Uh oh!
There was an error while loading. Please reload this page.
HBASE-24779 Report on the WAL edit buffer usage/limit for replication - #2193
Closed
joshelser wants to merge 6 commits into
Closed
HBASE-24779 Report on the WAL edit buffer usage/limit for replication#2193joshelser wants to merge 6 commits into
joshelser wants to merge 6 commits into
Conversation
Apache-HBase
commented
Aug 4, 2020
🎊 +1 overall
This message was automatically generated. |
busbey
reviewed
Aug 4, 2020
| @Override | ||
| public void setWALReaderEditsBufferBytes(long usage) { | ||
| //noop. Global limit, tracked globally. Do not need per-source metrics |
Contributor
There was a problem hiding this comment.
nit: wouldn't it still be useful to know if particular sources were eating up the global limit? or do we not have enough information tracked already to do that?
MemberAuthor
There was a problem hiding this comment.
Once we chuck something into this usage, we have zero insight back to which source put it there. Wellington was working on this buffer tracking in HBASE-24813, but I don't think per-source tracking was "in scope"
Contributor
There was a problem hiding this comment.
Once we chuck something into this usage, we have zero insight back to which source put it there.
I had the same question. I was wondering if a drill down by source would be helpful in addition to the global usage. Correct me if I'm wrong, looks like ReplicationSource class has access to the MetricsSource object.. we can just update the byte usage for that source? (and the global too at the same time). That way we can also get rid of the special logic to update one metric setWALReaderEditsBufferBytes(). Does that not work for some reason?
MemberAuthor
There was a problem hiding this comment.
looks like ReplicationSource class has access to the MetricsSource object.. we can just update the byte usage for that source? (and the global too at the same time). That way we can also get rid of the special logic to update one metric setWALReaderEditsBufferBytes()
You are correct that we could do that. I wanted to keep this change scoped on "make what we currently have reportable". I am all for doing a per-source tracking in addition to the globally-scoped tracking. I'd rather just keep these two things separate :)
Contributor
There was a problem hiding this comment.
Oh, I see. Looks like you pushed the metrics update logic into the source, the patch looks clean now.
MemberAuthor
There was a problem hiding this comment.
yup! Just made another interface to isolate the new additions which cleans this up a little. I think your suggestion is still good for better, fine-grained tracking. However, since HBASE-20417, hopefully no one else runs into this ;)
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
Apache-HBase
commented
Aug 4, 2020
💔 -1 overall
This message was automatically generated. |
wchevreuil
approved these changes
Aug 4, 2020
wchevreuil
left a comment
Contributor
There was a problem hiding this comment.
LGTM overall, just small nits, but good to go without it. There might be some new checkstyle issues, fine to commit it once the build reports ok.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
Apache-HBase
commented
Aug 4, 2020
💔 -1 overall
This message was automatically generated. |
busbey
approved these changes
Aug 4, 2020
busbey
left a comment
Contributor
There was a problem hiding this comment.
good to merge imho after fixing the license header
Uh oh!
There was an error while loading. Please reload this page.
bharathv
approved these changes
Aug 4, 2020
bharathv
left a comment
Contributor
There was a problem hiding this comment.
+1 once the qabot is green.
