You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@kudu.apache.org by "Todd Lipcon (Code Review)" <ge...@cloudera.org> on 2017/05/11 05:26:15 UTC
[kudu-CR] log: shut down appender thread when idle
Hello Adar Dembo,
I'd like you to do a code review. Please visit
http://gerrit.cloudera.org:8080/6856
to review the following change.
Change subject: log: shut down appender thread when idle
......................................................................
log: shut down appender thread when idle
This changes the log appender thread to be based on submitting tasks to a pool,
rather than a thread which is always running.
See the comments on Log::AppendThread in log.cc for notes on the design.
Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
---
M src/kudu/consensus/log-test.cc
M src/kudu/consensus/log.cc
M src/kudu/consensus/log.h
M src/kudu/util/blocking_queue-test.cc
M src/kudu/util/blocking_queue.h
M src/kudu/util/threadpool.h
6 files changed, 225 insertions(+), 84 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/56/6856/1
--
To view, visit http://gerrit.cloudera.org:8080/6856
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: newchange
Gerrit-Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Gerrit-PatchSet: 1
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
[kudu-CR] log: shut down appender thread when idle
Posted by "Todd Lipcon (Code Review)" <ge...@cloudera.org>.
Todd Lipcon has posted comments on this change.
Change subject: log: shut down appender thread when idle
......................................................................
Patch Set 1:
(1 comment)
http://gerrit.cloudera.org:8080/#/c/6856/1/src/kudu/consensus/log.cc
File src/kudu/consensus/log.cc:
Line 256: // Stopping is a bit tricky. We have to consider the following race:
> ok, lemme see if I can just add a variant of mt-log-test that you suggested
Done
--
To view, visit http://gerrit.cloudera.org:8080/6856
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Gerrit-PatchSet: 1
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-HasComments: Yes
[kudu-CR] log: shut down appender thread when idle
Posted by "Todd Lipcon (Code Review)" <ge...@cloudera.org>.
Todd Lipcon has posted comments on this change.
Change subject: log: shut down appender thread when idle
......................................................................
Patch Set 4: Verified+1
Looked into the flake but seems unrelated - somehow Subprocess::Start hung while starting a kudu-master. Perhaps a TSAN bug.
--
To view, visit http://gerrit.cloudera.org:8080/6856
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Gerrit-PatchSet: 4
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-HasComments: No
[kudu-CR] log: shut down appender thread when idle
Posted by "Adar Dembo (Code Review)" <ge...@cloudera.org>.
Adar Dembo has posted comments on this change.
Change subject: log: shut down appender thread when idle
......................................................................
Patch Set 2:
(1 comment)
http://gerrit.cloudera.org:8080/#/c/6856/1/src/kudu/consensus/log.cc
File src/kudu/consensus/log.cc:
Line 256: &worker_state_, WORKER_STOPPED, WORKER_ACTIVE);
> hard to repro in a test without injecting various sleeps. I did manually in
I looked at mt-log-test and I don't think it's geared to trigger this, because all of the writers are running all at once, and then not at all. That is, there's no staggering, so there's no window of time where the queue will be empty of entries for an entire second.
Maybe mt-log-test could reliably trigger this if you made kIdleThreshold configurable and turned it way down (say 1ms) for the test? That would probably be a good stress test in and of itself, since it'd trigger many active->stopped and stopped->active transitions.
Broadly speaking I don't find MAYBE... style fault injection to be polluting, but I don't want to ask to you spend hours building a test that, in the best of times, rarely triggers the race. So, if it's easily reproducible by tweaking the timeout (or injecting some latency), I'd support it, but not if it necessitates a new test with hundreds of lines of code.
--
To view, visit http://gerrit.cloudera.org:8080/6856
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Gerrit-PatchSet: 2
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-HasComments: Yes
[kudu-CR] log: shut down appender thread when idle
Posted by "Todd Lipcon (Code Review)" <ge...@cloudera.org>.
Hello Kudu Jenkins,
I'd like you to reexamine a change. Please visit
http://gerrit.cloudera.org:8080/6856
to look at the new patch set (#3).
Change subject: log: shut down appender thread when idle
......................................................................
log: shut down appender thread when idle
This changes the log appender thread to be based on submitting tasks to a pool,
rather than a thread which is always running.
See the comments on Log::AppendThread in log.cc for notes on the design.
