You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "ramkrishna.s.vasudevan (Created) (JIRA)" <ji...@apache.org> on 2012/02/02 18:46:53 UTC
[jira] [Created] (HBASE-5323) Need to handle assertion error if
split log through ServerShutDownHandler by shutting down the master
Need to handle assertion error if split log through ServerShutDownHandler by shutting down the master
-----------------------------------------------------------------------------------------------------
Key: HBASE-5323
URL: https://issues.apache.org/jira/browse/HBASE-5323
Project: HBase
Issue Type: Bug
Affects Versions: 0.90.5
Reporter: ramkrishna.s.vasudevan
Fix For: 0.94.0, 0.90.7
We know that while parsing the HLog we expect the proper length from HDFS.
In WALReaderFSDataInputStream
{code}
assert(realLength >= this.length);
{code}
We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
I think in this condition we abort the master by catching such exceptions.
Please do suggest.
--
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-5323) Need to handle assertion error
while splitting log through ServerShutDownHandler by shutting down the
master
Posted by "ramkrishna.s.vasudevan (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13234582#comment-13234582 ]
ramkrishna.s.vasudevan commented on HBASE-5323:
-----------------------------------------------
Currently am not working on this Lars. You need not include in RC. Thanks
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.90.7, 0.94.1
>
> Attachments: HBASE-5323.patch, HBASE-5323.patch
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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-5323) Need to handle assertion error while
splitting log through ServerShutDownHandler by shutting down the master
Posted by "Lars Hofhansl (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Lars Hofhansl updated HBASE-5323:
---------------------------------
Fix Version/s: (was: 0.94.3)
0.94.4
On to the greener pastures of 0.94.4
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.90.8, 0.94.4
>
> Attachments: HBASE-5323.patch, HBASE-5323.patch
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-5323) Need to handle assertion error while
splitting log through ServerShutDownHandler by shutting down the master
Posted by "Lars Hofhansl (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Lars Hofhansl updated HBASE-5323:
---------------------------------
Fix Version/s: (was: 0.94.1)
0.94.2
Thanks Ram... Let's try for the next release.
I'll leave it to Jon whether he wants this in 0.90.7.
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.90.7, 0.94.2
>
> Attachments: HBASE-5323.patch, HBASE-5323.patch
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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-5323) Need to handle assertion error
while splitting log through ServerShutDownHandler by shutting down the
master
Posted by "Lars Hofhansl (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13234566#comment-13234566 ]
Lars Hofhansl commented on HBASE-5323:
--------------------------------------
Are you still working on this Ram? Do you think this needs to go into 0.94?
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.90.7, 0.94.0
>
> Attachments: HBASE-5323.patch, HBASE-5323.patch
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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-5323) Need to handle assertion error
while splitting log through ServerShutDownHandler by shutting down the
master
Posted by "Zhihong Yu (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13199029#comment-13199029 ]
Zhihong Yu commented on HBASE-5323:
-----------------------------------
Suggested approach makes sense.
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Fix For: 0.94.0, 0.90.7
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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-5323) Need to handle assertion error while
splitting log through ServerShutDownHandler by shutting down the master
Posted by "ramkrishna.s.vasudevan (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
ramkrishna.s.vasudevan updated HBASE-5323:
------------------------------------------
Attachment: HBASE-5323.patch
Patch for 0.90.
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.94.0, 0.90.7
>
> Attachments: HBASE-5323.patch
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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-5323) Need to handle assertion error
while splitting log through ServerShutDownHandler by shutting down the
master
Posted by "Lars Hofhansl (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13200331#comment-13200331 ]
Lars Hofhansl commented on HBASE-5323:
--------------------------------------
+1 on aborting the Master when this happens
+1 on changing the assert to try/catch.
Generally assert should only be used with conditions that we can live with in a production environment (as Todd points out).
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.94.0, 0.90.7
>
> Attachments: HBASE-5323.patch
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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-5323) Need to handle assertion error while
splitting log through ServerShutDownHandler by shutting down the master
Posted by "ramkrishna.s.vasudevan (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
ramkrishna.s.vasudevan updated HBASE-5323:
------------------------------------------
Attachment: HBASE-5323.patch
Patch for 0.90
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.94.0, 0.90.7
>
> Attachments: HBASE-5323.patch, HBASE-5323.patch
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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-5323) Need to handle assertion error while
splitting log through ServerShutDownHandler by shutting down the master
Posted by "ramkrishna.s.vasudevan (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
ramkrishna.s.vasudevan updated HBASE-5323:
------------------------------------------
Summary: Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master (was: Need to handle assertion error if split log through ServerShutDownHandler by shutting down the master)
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Fix For: 0.94.0, 0.90.7
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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-5323) Need to handle assertion error
while splitting log through ServerShutDownHandler by shutting down the
master
Posted by "Todd Lipcon (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13199917#comment-13199917 ]
Todd Lipcon commented on HBASE-5323:
------------------------------------
If this is an actual error condition that can be seen in practice (rather than indication of a failed programmer assumption) it should be changed to an if/throw, not an assert. Most users don't have asserts enabled in production.
