[
https://issues.apache.org/jira/browse/HBASE-14239?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14703368#comment-14703368
]
Elliott Clark edited comment on HBASE-14239 at 8/19/15 5:03 PM:
----------------------------------------------------------------
I tested the v2 patch last night and things are better but not great yet.
I'm still getting some of these:
{code}
2015-08-19T09:19:24 Region server
hbase516.ash1.facebook.com,16020,1439963654848 reported a fatal error:
ABORTING region server hbase516.ash1.facebook.com,16020,1439963654848:
Exception running postOpenDeployTasks; region=142593938bea8fb99cb265258c19dfa1
Cause:
java.io.IOException: Failed to report opened region to master:
IntegegrationTestBigLinkedList.4,_\xCB%\xFC\xB2_\xCB
,1439965968754.142593938bea8fb99cb265258c19dfa1.
at
org.apache.hadoop.hbase.regionserver.HRegionServer.postOpenDeployTasks(HRegionServer.java:1927)
at
org.apache.hadoop.hbase.regionserver.handler.OpenRegionHandler$PostOpenDeployTasksThread.run(OpenRegionHandler.java:327)
{code}
I have all the regionservers set to restart so it wasn't too big a deal.
However for anyone without the auto-restart it would have been enough deaths to
kill the test.
was (Author: eclark):
I tested the v2 patch last night and things are better but not great yet.
I'm still getting some of these:
{code}
2015-08-19T09:19:24 Region server
hbase516.ash1.facebook.com,16020,1439963654848 reported a fatal error:
ABORTING region server hbase516.ash1.facebook.com,16020,1439963654848:
Exception running postOpenDeployTasks; region=142593938bea8fb99cb265258c19dfa1
Cause:
java.io.IOException: Failed to report opened region to master:
IntegegrationTestBigLinkedList.4,_\xCB%\xFC\xB2_\xCB
,1439965968754.142593938bea8fb99cb265258c19dfa1.
at
org.apache.hadoop.hbase.regionserver.HRegionServer.postOpenDeployTasks(HRegionServer.java:1927)
at
org.apache.hadoop.hbase.regionserver.handler.OpenRegionHandler$PostOpenDeployTasksThread.run(OpenRegionHandler.java:327)
{code}
I have all the regionserver set to restart so it wasn't too big a deal. However
for anyone without the auto-restart it would have been enough deaths to kill
the test.
> Branch-1.2 AM can get stuck when meta moves
> -------------------------------------------
>
> Key: HBASE-14239
> URL: https://issues.apache.org/jira/browse/HBASE-14239
> Project: HBase
> Issue Type: Sub-task
> Components: Region Assignment
> Affects Versions: 1.2.0
> Reporter: Elliott Clark
> Assignee: Elliott Clark
> Fix For: 1.2.0
>
> Attachments: HBASE-14239-v1.patch, HBASE-14239.patch
>
>
> When regions are moving master can get totally stuck trying to talk to meta.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