This also adds a new test case in mt-log-test which triggers the various
complexities of the appender lifecycle. I manually checked coverage using
LOG(WARNING) statements to make sure it does exercise the various cases.
Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
---
M src/kudu/consensus/log-test-base.h
M src/kudu/consensus/log-test.cc
M src/kudu/consensus/log.cc
M src/kudu/consensus/log.h
M src/kudu/consensus/mt-log-test.cc
M src/kudu/util/blocking_queue-test.cc
M src/kudu/util/blocking_queue.h
M src/kudu/util/threadpool.h
8 files changed, 301 insertions(+), 102 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/56/6856/3
--
To view, visit http://gerrit.cloudera.org:8080/6856
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: newpatchset
Gerrit-Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Gerrit-PatchSet: 3
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
[kudu-CR] log: shut down appender thread when idle
Posted by "Todd Lipcon (Code Review)" <ge...@cloudera.org>.
Todd Lipcon has submitted this change and it was merged.
Change subject: log: shut down appender thread when idle
......................................................................
log: shut down appender thread when idle
This changes the log appender thread to be based on submitting tasks to a pool,
rather than a thread which is always running.
See the comments on Log::AppendThread in log.cc for notes on the design.
This also adds a new test case in mt-log-test which triggers the various
complexities of the appender lifecycle. I manually checked coverage using
LOG(WARNING) statements to make sure it does exercise the various cases.
Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Reviewed-on: http://gerrit.cloudera.org:8080/6856
Reviewed-by: Adar Dembo <ad...@cloudera.com>
Tested-by: Todd Lipcon <to...@apache.org>
---
M src/kudu/consensus/log-test-base.h
M src/kudu/consensus/log-test.cc
M src/kudu/consensus/log.cc
M src/kudu/consensus/log.h
M src/kudu/consensus/mt-log-test.cc
M src/kudu/util/blocking_queue-test.cc
M src/kudu/util/blocking_queue.h
M src/kudu/util/threadpool.h
8 files changed, 301 insertions(+), 102 deletions(-)
Approvals:
Adar Dembo: Looks good to me, approved
Todd Lipcon: Verified
--
To view, visit http://gerrit.cloudera.org:8080/6856
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: merged
Gerrit-Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Gerrit-PatchSet: 5
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
[kudu-CR] log: shut down appender thread when idle
Posted by "Adar Dembo (Code Review)" <ge...@cloudera.org>.
Adar Dembo has posted comments on this change.
Change subject: log: shut down appender thread when idle
......................................................................
Patch Set 3:
Code looks good but the TSAN failure looks related.
--
To view, visit http://gerrit.cloudera.org:8080/6856
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Gerrit-PatchSet: 3
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-HasComments: No
[kudu-CR] log: shut down appender thread when idle
Posted by "Todd Lipcon (Code Review)" <ge...@cloudera.org>.
Todd Lipcon has posted comments on this change.
Change subject: log: shut down appender thread when idle
......................................................................
Patch Set 1:
(1 comment)
http://gerrit.cloudera.org:8080/#/c/6856/1/src/kudu/consensus/log.cc
File src/kudu/consensus/log.cc:
Line 256: // Stopping is a bit tricky. We have to consider the following race:
> I looked at mt-log-test and I don't think it's geared to trigger this, beca
ok, lemme see if I can just add a variant of mt-log-test that you suggested, and see if it covers these paths with some "poor man's coverage"
--
To view, visit http://gerrit.cloudera.org:8080/6856
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Gerrit-PatchSet: 1
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-HasComments: Yes
[kudu-CR] log: shut down appender thread when idle
Posted by "Todd Lipcon (Code Review)" <ge...@cloudera.org>.
Todd Lipcon has posted comments on this change.
Change subject: log: shut down appender thread when idle
......................................................................
Patch Set 1:
(1 comment)
http://gerrit.cloudera.org:8080/#/c/6856/1/src/kudu/consensus/log.cc
File src/kudu/consensus/log.cc:
Line 280: return false;
> warning: redundant boolean literal in conditional return statement [readabi
meh, I like it like this, since there is good comment value
--
To view, visit http://gerrit.cloudera.org:8080/6856
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Gerrit-PatchSet: 1
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-HasComments: Yes
[kudu-CR] log: shut down appender thread when idle
Posted by "Adar Dembo (Code Review)" <ge...@cloudera.org>.
Adar Dembo has posted comments on this change.
Change subject: log: shut down appender thread when idle
......................................................................
