You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "xufeng (Created) (JIRA)" <ji...@apache.org> on 2012/03/30 08:46:56 UTC
[jira] [Created] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
The master never does balance because duplicate openhandled the one region
--------------------------------------------------------------------------
Key: HBASE-5677
URL: https://issues.apache.org/jira/browse/HBASE-5677
Project: HBase
Issue Type: Bug
Affects Versions: 0.90.6
Environment: 0.90
Reporter: xufeng
Assignee: xufeng
If region be assigned When the master is doing initialization（before do processFailover）,the region will be duplicate openhandled.
it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Lars Hofhansl (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13252950#comment-13252950 ]
Lars Hofhansl commented on HBASE-5677:
--------------------------------------
I ran TestReplication with this patch (5577-proposal.txt) and it passed fine.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Lars Hofhansl (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13252533#comment-13252533 ]
Lars Hofhansl commented on HBASE-5677:
--------------------------------------
@Stack: This seems like a good addition to 0.94.
Let's say if we can track down the test failures today we'll put this in 0.94.0 otherwise 0.94.1
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "xufeng (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13252356#comment-13252356 ]
xufeng commented on HBASE-5677:
-------------------------------
@Ted @stack @Lars
I test it use trunk version.
then I got this in shell and my test case:
${noformat}
12/04/12 19:38:35 INFO client.HBaseAdmin: Started enable of Table02
org.apache.hadoop.hbase.PleaseHoldException: org.apache.hadoop.hbase.PleaseHoldException: Master is initializing
${noformat}
PleaseHoldException be added in HBASE-5454,the patch of this issue be integrated to trunk and 0.94 version.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Hadoop QA (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13252994#comment-13252994 ]
Hadoop QA commented on HBASE-5677:
----------------------------------
-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12522500/5677-proposal.txt
against trunk revision .
+1 @author. The patch does not contain any @author tags.
-1 tests included. The patch doesn't appear to include any new or modified tests.
Please justify why no new tests are needed for this patch.
Also please list what manual steps were performed to verify this patch.
+1 javadoc. The javadoc tool did not generate any warning messages.
+1 javac. The applied patch does not increase the total number of javac compiler warnings.
-1 findbugs. The patch appears to introduce 3 new Findbugs (version 1.3.9) warnings.
+1 release audit. The applied patch does not increase the total number of release audit warnings.
-1 core tests. The patch failed these unit tests:
org.apache.hadoop.hbase.replication.TestMasterReplication
org.apache.hadoop.hbase.replication.TestMultiSlaveReplication
org.apache.hadoop.hbase.regionserver.wal.TestHLog
org.apache.hadoop.hbase.regionserver.wal.TestHLogSplit
Test results: https://builds.apache.org/job/PreCommit-HBASE-Build/1502//testReport/
Findbugs warnings: https://builds.apache.org/job/PreCommit-HBASE-Build/1502//artifact/trunk/patchprocess/newPatchFindbugsWarnings.html
Console output: https://builds.apache.org/job/PreCommit-HBASE-Build/1502//console
This message is automatically generated.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
Posted by "Zhihong Yu (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Zhihong Yu updated HBASE-5677:
------------------------------
Fix Version/s: 0.96.0
0.94.0
0.92.2
0.90.7
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
Posted by "xufeng (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
xufeng updated HBASE-5677:
--------------------------
Attachment: surefire-report_patched_v1.html
surefire-report_no_patched_v1.html
HBASE-5677-90-v1.patch
@Ted
At first,I try to change the isMasterRunning like this:
public boolean isMasterRunning() {
return !isStopped() && isInitialized();
}
But Some class like HMerge,this tool just care the master is running or not.
So I think it is necessary to create a new method in HMasterInterface.
I create a 90 patch,pls review this and give me some suggestions,thanks.
I tested it by reproduce steps, shell can not work until the master completed the initialization.
And I also do the unit test,it seems ok.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Attachments: HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "xufeng (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13243668#comment-13243668 ]
xufeng commented on HBASE-5677:
-------------------------------
We can reproduce this issue by following steps with 0.90:
step1:start a cluster and create a table that has many regions.
step2:disable table created in step1 by shell.
step3:kill the active master.
step3:the backup master will become active one,when the master checkin regionservers. enable the table by shell.
result:the duplicate problem issue happened.