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.94.0, 0.90.7
>
> Attachments: HBASE-5323.patch
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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-5323) Need to handle assertion error while
splitting log through ServerShutDownHandler by shutting down the master
Posted by "ramkrishna.s.vasudevan (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
ramkrishna.s.vasudevan updated HBASE-5323:
------------------------------------------
Fix Version/s: (was: 0.94.0)
0.94.1
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.90.7, 0.94.1
>
> Attachments: HBASE-5323.patch, HBASE-5323.patch
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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-5323) Need to handle assertion error
while splitting log through ServerShutDownHandler by shutting down the
master
Posted by "Zhihong Yu (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13205610#comment-13205610 ]
Zhihong Yu commented on HBASE-5323:
-----------------------------------
Some comments for patch v2.
In MasterFileSystem.java, the following check should be put as first in if statement because it is fast:
{code}
e instanceof HLogLengthMisMatchException
{code}
In HLog.java:
{code}
+ if (e instanceof HLogLengthMisMatchException) {
+ throw new HLogLengthMisMatchException(
{code}
I don't think we need the above. Instead, we should check for e.cause in MasterFileSystem.java
For HLogLengthMisMatchException.java, the M in Match should be lower case.
I think it should be a generic exception. So maybe rename it to FileLengthMismatchException ?
For InstrumentedSequenceLogReader, please add javadoc for the class.
This class should be generic as well. How about passing the exception that getPos() is supposed to throw to ctor of InstrumentedSequenceLogReader ?
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.94.0, 0.90.7
>
> Attachments: HBASE-5323.patch, HBASE-5323.patch
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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] [Assigned] (HBASE-5323) Need to handle assertion error while
splitting log through ServerShutDownHandler by shutting down the master
Posted by "ramkrishna.s.vasudevan (Assigned) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
ramkrishna.s.vasudevan reassigned HBASE-5323:
---------------------------------------------
Assignee: ramkrishna.s.vasudevan
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.94.0, 0.90.7
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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-5323) Need to handle assertion error
while splitting log through ServerShutDownHandler by shutting down the
master
Posted by "Zhihong Yu (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13205612#comment-13205612 ]
Zhihong Yu commented on HBASE-5323:
-----------------------------------
I would suggest starting with a patch for TRUNK which can be verified by Hadoop QA.
Once code review passes, you can backport to 0.90 branch.
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.94.0, 0.90.7
>
> Attachments: HBASE-5323.patch, HBASE-5323.patch
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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-5323) Need to handle assertion error while
splitting log through ServerShutDownHandler by shutting down the master
Posted by "Jonathan Hsieh (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Hsieh updated HBASE-5323:
----------------------------------
Fix Version/s: (was: 0.90.7)
0.90.8
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.94.2, 0.90.8
>
> Attachments: HBASE-5323.patch, HBASE-5323.patch
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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-5323) Need to handle assertion error
while splitting log through ServerShutDownHandler by shutting down the
master
Posted by "Jonathan Hsieh (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13412366#comment-13412366 ]
Jonathan Hsieh commented on HBASE-5323:
---------------------------------------
If it isn't making 0.94.1 then I'm going to bump it from 0.90.7.
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.94.2, 0.90.8
>
> Attachments: HBASE-5323.patch, HBASE-5323.patch
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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-5323) Need to handle assertion error
while splitting log through ServerShutDownHandler by shutting down the
master
Posted by "ramkrishna.s.vasudevan (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-5323?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13200329#comment-13200329 ]
ramkrishna.s.vasudevan commented on HBASE-5323:
-----------------------------------------------
@Todd
Ok.. I can update the patch accordingly. I agree with your suggestion.
> Need to handle assertion error while splitting log through ServerShutDownHandler by shutting down the master
> ------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5323
> URL: https://issues.apache.org/jira/browse/HBASE-5323
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.90.5
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.94.0, 0.90.7
>
> Attachments: HBASE-5323.patch
>
>
> We know that while parsing the HLog we expect the proper length from HDFS.
> In WALReaderFSDataInputStream
> {code}
> assert(realLength >= this.length);
> {code}
> We are trying to come out if the above condition is not satisfied. But if SSH.splitLog() gets this problem then it lands in the run method of EventHandler. This kills the SSH thread and so further assignment does not happen. If ROOT and META are to be assigned they cannot be.
> I think in this condition we abort the master by catching such exceptions.
> Please do suggest.
--
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