Patch Set 1:
(10 comments)
http://gerrit.cloudera.org:8080/#/c/6856/1/src/kudu/consensus/log-test.cc
File src/kudu/consensus/log-test.cc:
Line 1105: ASSERT_TRUE(log_->append_thread_active_for_tests());
Could be flaky if enough time passes between Append() and now, right?
http://gerrit.cloudera.org:8080/#/c/6856/1/src/kudu/consensus/log.cc
File src/kudu/consensus/log.cc:
Line 203: // Tries to transition back to WORKER_STOPPED state. If successful, returns true.
Nit: Add a blank line before.
Line 229: gscoped_ptr<ThreadPool> append_pool_;
Given this design and given that the append thread is blocked on IO when running, I take it we don't want to consolidate pools across tablets?
Line 243: .set_idle_timeout(MonoDelta::FromSeconds(0))
Add a comment for this? I'm guessing it's because we handle idleness ourselves, so there's no reason for the threadpool to keep a thread alive if we've deemed it unnecessary.
Line 248: void Log::AppendThread::Wake() {
DCHECK(append_pool_) maybe?
Line 256: // Stopping is a bit tricky. We have to consider the following race:
Do we have any unit tests that can semi-reliably produce this race? A multi-threaded stress test for the log?
Line 288: void Log::AppendThread::DoWork() {
Maybe DCHECK that worker_state_ is WORKER_ACTIVE?
Line 365: append_pool_->Wait();
Why call Wait() prior to Shutdown()? IIRC, Shutdown() will wait for outstanding tasks and cancel all the pending tasks; isn't that what we want?
http://gerrit.cloudera.org:8080/#/c/6856/1/src/kudu/util/blocking_queue-test.cc
File src/kudu/util/blocking_queue-test.cc:
Line 64: ASSERT_OK(test_queue.BlockingDrainTo(&out, MonoTime::Now() + MonoDelta::FromSeconds(5)));
May be flaky if enough time passes between creation of the deadline and the call to BlockingDrainTo. Maybe make it obnoxiously higher, like 30s?
http://gerrit.cloudera.org:8080/#/c/6856/1/src/kudu/util/threadpool.h
File src/kudu/util/threadpool.h:
Line 89: // timeout: How long we'll keep around an idle thread before timing it out.
Change this to idle_timeout too?
--
To view, visit http://gerrit.cloudera.org:8080/6856
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Gerrit-PatchSet: 1
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-HasComments: Yes
[kudu-CR] log: shut down appender thread when idle
Posted by "Adar Dembo (Code Review)" <ge...@cloudera.org>.
Adar Dembo has posted comments on this change.
Change subject: log: shut down appender thread when idle
......................................................................
Patch Set 4: Code-Review+2
--
To view, visit http://gerrit.cloudera.org:8080/6856
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Gerrit-PatchSet: 4
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-HasComments: No
[kudu-CR] log: shut down appender thread when idle
Posted by "Todd Lipcon (Code Review)" <ge...@cloudera.org>.
Todd Lipcon has posted comments on this change.
Change subject: log: shut down appender thread when idle
......................................................................
Patch Set 1:
(11 comments)
http://gerrit.cloudera.org:8080/#/c/6856/1/src/kudu/consensus/log-test.cc
File src/kudu/consensus/log-test.cc:
Line 1105: ASSERT_TRUE(log_->append_thread_active_for_tests());
> Could be flaky if enough time passes between Append() and now, right?
yea, would have to be a whole second, but lemme see if I can make it less flaky-prone with a loop
http://gerrit.cloudera.org:8080/#/c/6856/1/src/kudu/consensus/log.cc
File src/kudu/consensus/log.cc:
Line 203: // Tries to transition back to WORKER_STOPPED state. If successful, returns true.
> Nit: Add a blank line before.
Done
Line 229: gscoped_ptr<ThreadPool> append_pool_;
> Given this design and given that the append thread is blocked on IO when ru
yea, I don't know if there's a lot of benefit in doing that... suppose it would be possible at some point but this seemed easier and we still get the end goal of no threads running for cold tablets
Line 243: .set_idle_timeout(MonoDelta::FromSeconds(0))
> Add a comment for this? I'm guessing it's because we handle idleness oursel
Done
Line 248: void Log::AppendThread::Wake() {
> DCHECK(append_pool_) maybe?