I think the master should not provide service when it did not complete the initialization.
We can add a method in HMasterInterface
like:
{noformat}
public boolean isMasterAvailable();
//the master is running and it can provide service
public boolean isMasterAvailable() {
return !isStopped() && isActiveMaster() && isInitialized();
}
{noformat}
When the client getMaster,we can check it.
pls give me the suggestions,thanks.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Lars Hofhansl (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13251189#comment-13251189 ]
Lars Hofhansl commented on HBASE-5677:
--------------------------------------
bq. public boolean isMasterRunning() { return !isStopped() && isInitialized(); } But Some class like HMerge,this tool just care the master is running or not.
Is that actually a problem? HMerge would need to wait until the master is initialized. Seems generally a better a better condition for "running" than just having the process up.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Hadoop QA (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13253593#comment-13253593 ]
Hadoop QA commented on HBASE-5677:
----------------------------------
-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12522508/5677-proposal.txt
against trunk revision .
+1 @author. The patch does not contain any @author tags.
-1 tests included. The patch doesn't appear to include any new or modified tests.
Please justify why no new tests are needed for this patch.
Also please list what manual steps were performed to verify this patch.
+1 javadoc. The javadoc tool did not generate any warning messages.
+1 javac. The applied patch does not increase the total number of javac compiler warnings.
-1 findbugs. The patch appears to introduce 3 new Findbugs (version 1.3.9) warnings.
+1 release audit. The applied patch does not increase the total number of release audit warnings.
-1 core tests. The patch failed these unit tests:
org.apache.hadoop.hbase.mapreduce.TestWALPlayer
Test results: https://builds.apache.org/job/PreCommit-HBASE-Build/1514//testReport/
Findbugs warnings: https://builds.apache.org/job/PreCommit-HBASE-Build/1514//artifact/trunk/patchprocess/newPatchFindbugsWarnings.html
Console output: https://builds.apache.org/job/PreCommit-HBASE-Build/1514//console
This message is automatically generated.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HBASE-5677:
-------------------------
Resolution: Won't Fix
Status: Resolved (was: Patch Available)
Will be fixed over in HBASE-5850.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, Backport-HBASE-5454-to-90.patch, Backport-HBASE-5454-to-92.patch, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Lars Hofhansl (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13250445#comment-13250445 ]
Lars Hofhansl commented on HBASE-5677:
--------------------------------------
Seems this should in 0.94.0. Agreed?
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Attachments: HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
Posted by "Lars Hofhansl (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Lars Hofhansl updated HBASE-5677:
---------------------------------
Attachment: 5677-proposal.txt
One more time
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Issue Comment Edited] (HBASE-5677) The master never does
balance because duplicate openhandled the one region
Posted by "Lars Hofhansl (Issue Comment Edited) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13252516#comment-13252516 ]
Lars Hofhansl edited comment on HBASE-5677 at 4/12/12 4:00 PM:
---------------------------------------------------------------
@xufeng: Are you saying your change is good to go in?
was (Author: lhofhansl):
@xufeng: Are you saying you change is good to go in?
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Lars Hofhansl (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13251880#comment-13251880 ]
Lars Hofhansl commented on HBASE-5677:
--------------------------------------
@xufeng: OK, thanks. This is caused by the client?
Back to my previous comment. I think a good definition for declaring the master "running" is when it is fully initialized. I'd still just add this extra check to isMasterRunning. Is anything failing with that? If so I'd like to understand why?
BTW. This is the only issue holding up the next RC for the first 0.94 release.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Lars Hofhansl (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13251006#comment-13251006 ]
Lars Hofhansl commented on HBASE-5677:
--------------------------------------
Is this at all similar to HBASE-5615?
Maybe here too, there is no point holding up 0.94.0 for this.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Issue Comment Edited] (HBASE-5677) The master never does
balance because duplicate openhandled the one region
Posted by "Zhihong Yu (Issue Comment Edited) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13250443#comment-13250443 ]
Zhihong Yu edited comment on HBASE-5677 at 4/10/12 3:58 AM:
------------------------------------------------------------
Some comments about coding style:
{code}
+ //the master is running and it can provide service
+ public boolean isMasterAvailable() {
+ return !isStopped() && isInitialized();
+ }
{code}
@Override is missing for the above method.
