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 2016/08/30 21:34:26 UTC

[kudu-CR] Add trace metrics for maintenance ops, LBM writes

Hello Dan Burkert,

I'd like you to do a code review.  Please visit

    http://gerrit.cloudera.org:8080/4172

to review the following change.

Change subject: Add trace metrics for maintenance ops, LBM writes
......................................................................

Add trace metrics for maintenance ops, LBM writes

This adds a trace-scoped counter for LogBlockManager writes, and also
adds traces and counter printing to the maintenance manager.

With this patch, flushes and compactions now output log messages like:

I0830 13:42:28.809852  2208 maintenance_manager.cc:356] Time spent running FlushMRSOp(4301e24175a2469fba37184430567acc): real 9.510s user 4.432s     sys 0.860s
I0830 13:42:28.810760  2208 maintenance_manager.cc:362] FlushMRSOp(4301e24175a2469fba37184430567acc) metrics: {"cfile_init":14,"compiler_manager_pool.queue_time_us":106,"compiler_manager_pool.run_cpu_time_us":219218,"compiler_manager_pool.run_wall_time_us":232004,"fdatasync":533,"fdatasync_us":67137,"lbm root 0.queue_time_us":1605,"lbm root 0.run_cpu_time_us":1280,"lbm root 0.run_wall_time_us":344957,"lbm_read_time_us":94,"lbm_reads_lt_1ms":56,"lbm_write_time_us":709833,"lbm_writes_gt_100_ms":1,"lbm_writes_lt_1ms":11564,"mutex_wait_us":343945,"tcmalloc_contention_cycles":2759552,"thread_start_us":1304,"threads_started":13}

This can be handy for understanding why a particular compaction or flush
was slow.

Change-Id: I39cd9438071aabbd2d7bdeca3269af8b83f2d55b
---
M src/kudu/fs/log_block_manager.cc
M src/kudu/util/maintenance_manager.cc
M src/kudu/util/trace.h
3 files changed, 30 insertions(+), 10 deletions(-)


  git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/72/4172/1
-- 
To view, visit http://gerrit.cloudera.org:8080/4172
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: newchange
Gerrit-Change-Id: I39cd9438071aabbd2d7bdeca3269af8b83f2d55b
Gerrit-PatchSet: 1
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Dan Burkert <da...@cloudera.com>

[kudu-CR] Add trace metrics for maintenance ops, LBM writes

Posted by "Dan Burkert (Code Review)" <ge...@cloudera.org>.
Dan Burkert has posted comments on this change.

Change subject: Add trace metrics for maintenance ops, LBM writes
......................................................................


Patch Set 1: Code-Review+2

(1 comment)

http://gerrit.cloudera.org:8080/#/c/4172/1/src/kudu/util/maintenance_manager.cc
File src/kudu/util/maintenance_manager.cc:

Line 357:     ADOPT_TRACE(trace.get());
What does the ADOPT_TRACE change?  Did this work before?


-- 
To view, visit http://gerrit.cloudera.org:8080/4172
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I39cd9438071aabbd2d7bdeca3269af8b83f2d55b
Gerrit-PatchSet: 1
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Dan Burkert <da...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-HasComments: Yes

[kudu-CR] Add trace metrics for maintenance ops, LBM writes

Posted by "Kudu Jenkins (Code Review)" <ge...@cloudera.org>.
Kudu Jenkins has posted comments on this change.

Change subject: Add trace metrics for maintenance ops, LBM writes
......................................................................


Patch Set 1:

Build Started http://104.196.14.100/job/kudu-gerrit/3150/

-- 
To view, visit http://gerrit.cloudera.org:8080/4172
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I39cd9438071aabbd2d7bdeca3269af8b83f2d55b
Gerrit-PatchSet: 1
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Dan Burkert <da...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-HasComments: No

[kudu-CR] Add trace metrics for maintenance ops, LBM writes

Posted by "Todd Lipcon (Code Review)" <ge...@cloudera.org>.
Todd Lipcon has posted comments on this change.

Change subject: Add trace metrics for maintenance ops, LBM writes
......................................................................


Patch Set 1:

(1 comment)

http://gerrit.cloudera.org:8080/#/c/4172/1/src/kudu/util/maintenance_manager.cc
File src/kudu/util/maintenance_manager.cc:

Line 357:     ADOPT_TRACE(trace.get());
> What does the ADOPT_TRACE change?  Did this work before?
the maintenance manager doesn't run in the context of an RPC, so there was no trace associated with the thread when it was running. So we have to manually construct one and adopt it here, or else the counters don't end up going anywhere.


-- 
To view, visit http://gerrit.cloudera.org:8080/4172
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I39cd9438071aabbd2d7bdeca3269af8b83f2d55b
Gerrit-PatchSet: 1
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Dan Burkert <da...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-HasComments: Yes

[kudu-CR] Add trace metrics for maintenance ops, LBM writes

Posted by "Todd Lipcon (Code Review)" <ge...@cloudera.org>.
Todd Lipcon has submitted this change and it was merged.

Change subject: Add trace metrics for maintenance ops, LBM writes
......................................................................


Add trace metrics for maintenance ops, LBM writes

This adds a trace-scoped counter for LogBlockManager writes, and also
adds traces and counter printing to the maintenance manager.

With this patch, flushes and compactions now output log messages like:

I0830 13:42:28.809852  2208 maintenance_manager.cc:356] Time spent running FlushMRSOp(4301e24175a2469fba37184430567acc): real 9.510s user 4.432s     sys 0.860s
I0830 13:42:28.810760  2208 maintenance_manager.cc:362] FlushMRSOp(4301e24175a2469fba37184430567acc) metrics: {"cfile_init":14,"compiler_manager_pool.queue_time_us":106,"compiler_manager_pool.run_cpu_time_us":219218,"compiler_manager_pool.run_wall_time_us":232004,"fdatasync":533,"fdatasync_us":67137,"lbm root 0.queue_time_us":1605,"lbm root 0.run_cpu_time_us":1280,"lbm root 0.run_wall_time_us":344957,"lbm_read_time_us":94,"lbm_reads_lt_1ms":56,"lbm_write_time_us":709833,"lbm_writes_gt_100_ms":1,"lbm_writes_lt_1ms":11564,"mutex_wait_us":343945,"tcmalloc_contention_cycles":2759552,"thread_start_us":1304,"threads_started":13}

This can be handy for understanding why a particular compaction or flush
was slow.

Change-Id: I39cd9438071aabbd2d7bdeca3269af8b83f2d55b
Reviewed-on: http://gerrit.cloudera.org:8080/4172
Tested-by: Kudu Jenkins
Reviewed-by: Dan Burkert <da...@cloudera.com>
---
M src/kudu/fs/log_block_manager.cc
M src/kudu/util/maintenance_manager.cc
M src/kudu/util/trace.h
3 files changed, 30 insertions(+), 10 deletions(-)

Approvals:
  Dan Burkert: Looks good to me, approved
  Kudu Jenkins: Verified



-- 
To view, visit http://gerrit.cloudera.org:8080/4172
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: merged
Gerrit-Change-Id: I39cd9438071aabbd2d7bdeca3269af8b83f2d55b
Gerrit-PatchSet: 2
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <to...@apache.org>
Gerrit-Reviewer: Dan Burkert <da...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>