joshelser
commented
Aug 4, 2020
MemberAuthor
Thanks folks! Just pushed a couple more commits for cleanup on QA and wellington's suggestions. I'll merge when QA is happy. |
Apache-HBase
commented
Aug 4, 2020
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Aug 4, 2020
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Aug 4, 2020
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Aug 4, 2020
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Aug 4, 2020
💔 -1 overall
This message was automatically generated. |
joshelser
commented
Aug 4, 2020
MemberAuthor
Interesting. TestWALEntryStream timed out for me too. Will dig in. |
joshelser
commented
Aug 4, 2020
MemberAuthor
Ah, a mocking issue. Fixing. |
* Fix up metrics source interface hierarchy * Docs for test replication endpoint * Push down metrics updating to mutation of the buffer usage variable (rather than on a timer)
joshelserforce-pushed
the
24779-check-quota-insight
branch
from
August 6, 2020 21:10
e2c75b5 to
7695a70Comparejoshelser
commented
Aug 6, 2020
MemberAuthor
I got 99 problems and all of them are due to mocking. The second UT failure was also due to a mock returning a value of 0 instead of the |
Apache-HBase
commented
Aug 6, 2020
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Aug 7, 2020
💔 -1 overall
This message was automatically generated. |
joshelser
commented
Aug 7, 2020
MemberAuthor
TestReplicationSource failure appears to be due to HBASE-24817 (#2198) We end up filtering out an edit because it's empty, but we expect to not filter it because it's not a system table edit. I'll make the same mocking changes to prevent future pain in TestReplicationSource (like I did for TestWALEntryStream) in a follow-on, as well as fix this test. FYI @saintstack Pushing this to make some progress. |
asfgit pushed a commit
that referenced
this pull request
Aug 7, 2020
Closes#2193 Signed-off-by: Bharath Vissapragada <bharathv@apache.org> Signed-off-by: Sean Busbey <busbey@apache.org> Signed-off-by: Wellington Chevreuil <wchevreuil@apache.org>
joshelser
commented
Aug 7, 2020
MemberAuthor
Filed https://issues.apache.org/jira/browse/HBASE-24834 to address the TestReplicationSource failure mentioned above. |
tamasadami pushed a commit
to tamasadami/hbase
that referenced
this pull request
Sep 9, 2020
Closesapache#2193 Signed-off-by: Bharath Vissapragada <bharathv@apache.org> Signed-off-by: Sean Busbey <busbey@apache.org> Signed-off-by: Wellington Chevreuil <wchevreuil@apache.org> (cherry picked from commit 303db63) Change-Id: If1b3662792304747d2942dbc52338e2108b1a764
tamasadami pushed a commit
to tamasadami/hbase
that referenced
this pull request
Sep 9, 2020
Closesapache#2193 Signed-off-by: Bharath Vissapragada <bharathv@apache.org> Signed-off-by: Sean Busbey <busbey@apache.org> Signed-off-by: Wellington Chevreuil <wchevreuil@apache.org> (cherry picked from commit 303db63)
tamasadami pushed a commit
to tamasadami/hbase
that referenced
this pull request
Oct 19, 2020
Closesapache#2193 Signed-off-by: Bharath Vissapragada <bharathv@apache.org> Signed-off-by: Sean Busbey <busbey@apache.org> Signed-off-by: Wellington Chevreuil <wchevreuil@apache.org> (cherry picked from commit 303db63) Change-Id: If1b3662792304747d2942dbc52338e2108b1a764
tamasadami pushed a commit
to tamasadami/hbase
that referenced
this pull request
Oct 19, 2020
Closesapache#2193 Signed-off-by: Bharath Vissapragada <bharathv@apache.org> Signed-off-by: Sean Busbey <busbey@apache.org> Signed-off-by: Wellington Chevreuil <wchevreuil@apache.org> (cherry picked from commit 303db63) Change-Id: If1b3662792304747d2942dbc52338e2108b1a764
clarax pushed a commit
to clarax/hbase
that referenced
this pull request
Nov 15, 2020
Closesapache#2193 Signed-off-by: Bharath Vissapragada <bharathv@apache.org> Signed-off-by: Sean Busbey <busbey@apache.org> Signed-off-by: Wellington Chevreuil <wchevreuil@apache.org>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for freeto join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Testing done:
And then to observe:
replication.stats.thread.period.secondsto15in hbase-site.xml and tail regionserver logwatch -n 15 "curl -s http://mizar.local:16030/jmx | jq '.beans[] | select(.name == \"Hadoop:service=HBase,name=RegionServer,sub=Replication\") | .\"source.walReaderEditsBufferUsage\"'"Can see the buffer grow. When the puts run out, the buffer eventually drains.