HBASE-23213 : Reopen regions with very high Store Ref Counts(backport… #761
Conversation
This file contains 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
|
💔 -1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
virajjasani
force-pushed
the
HBASE-23213-branch-1
branch
from
e52c85e to
88232a9
Compare
|
💔 -1 overall
This message was automatically generated. |
|
TestAdmin1 is passing locally |
|
This is branch-1 backport PR. |
apurtell
approved these changes
Oct 28, 2019
| /** | ||
| * @return map of the names of region servers on the live list with associated ServerLoad | ||
| */ | ||
| public Map<ServerName, ServerLoad> getLiveServersLoad() { |
There was a problem hiding this comment.
Additive change to Public interface is fine. We are already also changing RegionLoad.
| * The max number of references active on single store file among all store files | ||
| * that belong to given region | ||
| */ | ||
| optional int32 max_store_file_ref_count = 22 [default = 0]; |
There was a problem hiding this comment.
Checked branch-2 and master proto def, this is consistent with them, good
|
+1 from me, merging soon. Shout if you want to stop it. |
|
I am not very comfortable having this test failing here: This test is not part of our branch-1 flaky tests list:https://builds.apache.org/view/H-L/view/HBase/job/HBase-Find-Flaky-Tests/job/branch-1/lastSuccessfulBuild/artifact/dashboard.html#job_2 And having value 2 here meaning the region merge failed. |
|
Ok @xcangCRM , holding off |
|
So the question is this test unstable before this change or caused by this change. @virajjasani if you don't have time to run TestAdmin1 at head of branch-1 in a loop 20 times or so to check that all runs are green before your change (or not), I'll pick it up tomorrow. |
|
@apurtell @xcangCRM With and without this patch applied, the results remain same:
Will try more runs by EOD and check if any discrepancy is found between branch-1 and this patch. |
|
ok so now I just ran TestAdmin1.testMergeRegions 50 times in loop with this patch and it went smooth without any failure. |
|
I'll do the same. If we can't correlate the failure, I will proceed. |
|
Sounds good to me! thank you. @virajjasani @apurtell |
|
On my new dev laptop testing TestAdmin1 is problematic: This happens before applying the patch here. I'm going to bisect to find where this was introduced. Edit: This looks like an old problem with JVMs doing weird things with binding to localhost when on VPN on MacOS. The Azul JVM may be at issue. I will test on a linux host to work around this. |
|
25 iterations with this patch look good. @xcangCRM if you continue to see instability with TestAdmin1 let's open a JIRA, and please post logs from failed runs there. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
… HBASE-22460)