Done
Line 256: // Stopping is a bit tricky. We have to consider the following race:
> Do we have any unit tests that can semi-reliably produce this race? A multi
hard to repro in a test without injecting various sleeps. I did manually inject sleeps in the right places to cover all the code paths, and we do have some multi-threaded tests like mt-log-test as well as any multi-threaded integration tests covering this.
Do you think it's worth polluting the code with MAYBE_INJECT_RANDOM_LATENCY() calls to trigger potential races?
Line 288: void Log::AppendThread::DoWork() {
> Maybe DCHECK that worker_state_ is WORKER_ACTIVE?
Done
Line 365: append_pool_->Wait();
> Why call Wait() prior to Shutdown()? IIRC, Shutdown() will wait for outstan
I don't think we want to cancel the pending ones, because then I think we'll end up orphaning entries in the blocking queue. We want the task to drain the queue and run the callbacks, I believe. (The previous behavior by joining is doing the same thing, allowing the appender thread to drain the queue)
http://gerrit.cloudera.org:8080/#/c/6856/1/src/kudu/util/blocking_queue-test.cc
File src/kudu/util/blocking_queue-test.cc:
Line 31: using std::shared_ptr;
> warning: using decl 'shared_ptr' is unused [misc-unused-using-decls]
Done
Line 64: ASSERT_OK(test_queue.BlockingDrainTo(&out, MonoTime::Now() + MonoDelta::FromSeconds(5)));
> May be flaky if enough time passes between creation of the deadline and the
Done
http://gerrit.cloudera.org:8080/#/c/6856/1/src/kudu/util/threadpool.h
File src/kudu/util/threadpool.h:
Line 89: // timeout: How long we'll keep around an idle thread before timing it out.
> Change this to idle_timeout too?
Done
--
To view, visit http://gerrit.cloudera.org:8080/6856
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Gerrit-PatchSet: 1
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-HasComments: Yes
[kudu-CR] log: shut down appender thread when idle
Posted by "Todd Lipcon (Code Review)" <ge...@cloudera.org>.
Hello Adar Dembo, Kudu Jenkins,
I'd like you to reexamine a change. Please visit
http://gerrit.cloudera.org:8080/6856
to look at the new patch set (#4).
Change subject: log: shut down appender thread when idle
......................................................................
log: shut down appender thread when idle
This changes the log appender thread to be based on submitting tasks to a pool,
rather than a thread which is always running.
See the comments on Log::AppendThread in log.cc for notes on the design.
This also adds a new test case in mt-log-test which triggers the various
complexities of the appender lifecycle. I manually checked coverage using
LOG(WARNING) statements to make sure it does exercise the various cases.
Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
---
M src/kudu/consensus/log-test-base.h
M src/kudu/consensus/log-test.cc
M src/kudu/consensus/log.cc
M src/kudu/consensus/log.h
M src/kudu/consensus/mt-log-test.cc
M src/kudu/util/blocking_queue-test.cc
M src/kudu/util/blocking_queue.h
M src/kudu/util/threadpool.h
8 files changed, 301 insertions(+), 102 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/56/6856/4
--
To view, visit http://gerrit.cloudera.org:8080/6856
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: newpatchset
Gerrit-Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Gerrit-PatchSet: 4
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
[kudu-CR] log: shut down appender thread when idle
Posted by "Todd Lipcon (Code Review)" <ge...@cloudera.org>.
Hello Adar Dembo, Kudu Jenkins,
I'd like you to reexamine a change. Please visit
http://gerrit.cloudera.org:8080/6856
to look at the new patch set (#2).
Change subject: log: shut down appender thread when idle
......................................................................
log: shut down appender thread when idle
This changes the log appender thread to be based on submitting tasks to a pool,
rather than a thread which is always running.
See the comments on Log::AppendThread in log.cc for notes on the design.
Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
---
M src/kudu/consensus/log-test.cc
M src/kudu/consensus/log.cc
M src/kudu/consensus/log.h
M src/kudu/util/blocking_queue-test.cc
M src/kudu/util/blocking_queue.h
M src/kudu/util/threadpool.h
6 files changed, 243 insertions(+), 86 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/56/6856/2
--
To view, visit http://gerrit.cloudera.org:8080/6856
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: newpatchset
Gerrit-Change-Id: Id40a4cfcec96198b537c2f50be7ff1204faf96d2
Gerrit-PatchSet: 2
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Adar Dembo <ad...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>