[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16469838#comment-16469838
]
Zheng Hu commented on HBASE-20475:
--
Sure, filed HBASE-20560 for it .
> Fix the flaky
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16468832#comment-16468832
]
Duo Zhang commented on HBASE-20475:
---
The patch is a bit big as an addendum, let's open a new issue for
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16468727#comment-16468727
]
Zheng Hu commented on HBASE-20475:
--
Ping [~Apache9]
> Fix the flaky TestReplicationDroppedTables unit
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16467225#comment-16467225
]
Zheng Hu commented on HBASE-20475:
--
Let me explain the changes in UT refactor :
1. Let
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16465790#comment-16465790
]
Zheng Hu commented on HBASE-20475:
--
I guess we were stopping a ReplicationSourceShipper which has not
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16465766#comment-16465766
]
Duo Zhang commented on HBASE-20475:
---
What about the NPE posted by me above? Is there a race?
> Fix the
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16465759#comment-16465759
]
Zheng Hu commented on HBASE-20475:
--
Checked the UT & log again, the phenomenon is:
{code}
>
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16463684#comment-16463684
]
Zheng Hu commented on HBASE-20475:
--
Filed HBASE-20531 to address the above NPE.
> Fix the flaky
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16463322#comment-16463322
]
Zheng Hu commented on HBASE-20475:
--
Oh, the hadoop QA results
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16463285#comment-16463285
]
Zheng Hu commented on HBASE-20475:
--
Let me dig the NPE.
> Fix the flaky TestReplicationDroppedTables
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16457702#comment-16457702
]
Hudson commented on HBASE-20475:
Results for branch branch-2
[build #670 on
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16457593#comment-16457593
]
Hudson commented on HBASE-20475:
Results for branch master
[build #314 on
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16457584#comment-16457584
]
Duo Zhang commented on HBASE-20475:
---
I think there is a race... Need to dig more, our locking still has
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16457582#comment-16457582
]
Duo Zhang commented on HBASE-20475:
---
{noformat}
2018-04-28 11:21:41,825 WARN
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16457512#comment-16457512
]
Duo Zhang commented on HBASE-20475:
---
[~openinx] is probably on the train to home. Let me help
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16457456#comment-16457456
]
Duo Zhang commented on HBASE-20475:
---
It works. Please merge the patches and commit it to branch-2.
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16457281#comment-16457281
]
Zheng Hu commented on HBASE-20475:
--
The TestReplicationDroppedTables is becoming more stable now, still
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16456682#comment-16456682
]
Hadoop QA commented on HBASE-20475:
---
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16456444#comment-16456444
]
Zheng Hu commented on HBASE-20475:
--
Thanks [~Apache9] for reviewing, Pushed the addendum to master
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16456321#comment-16456321
]
Duo Zhang commented on HBASE-20475:
---
{code}
peersSnapshot = new HashMap<>();
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16456255#comment-16456255
]
Hadoop QA commented on HBASE-20475:
---
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16454297#comment-16454297
]
Zheng Hu commented on HBASE-20475:
--
Found the cause why the test to be flaky again, and it happened
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16454240#comment-16454240
]
Duo Zhang commented on HBASE-20475:
---
Please open an issue to fix it...
> Fix the flaky
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16454213#comment-16454213
]
Zheng Hu commented on HBASE-20475:
--
Found the above bug, because I notice the log :
{code}
2018-04-25
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16454202#comment-16454202
]
Zheng Hu commented on HBASE-20475:
--
Found an unrelated bug in
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16453788#comment-16453788
]
Hadoop QA commented on HBASE-20475:
---
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16453579#comment-16453579
]
Zheng Hu commented on HBASE-20475:
--
Two possible causes , I think
1. The host asf915.gq1.ygridcore.net
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16453565#comment-16453565
]
Zheng Hu commented on HBASE-20475:
--
Checked the console log
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16453377#comment-16453377
]
Duo Zhang commented on HBASE-20475:
---
See http://104.198.223.121:8080/job/HBASE-Flaky-Tests/36000/
>
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16452183#comment-16452183
]
Hudson commented on HBASE-20475:
Results for branch master
[build #311 on
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16451869#comment-16451869
]
Zheng Hu commented on HBASE-20475:
--
Checked branch-2, it's flaky too. If the UT is not flaky in master
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16451853#comment-16451853
]
Zheng Hu commented on HBASE-20475:
--
Thanks [~Apache9] for checking, Pushed to master branch now. I'm
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16451806#comment-16451806
]
Duo Zhang commented on HBASE-20475:
---
Checked locally, it works for me.
[~openinx] Please commit to
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16451731#comment-16451731
]
Hadoop QA commented on HBASE-20475:
---
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16449840#comment-16449840
]
Duo Zhang commented on HBASE-20475:
---
It is known to be flakey so the pre commit will not run it. But it
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16449833#comment-16449833
]
Zheng Hu commented on HBASE-20475:
--
I've tried to checkout some commits before HBASE-20128, and run the
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16449822#comment-16449822
]
Duo Zhang commented on HBASE-20475:
---
But we haven't touched the related code recently, so why it did
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16449791#comment-16449791
]
Zheng Hu commented on HBASE-20475:
--
It will block the progress of replication if we deleted the table
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16449775#comment-16449775
]
Zheng Hu commented on HBASE-20475:
--
I think I found the root cause finally. the source replicate a
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16449563#comment-16449563
]
Zheng Hu commented on HBASE-20475:
--
OK, the concurrent batch replication for serial replication is not
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16449555#comment-16449555
]
Duo Zhang commented on HBASE-20475:
---
I think we can introduce a flag. If the peer is serial then we
[
https://issues.apache.org/jira/browse/HBASE-20475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16449514#comment-16449514
]
Zheng Hu commented on HBASE-20475:
--
When dig the UT, I found that HBaseInterClusterReplicationEndpoint
42 matches
Mail list logo