Please leave a space between // and the
Indentation for the return line should be 4 spaces. i.e. 'r' of return should be under 'b' of public.
{code}
+ if(isAvailable) {
{code}
Please leave a space between if and left parenthesis.
{code}
+ throw new MasterNotRunningException();
{code}
You can create a new exception in place of MasterNotRunningException above.
was (Author: zhihyu@ebaysf.com):
Some comments about coding style:
{code}
+ //the master is running and it can provide service
+ public boolean isMasterAvailable() {
+ return !isStopped() && isInitialized();
+ }
{code}
@Override is missing for the above method.
Please leave a space between // and the
Indentation for the return line should be 4 spaces. i.e. 'r' of return should be under 'b' of public.
{code}
+ if(isAvailable) {
{code}
Please leave a space between if and left parenthesis.
{code}
+ throw new MasterNotRunningException();
{code}
You can create a new exception or provide cause to MasterNotRunningException.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Attachments: HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Lars Hofhansl (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13252057#comment-13252057 ]
Lars Hofhansl commented on HBASE-5677:
--------------------------------------
Sigh... I think these are test-related issues. In a real cluster the Master is useless unless initialized (although I wouldn't know if that holds for a backup master as well).
I'll have a look at these when I get a chance.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Lars Hofhansl (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13252516#comment-13252516 ]
Lars Hofhansl commented on HBASE-5677:
--------------------------------------
@xufeng: Are you saying you change is good to go in?
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "xufeng (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13258171#comment-13258171 ]
xufeng commented on HBASE-5677:
-------------------------------
@stack
Yes,we should close this issue.
I will create a new issue to backport HBASE-5454 to 0.90,0.92.2 version.
And submit the patch that the checkinitialized method in createTable for trunk and 0.94 version.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, Backport-HBASE-5454-to-90.patch, Backport-HBASE-5454-to-92.patch, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Zhihong Yu (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13250442#comment-13250442 ]
Zhihong Yu commented on HBASE-5677:
-----------------------------------
The new method has traces of the existing method:
{code}
+ public boolean isMasterAvailable()
+ throws MasterNotRunningException, ZooKeeperConnectionException {
{code}
We should wait for other reviewers' comment about proper naming.
As always, please provide patch for trunk, 0.94 and 0.92.
Attaching patch for trunk would allow Hadoop QA to run.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Attachments: HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "xufeng (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13253196#comment-13253196 ]
xufeng commented on HBASE-5677:
-------------------------------
@Lars
Sorry,Something I can not undestand.
I think that this issue can be fixed by HBASE-5454.
Why we need 5677-proposal.txt patch for it?
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Zhihong Yu (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13252007#comment-13252007 ]
Zhihong Yu commented on HBASE-5677:
-----------------------------------
I saw the following in org.apache.hadoop.hbase.mapreduce.TestImportTsv.txt when I tested HBASE-5741 in 0.94, with the proposed patch in place:
{code}
Caused by: java.lang.RuntimeException: Master not initialized after 200 seconds
at org.apache.hadoop.hbase.util.JVMClusterUtil.startup(JVMClusterUtil.java:206)
at org.apache.hadoop.hbase.LocalHBaseCluster.startup(LocalHBaseCluster.java:422)
at org.apache.hadoop.hbase.MiniHBaseCluster.init(MiniHBaseCluster.java:196)
{code}
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
Posted by "Zhihong Yu (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Zhihong Yu updated HBASE-5677:
------------------------------
Attachment: (was: 5677-proposal.txt)
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "stack (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13256921#comment-13256921 ]
stack commented on HBASE-5677:
------------------------------
Xufeng So we should close this issue and backport hbase-5454 to 0.90 and to 0.92.2? Or would you rather make a new issue that adds check initialized to createTable for trunk and 0.94 and that has a new version of hbase-5454 that includes checkinitialized in the patch we put on 0.90 and 0.92?
