M src/kudu/tserver/tablet_server.h
M src/kudu/tserver/ts_tablet_manager.cc
11 files changed, 247 insertions(+), 12 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/77/14977/5
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: newpatchset
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 5
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Andrew Wong has removed a vote on this change.
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Removed Verified-1 by Kudu Jenkins (120)
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: deleteVote
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 10
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Andrew Wong has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 10: Verified+1
(1 comment)
Test failures were unrelated.
http://gerrit.cloudera.org:8080/#/c/14977/7//COMMIT_MSG
Commit Message:
http://gerrit.cloudera.org:8080/#/c/14977/7//COMMIT_MSG@13
PS7, Line 13: election started
> SGTM. Probably, the better optimization is to avoid tablet copying at the
Right, I intend on leveraging maintenance mode for that.
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 10
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Tue, 07 Jan 2020 07:37:41 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Andrew Wong has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 6:
(1 comment)
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc
File src/kudu/consensus/raft_consensus.cc:
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc@382
PS4, Line 382: WARN_NOT_OK(StartElection(NORMAL_ELECTION, INITIAL_SINGLE_NODE_ELECTION),
: "Couldn't start leader election");
> I was worried about IO-related failures e.g. updating cmeta or writing a WA
Failing to write cmeta or failing to write a WAL entry would lead to a crash, since failures of WALs and metadata aren't tolerated right now. One such pathway is in StartElection() we call SetVotedForCurrentTermUnlocked(), which does IO to write the cmeta -- that's wrapped in a CHECK (as are many WAL and metadata operations, probably from a time before it was considered better practice to LOG(FATAL) instead).
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 6
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Mon, 06 Jan 2020 21:59:35 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Alexey Serbin (Code Review)" <ge...@cloudera.org>.
Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 8:
(3 comments)
http://gerrit.cloudera.org:8080/#/c/14977/8/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/14977/8/src/kudu/integration-tests/tablet_server_quiescing-itest.cc@72
PS8, Line 72: const string kTableName = "foo";
Is it needed?
http://gerrit.cloudera.org:8080/#/c/14977/8/src/kudu/integration-tests/tablet_server_quiescing-itest.cc@129
PS8, Line 129: const string kTableName = "foo";
ditto
http://gerrit.cloudera.org:8080/#/c/14977/8/src/kudu/integration-tests/tablet_server_quiescing-itest.cc@178
PS8, Line 178: const string kTableName = "foo";
ditto
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 8
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Tue, 07 Jan 2020 00:31:36 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Andrew Wong has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 4:
(1 comment)
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc
File src/kudu/consensus/raft_consensus.cc:
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc@382
PS4, Line 382: WARN_NOT_OK(StartElection(NORMAL_ELECTION, INITIAL_SINGLE_NODE_ELECTION),
: "Couldn't start leader election");
> So for singleton replicas, when should we expect this to fail? What's the o
If the replica is being shut down, StartElection() can fail. There also seem to be some invariant-related error statuses that can surface here. The main "expected" error is that for quiescing, which is tested in TServerQuiescingParamITest.TestNoElectionsForNewReplicas: it'll be a follower until we stop quiescing the server and we can start an election. I.e. when we stop quiescing the server, the failure detector will pop and we'll see elections happen.
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 4
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Mon, 06 Jan 2020 21:26:46 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Alexey Serbin (Code Review)" <ge...@cloudera.org>.
Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 8:
(2 comments)
http://gerrit.cloudera.org:8080/#/c/14977/7//COMMIT_MSG
Commit Message:
PS7:
With this approach when StartElection() returns an error, would it be a flood of warning messages in the tablet server's logs when the tablet server enters a quiescent period?
http://gerrit.cloudera.org:8080/#/c/14977/7/src/kudu/integration-tests/tablet_server_quiescing-itest.cc
File src/kudu/integration-tests/tablet_server_quiescing-itest.cc:
PS7:
Does it make sense to add a test to verify that even with a majority of replicas in quiescent state, a tablet still can elect a leader if the election is started by a replica that is in normal state?
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 8
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Tue, 07 Jan 2020 00:04:40 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Hello Tidy Bot, Alexey Serbin, Kudu Jenkins, Adar Dembo,
I'd like you to reexamine a change. Please visit
http://gerrit.cloudera.org:8080/14977
to look at the new patch set (#9).
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
KUDU-3011 p2: mechanism to quiesce leadership
This plumbs a shared flag from the TabletServer down to the
RaftConsensus instance of each TabletReplica that indicates whether the
server is in a quiescing state.
When in this state, any election started by a RaftConsensus instance on
the server will fail with an error.
Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
---
M src/kudu/consensus/raft_consensus.cc
M src/kudu/consensus/raft_consensus.h
M src/kudu/consensus/raft_consensus_quorum-test.cc
M src/kudu/integration-tests/CMakeLists.txt
A src/kudu/integration-tests/tablet_server_quiescing-itest.cc
M src/kudu/master/sys_catalog.cc
M src/kudu/tablet/tablet_replica-test.cc
M src/kudu/tserver/tablet_copy_source_session-test.cc
M src/kudu/tserver/tablet_server.cc
M src/kudu/tserver/tablet_server.h
M src/kudu/tserver/ts_tablet_manager.cc
11 files changed, 263 insertions(+), 12 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/77/14977/9
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: newpatchset
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 9
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Andrew Wong has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 9:
(5 comments)
http://gerrit.cloudera.org:8080/#/c/14977/7//COMMIT_MSG
Commit Message:
PS7:
> With this approach when StartElection() returns an error, would it be a flo
Yeah, I would rather address that in a later patch, if you don't mind -- I think throttling would be non-negligible code and would distract from this patch.
http://gerrit.cloudera.org:8080/#/c/14977/7/src/kudu/integration-tests/tablet_server_quiescing-itest.cc
File src/kudu/integration-tests/tablet_server_quiescing-itest.cc:
PS7:
> Does it make sense to add a test to verify that even with a majority of rep
Done
http://gerrit.cloudera.org:8080/#/c/14977/8/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/14977/8/src/kudu/integration-tests/tablet_server_quiescing-itest.cc@72
PS8, Line 72: if (tablet_ids) {
> Is it needed?
Done
http://gerrit.cloudera.org:8080/#/c/14977/8/src/kudu/integration-tests/tablet_server_quiescing-itest.cc@129
PS8, Line 129: vector<string> tablet_ids;
> ditto
Done
http://gerrit.cloudera.org:8080/#/c/14977/8/src/kudu/integration-tests/tablet_server_quiescing-itest.cc@178
PS8, Line 178: *ts->server()->mutable_quiescing
> ditto
Done
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 9
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Tue, 07 Jan 2020 04:09:04 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Hello Tidy Bot, Kudu Jenkins,
I'd like you to reexamine a change. Please visit
http://gerrit.cloudera.org:8080/14977
to look at the new patch set (#2).
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
KUDU-3011 p2: mechanism to quiesce leadership
This plumbs a functor from the TabletServer down to the RaftConsensus
instance of each TabletReplica that indicates whether the server is in a
quiescing state.
When in this state, any election started by a RaftConsensus instance on
the server will fail with an error.
Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
---
M src/kudu/consensus/raft_consensus.cc
M src/kudu/consensus/raft_consensus.h
M src/kudu/consensus/raft_consensus_quorum-test.cc
M src/kudu/integration-tests/CMakeLists.txt
A src/kudu/integration-tests/tablet_server_quiescing-itest.cc
M src/kudu/master/sys_catalog.cc
M src/kudu/tablet/tablet_replica-test.cc
M src/kudu/tserver/tablet_copy_source_session-test.cc
M src/kudu/tserver/tablet_server.cc
M src/kudu/tserver/tablet_server.h
M src/kudu/tserver/ts_tablet_manager.cc
M src/kudu/tserver/ts_tablet_manager.h
12 files changed, 205 insertions(+), 11 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/77/14977/2
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: newpatchset
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 2
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Hello Tidy Bot, Alexey Serbin, Kudu Jenkins, Adar Dembo,
I'd like you to reexamine a change. Please visit
http://gerrit.cloudera.org:8080/14977
to look at the new patch set (#8).
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
KUDU-3011 p2: mechanism to quiesce leadership
This plumbs a shared flag from the TabletServer down to the
RaftConsensus instance of each TabletReplica that indicates whether the
server is in a quiescing state.
When in this state, any election started by a RaftConsensus instance on
the server will fail with an error.
Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
---
M src/kudu/consensus/raft_consensus.cc
M src/kudu/consensus/raft_consensus.h
M src/kudu/consensus/raft_consensus_quorum-test.cc
M src/kudu/integration-tests/CMakeLists.txt
A src/kudu/integration-tests/tablet_server_quiescing-itest.cc
M src/kudu/master/sys_catalog.cc
M src/kudu/tablet/tablet_replica-test.cc
M src/kudu/tserver/tablet_copy_source_session-test.cc
M src/kudu/tserver/tablet_server.cc
M src/kudu/tserver/tablet_server.h
M src/kudu/tserver/ts_tablet_manager.cc
11 files changed, 247 insertions(+), 12 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/77/14977/8
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: newpatchset
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 8
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Alexey Serbin (Code Review)" <ge...@cloudera.org>.
Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 4:
(2 comments)
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc
File src/kudu/consensus/raft_consensus.cc:
http://gerrit.cloudera.org:8080/#/c/14977/4/src/kudu/consensus/raft_consensus.cc@382
PS4, Line 382: WARN_NOT_OK(StartElection(NORMAL_ELECTION, INITIAL_SINGLE_NODE_ELECTION),
: "Couldn't start leader election");
> Should
Whoops, it seems I started some comment and left it incomplete. I was about to ask whether the single-replica scenario case should be covered by a test, i.e. making sure that a single replica tablet starts an election and elects its single replica a leader once quiescent period is over.
http://gerrit.cloudera.org:8080/#/c/14977/7/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/14977/7/src/kudu/integration-tests/tablet_server_quiescing-itest.cc@64
PS7, Line 64: hitting
drop ?
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 4
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Mon, 06 Jan 2020 23:28:42 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Hello Tidy Bot, Kudu Jenkins, Adar Dembo,
I'd like you to reexamine a change. Please visit
http://gerrit.cloudera.org:8080/14977
to look at the new patch set (#6).
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
KUDU-3011 p2: mechanism to quiesce leadership
This plumbs a shared flag from the TabletServer down to the
RaftConsensus instance of each TabletReplica that indicates whether the
server is in a quiescing state.
When in this state, any election started by a RaftConsensus instance on
the server will fail with an error.
Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
---
M src/kudu/consensus/raft_consensus.cc
M src/kudu/consensus/raft_consensus.h
M src/kudu/consensus/raft_consensus_quorum-test.cc
M src/kudu/integration-tests/CMakeLists.txt
A src/kudu/integration-tests/tablet_server_quiescing-itest.cc
M src/kudu/master/sys_catalog.cc
M src/kudu/tablet/tablet_replica-test.cc
M src/kudu/tserver/tablet_copy_source_session-test.cc
M src/kudu/tserver/tablet_server.cc
M src/kudu/tserver/tablet_server.h
M src/kudu/tserver/ts_tablet_manager.cc
11 files changed, 247 insertions(+), 12 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/77/14977/6
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: newpatchset
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 6
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Adar Dembo (Code Review)" <ge...@cloudera.org>.
Adar Dembo has posted comments on this change. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
Patch Set 2:
(2 comments)
http://gerrit.cloudera.org:8080/#/c/14977/2/src/kudu/consensus/raft_consensus.cc
File src/kudu/consensus/raft_consensus.cc:
http://gerrit.cloudera.org:8080/#/c/14977/2/src/kudu/consensus/raft_consensus.cc@444
PS2, Line 444: return Status::IllegalState("leader elections are disabled");
Will it be possible to quiesce a tserver as it is bootstrapping replicas? If so, it looks like this failure will percolate up into outright replica failure, which seems undesirable.
Though I guess this also applies to new replica creation: a failure here will cause the newly created replica to be marked as failed (see RaftConsensus::Start).
http://gerrit.cloudera.org:8080/#/c/14977/2/src/kudu/tserver/tablet_server.h
File src/kudu/tserver/tablet_server.h:
http://gerrit.cloudera.org:8080/#/c/14977/2/src/kudu/tserver/tablet_server.h@107
PS2, Line 107: mutable simple_spinlock quiescing_lock_;
: bool quiescing_;
:
Make it an std::atomic and forgo the lock?
Then you could also plumb the member directly into the replicas rather than indirecting thru a functor, right?
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 2
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Mon, 06 Jan 2020 05:36:16 +0000
Gerrit-HasComments: Yes
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Andrew Wong has submitted this change and it was merged. ( http://gerrit.cloudera.org:8080/14977 )
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
KUDU-3011 p2: mechanism to quiesce leadership
This plumbs a shared flag from the TabletServer down to the
RaftConsensus instance of each TabletReplica that indicates whether the
server is in a quiescing state.
When in this state, any election started by a RaftConsensus instance on
the server will fail with an error.
Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Reviewed-on: http://gerrit.cloudera.org:8080/14977
Reviewed-by: Alexey Serbin <as...@cloudera.com>
Tested-by: Andrew Wong <aw...@cloudera.com>
---
M src/kudu/consensus/raft_consensus.cc
M src/kudu/consensus/raft_consensus.h
M src/kudu/consensus/raft_consensus_quorum-test.cc
M src/kudu/integration-tests/CMakeLists.txt
A src/kudu/integration-tests/tablet_server_quiescing-itest.cc
M src/kudu/master/sys_catalog.cc
M src/kudu/tablet/tablet_replica-test.cc
M src/kudu/tserver/tablet_copy_source_session-test.cc
M src/kudu/tserver/tablet_server.cc
M src/kudu/tserver/tablet_server.h
M src/kudu/tserver/ts_tablet_manager.cc
11 files changed, 264 insertions(+), 12 deletions(-)
Approvals:
Alexey Serbin: Looks good to me, approved
Andrew Wong: Verified
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: merged
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 11
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Hello Tidy Bot, Kudu Jenkins, Adar Dembo,
I'd like you to reexamine a change. Please visit
http://gerrit.cloudera.org:8080/14977
to look at the new patch set (#4).
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
KUDU-3011 p2: mechanism to quiesce leadership
This plumbs a functor from the TabletServer down to the RaftConsensus
instance of each TabletReplica that indicates whether the server is in a
quiescing state.
When in this state, any election started by a RaftConsensus instance on
the server will fail with an error.
Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
---
M src/kudu/consensus/raft_consensus.cc
M src/kudu/consensus/raft_consensus.h
M src/kudu/consensus/raft_consensus_quorum-test.cc
M src/kudu/integration-tests/CMakeLists.txt
A src/kudu/integration-tests/tablet_server_quiescing-itest.cc
M src/kudu/master/sys_catalog.cc
M src/kudu/tablet/tablet_replica-test.cc
M src/kudu/tserver/tablet_copy_source_session-test.cc
M src/kudu/tserver/tablet_server.cc
M src/kudu/tserver/tablet_server.h
M src/kudu/tserver/ts_tablet_manager.cc
11 files changed, 247 insertions(+), 12 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/77/14977/4
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: newpatchset
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 4
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Hello Tidy Bot, Alexey Serbin, Kudu Jenkins, Adar Dembo,
I'd like you to reexamine a change. Please visit
http://gerrit.cloudera.org:8080/14977
to look at the new patch set (#10).
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
KUDU-3011 p2: mechanism to quiesce leadership
This plumbs a shared flag from the TabletServer down to the
RaftConsensus instance of each TabletReplica that indicates whether the
server is in a quiescing state.
When in this state, any election started by a RaftConsensus instance on
the server will fail with an error.
Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
---
M src/kudu/consensus/raft_consensus.cc
M src/kudu/consensus/raft_consensus.h
M src/kudu/consensus/raft_consensus_quorum-test.cc
M src/kudu/integration-tests/CMakeLists.txt
A src/kudu/integration-tests/tablet_server_quiescing-itest.cc
M src/kudu/master/sys_catalog.cc
M src/kudu/tablet/tablet_replica-test.cc
M src/kudu/tserver/tablet_copy_source_session-test.cc
M src/kudu/tserver/tablet_server.cc
M src/kudu/tserver/tablet_server.h
M src/kudu/tserver/ts_tablet_manager.cc
11 files changed, 264 insertions(+), 12 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/77/14977/10
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: newpatchset
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 10
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
[kudu-CR] KUDU-3011 p2: mechanism to quiesce leadership
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Hello Tidy Bot, Kudu Jenkins, Adar Dembo,
I'd like you to reexamine a change. Please visit
http://gerrit.cloudera.org:8080/14977
to look at the new patch set (#3).
Change subject: KUDU-3011 p2: mechanism to quiesce leadership
......................................................................
KUDU-3011 p2: mechanism to quiesce leadership
This plumbs a functor from the TabletServer down to the RaftConsensus
instance of each TabletReplica that indicates whether the server is in a
quiescing state.
When in this state, any election started by a RaftConsensus instance on
the server will fail with an error.
Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
---
M src/kudu/consensus/raft_consensus.cc
M src/kudu/consensus/raft_consensus.h
M src/kudu/consensus/raft_consensus_quorum-test.cc
M src/kudu/integration-tests/CMakeLists.txt
A src/kudu/integration-tests/tablet_server_quiescing-itest.cc
M src/kudu/master/sys_catalog.cc
M src/kudu/tablet/tablet_replica-test.cc
M src/kudu/tserver/tablet_copy_source_session-test.cc
M src/kudu/tserver/tablet_server.cc
M src/kudu/tserver/tablet_server.h
M src/kudu/tserver/ts_tablet_manager.cc
11 files changed, 251 insertions(+), 12 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/77/14977/3
--
To view, visit http://gerrit.cloudera.org:8080/14977
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: newpatchset
Gerrit-Change-Id: I0eb3778fe094647bc5f61f97971087d7efb8af5e
Gerrit-Change-Number: 14977
Gerrit-PatchSet: 3
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)