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, Backport-HBASE-5454-to-90.patch, Backport-HBASE-5454-to-92.patch, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
Posted by "Zhihong Yu (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Zhihong Yu updated HBASE-5677:
------------------------------
Status: Patch Available (was: Open)
Run Lars' proposal through Hadoop QA
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
Posted by "Lars Hofhansl (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Lars Hofhansl updated HBASE-5677:
---------------------------------
Attachment: 5677-proposal.txt
Reattaching for new test run.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "stack (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13252218#comment-13252218 ]
stack commented on HBASE-5677:
------------------------------
I do not think this an issue to hold up a release. Its on failover of master and we have hbck to do fixup if needed.
The problem described is another form of the well-known case where concurrent handler operations coming in while a failover master is coming up causes the master confusion. We need to do a fundamental fix to plug this problem and its many manifestations.
-1 on adding an isMasterAvailable method. As stated above by Ted, it will cause confusion when put beside isMasterRunning. The Lars version, where it takes' xufeng's idea and adds isInitialized to isMasterRunning seems like a good compromise.
If you want to get this into 0.94 Lars, say so, and I'll try help out w/ the above failures tomorrow.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Hadoop QA (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13254572#comment-13254572 ]
Hadoop QA commented on HBASE-5677:
----------------------------------
-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12522752/Backport-HBASE-5454-to-92.patch
against trunk revision .
+1 @author. The patch does not contain any @author tags.
-1 tests included. The patch doesn't appear to include any new or modified tests.
Please justify why no new tests are needed for this patch.
Also please list what manual steps were performed to verify this patch.
-1 patch. The patch command could not apply the patch.
Console output: https://builds.apache.org/job/PreCommit-HBASE-Build/1536//console
This message is automatically generated.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, Backport-HBASE-5454-to-90.patch, Backport-HBASE-5454-to-92.patch, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "xufeng (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13249753#comment-13249753 ]
xufeng commented on HBASE-5677:
-------------------------------
@Ted
I test the 0.92 in my cluster by reproduce steps.
then I run the hbck tool to check the health of cluster and found many multiply error.
I think it also has problem in 0.92.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "xufeng (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13249754#comment-13249754 ]
xufeng commented on HBASE-5677:
-------------------------------
I got lasted 0.92 version(revision 1311105) from https://svn.apache.org/repos/asf/hbase/branches/0.92
then compiled it.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Hadoop QA (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13253031#comment-13253031 ]
Hadoop QA commented on HBASE-5677:
----------------------------------
-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12522508/5677-proposal.txt
against trunk revision .
+1 @author. The patch does not contain any @author tags.
-1 tests included. The patch doesn't appear to include any new or modified tests.
Please justify why no new tests are needed for this patch.
Also please list what manual steps were performed to verify this patch.
+1 javadoc. The javadoc tool did not generate any warning messages.
+1 javac. The applied patch does not increase the total number of javac compiler warnings.
-1 findbugs. The patch appears to introduce 3 new Findbugs (version 1.3.9) warnings.
+1 release audit. The applied patch does not increase the total number of release audit warnings.
-1 core tests. The patch failed these unit tests:
org.apache.hadoop.hbase.regionserver.wal.TestHLogSplit
org.apache.hadoop.hbase.replication.TestMultiSlaveReplication
org.apache.hadoop.hbase.regionserver.wal.TestHLog
org.apache.hadoop.hbase.replication.TestMasterReplication
Test results: https://builds.apache.org/job/PreCommit-HBASE-Build/1503//testReport/
Findbugs warnings: https://builds.apache.org/job/PreCommit-HBASE-Build/1503//artifact/trunk/patchprocess/newPatchFindbugsWarnings.html
Console output: https://builds.apache.org/job/PreCommit-HBASE-Build/1503//console
This message is automatically generated.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Lars Hofhansl (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13253549#comment-13253549 ]
Lars Hofhansl commented on HBASE-5677:
--------------------------------------
Arghh... OK. So:
* in 0.94+ this is fixed, correct?
* you like to backport HBASE-5454 to 0.90 and 0.92, right?
So let's close this one then?
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "xufeng (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13242118#comment-13242118 ]
xufeng commented on HBASE-5677:
-------------------------------
If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
I use the 0.90 vsersion.
I found this issue in my cluster.
1.The system did not do balance:
{noformat}
Not running balancer because 2 region(s) in transition:
{f4ff609df50e5bc9049fe202bb90f22e=hbase0205test,0038613850505050,1333033465665.f4ff609df50e5bc9049fe202bb90f22e.
state=OPEN, ts=1333036748502,
febe5bb42ec841f7a9086d3b7bf0637c=hbase0205test,0038613802020202,1333033465474.febe5bb42ec841f7a9086d3b7bf0637c...
{noformat}
2.Choose f4ff609df50e5bc9049fe202bb90f22e as a simple to track.
3.In master log I found:
logA：
{noformat}
Line 17884: [2012-03-29 15:05:08,082] [DEBUG] [MASTER_OPEN_REGION-158-1-130-18:20000-1] [org.apache.hadoop.hbase.master.handler.OpenedRegionHandler 138] The master has opened the region hbase0205test,0038613850505050,1333033465665.f4ff609df50e5bc9049fe202bb90f22e. that was online on serverName=158-1-130-18,20020,1332952904731, load=(requests=4444, regions=728, usedHeap=141, maxHeap=8165)
{noformat}
logB：
{noformat}
=Line 17885: [2012-03-29 15:05:08,082] [DEBUG] [master-158-1-130-18:20000] [org.apache.hadoop.hbase.master.handler.OpenedRegionHandler 138] Handling OPENED event for hbase0205test,0038613850505050,1333033465665.f4ff609df50e5bc9049fe202bb90f22e. from serverName=158-1-130-18,20020,1332952904731, load=(requests=245, regions=758, usedHeap=145, maxHeap=8165); deleting unassigned node
Line 17897: [2012-03-29 15:05:08,084] [DEBUG] [master-158-1-130-18:20000] [org.apache.hadoop.hbase.zookeeper.ZKAssign 511] master:20000-0x236552a09e20353 Deleting existing unassigned node for f4ff609df50e5bc9049fe202bb90f22e that is in expected state RS_ZK_REGION_OPENED
Line 17898: [2012-03-29 15:05:08,092] [WARN ] [master-158-1-130-18:20000] [org.apache.hadoop.hbase.master.handler.OpenedRegionHandler 123] The znode of the region hbase0205test,0038613850505050,1333033465665.f4ff609df50e5bc9049fe202bb90f22e. would have already been deleted
Line 17899: [2012-03-29 15:05:08,092] [ERROR] [master-158-1-130-18:20000] [org.apache.hadoop.hbase.master.handler.OpenedRegionHandler 97] The znode of region hbase0205test,0038613850505050,1333033465665.f4ff609df50e5bc9049fe202bb90f22e. could not be deleted.
{noformat}
4.The logA and logB should not appear at the same time,because belong to the same code in the region open flow.
5.So I ensure that this region has been handled duplicate.
6.Those log can explain what I write in Description:
Enable the table:
{noformat}
Line 16925: [2012-03-29 15:04:59,875] [DEBUG] [158-1-130-18:20000-org.apache.hadoop.hbase.master.handler.EnableTableHandler$BulkEnabler-0] [org.apache.hadoop.hbase.zookeeper.ZKAssign 289] master:20000-0x236552a09e20353 Creating (or updating) unassigned node for f4ff609df50e5bc9049fe202bb90f22e with OFFLINE state
{noformat}
Failover:
{noformat}
[2012-03-29 15:05:00,906] [INFO ] [master-158-1-130-18:20000] [org.apache.hadoop.hbase.master.AssignmentManager 284] Failed-over master needs to process 66 regions in transition
{noformat}
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
>
> If region be assigned When the master is doing initialization（before do processFailover）,the region will be duplicate openhandled.
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Zhihong Yu (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13250443#comment-13250443 ]
Zhihong Yu commented on HBASE-5677:
-----------------------------------
Some comments about coding style:
{code}
+ //the master is running and it can provide service
+ public boolean isMasterAvailable() {
+ return !isStopped() && isInitialized();
+ }
{code}
@Override is missing for the above method.
Please leave a space between // and the
Indentation for the return line should be 4 spaces. i.e. 'r' of return should be under 'b' of public.
{code}
+ if(isAvailable) {
{code}
Please leave a space between if and left parenthesis.
{code}
+ throw new MasterNotRunningException();
{code}
You can create a new exception or provide cause to MasterNotRunningException.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Attachments: HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
Posted by "xufeng (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
xufeng updated HBASE-5677:
--------------------------
Description:
If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
it cause the region in RIT,thus the master never does balance.
was:
If region be assigned When the master is doing initialization（before do processFailover）,the region will be duplicate openhandled.
it cause the region in RIT,thus the master never does balance.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "xufeng (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13253961#comment-13253961 ]
xufeng commented on HBASE-5677:
-------------------------------
@Lars
>>in 0.94+ this is fixed, correct?
yes.
>>you like to backport HBASE-5454 to 0.90 and 0.92, right?
ok.
But I also have a question about HBASE-5454(why did not add checkInitialized() in HMaster#createTable),I commented it in HBASE-5454.
Now I am at home,So no env to test it and create patch to backport in 90 and 92.
I plan to do it on Monday.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
Posted by "Ted Yu (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Ted Yu updated HBASE-5677:
--------------------------
Comment: was deleted
(was: -1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12522500/5677-proposal.txt
against trunk revision .
+1 @author. The patch does not contain any @author tags.
-1 tests included. The patch doesn't appear to include any new or modified tests.
Please justify why no new tests are needed for this patch.
Also please list what manual steps were performed to verify this patch.
+1 javadoc. The javadoc tool did not generate any warning messages.
+1 javac. The applied patch does not increase the total number of javac compiler warnings.
-1 findbugs. The patch appears to introduce 3 new Findbugs (version 1.3.9) warnings.
+1 release audit. The applied patch does not increase the total number of release audit warnings.
-1 core tests. The patch failed these unit tests:
org.apache.hadoop.hbase.replication.TestMasterReplication
org.apache.hadoop.hbase.replication.TestMultiSlaveReplication
org.apache.hadoop.hbase.regionserver.wal.TestHLog
org.apache.hadoop.hbase.regionserver.wal.TestHLogSplit
Test results: https://builds.apache.org/job/PreCommit-HBASE-Build/1502//testReport/
Findbugs warnings: https://builds.apache.org/job/PreCommit-HBASE-Build/1502//artifact/trunk/patchprocess/newPatchFindbugsWarnings.html
Console output: https://builds.apache.org/job/PreCommit-HBASE-Build/1502//console
This message is automatically generated.)
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "xufeng (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13251379#comment-13251379 ]
xufeng commented on HBASE-5677:
-------------------------------
@Lars
This issue cased by client.I think that it is not similar to HBASE-5615 in 0.90 at least.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "xufeng (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13258774#comment-13258774 ]
xufeng commented on HBASE-5677:
-------------------------------
@stack
see https://issues.apache.org/jira/browse/HBASE-5850
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, Backport-HBASE-5454-to-90.patch, Backport-HBASE-5454-to-92.patch, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Zhihong Yu (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13249821#comment-13249821 ]
Zhihong Yu commented on HBASE-5677:
-----------------------------------
There is this method in HMasterInterface:
{code}
/** @return true if master is available */
public boolean isMasterRunning();
{code}
If we introduce isMasterAvailable(), that would create confusion, right ?
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Hadoop QA (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13251954#comment-13251954 ]
Hadoop QA commented on HBASE-5677:
----------------------------------
-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12522314/5677-proposal.txt
against trunk revision .
+1 @author. The patch does not contain any @author tags.
-1 tests included. The patch doesn't appear to include any new or modified tests.
Please justify why no new tests are needed for this patch.
Also please list what manual steps were performed to verify this patch.
+1 javadoc. The javadoc tool did not generate any warning messages.
+1 javac. The applied patch does not increase the total number of javac compiler warnings.
+1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.
+1 release audit. The applied patch does not increase the total number of release audit warnings.
-1 core tests. The patch failed these unit tests:
org.apache.hadoop.hbase.replication.TestReplication
Test results: https://builds.apache.org/job/PreCommit-HBASE-Build/1484//testReport/
Findbugs warnings: https://builds.apache.org/job/PreCommit-HBASE-Build/1484//artifact/trunk/patchprocess/newPatchFindbugsWarnings.html
Console output: https://builds.apache.org/job/PreCommit-HBASE-Build/1484//console
This message is automatically generated.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Lars Hofhansl (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13253046#comment-13253046 ]
Lars Hofhansl commented on HBASE-5677:
--------------------------------------
@xufeng: I made a trunk patch, so that I can get a HadoopQA test run.
All the failures are unrelated, though. In fact they are all because of negative array sizes if WALEdit de-serialization, which is suspicious.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
Posted by "Zhihong Yu (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Zhihong Yu updated HBASE-5677:
------------------------------
Attachment: 5677-proposal.txt
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "xufeng (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13253198#comment-13253198 ]
xufeng commented on HBASE-5677:
-------------------------------
should we integrate the HBASE-5454 to 0.90 version?
I integrated the HBASE-5454 patch to 0.90 in my cluster,and it can work.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "xufeng (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13256184#comment-13256184 ]
xufeng commented on HBASE-5677:
-------------------------------
Pls review and if no problem,can we integrate it to 90 and 92?
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, Backport-HBASE-5454-to-90.patch, Backport-HBASE-5454-to-92.patch, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Ted Yu (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13243747#comment-13243747 ]
Ted Yu commented on HBASE-5677:
-------------------------------
Interesting.
Chunhui proposed safe mode for Master in HBASE-5270. See https://issues.apache.org/jira/browse/HBASE-5270?focusedCommentId=13214394&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13214394
Can you verify that this issue has been fixed in 0.92.2 ?
Thanks
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
Posted by "Lars Hofhansl (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Lars Hofhansl updated HBASE-5677:
---------------------------------
Status: Patch Available (was: Open)
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "xufeng (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13252358#comment-13252358 ]
xufeng commented on HBASE-5677:
-------------------------------
Test by trunk version is ok.
master do nothing if it has not initialized.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Lars Hofhansl (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13252953#comment-13252953 ]
Lars Hofhansl commented on HBASE-5677:
--------------------------------------
TestImportTsv passes as well.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
Posted by "Lars Hofhansl (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Lars Hofhansl updated HBASE-5677:
---------------------------------
Status: Open (was: Patch Available)
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "Zhihong Yu (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13251969#comment-13251969 ]
Zhihong Yu commented on HBASE-5677:
-----------------------------------
I ran TestReplication on MacBook twice and they both failed:
{code}
Failed tests: queueFailover(org.apache.hadoop.hbase.replication.TestReplication): Waited too much time for queueFailover replication
{code}
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
Posted by "xufeng (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
xufeng updated HBASE-5677:
--------------------------
Attachment: Backport-HBASE-5454-to-92.patch
Backport-HBASE-5454-to-90.patch
Backport-HBASE-5454-to-92.patch(this patch for 92):All unit tests passed..and also verified this patch in real cluste
Backport-HBASE-5454-to-90.patch(this patch for 90):Some test error,but those error also exist if no patched.and also verified this patch in real cluste
pls review,thanks.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, Backport-HBASE-5454-to-90.patch, Backport-HBASE-5454-to-92.patch, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-5677) The master never does balance
because duplicate openhandled the one region
Posted by "xufeng (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13253024#comment-13253024 ]
xufeng commented on HBASE-5677:
-------------------------------
@Lars
I did not change anything in trunk.
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2, 0.94.0, 0.96.0
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5677) The master never does balance because
duplicate openhandled the one region
Posted by "Lars Hofhansl (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Lars Hofhansl updated HBASE-5677:
---------------------------------
Fix Version/s: (was: 0.96.0)
(was: 0.94.0)
Removed 0.94 and 0.96 from "Fix Version/s"
> The master never does balance because duplicate openhandled the one region
> --------------------------------------------------------------------------
>
> Key: HBASE-5677
> URL: https://issues.apache.org/jira/browse/HBASE-5677
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.6
> Environment: 0.90
> Reporter: xufeng
> Assignee: xufeng
> Fix For: 0.90.7, 0.92.2
>
> Attachments: 5677-proposal.txt, 5677-proposal.txt, 5677-proposal.txt, HBASE-5677-90-v1.patch, surefire-report_no_patched_v1.html, surefire-report_patched_v1.html
>
>
> If region be assigned When the master is doing initialization(before do processFailover),the region will be duplicate openhandled.
> because the unassigned node in zookeeper will be handled again in AssignmentManager#processFailover()
> it cause the region in RIT,thus the master never does balance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira