You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by "stack (JIRA)" <ji...@apache.org> on 2007/07/07 03:23:04 UTC
[jira] Created: (HADOOP-1574) [hbase] Concurrent table creates all
succeed
[hbase] Concurrent table creates all succeed
--------------------------------------------
Key: HADOOP-1574
URL: https://issues.apache.org/jira/browse/HADOOP-1574
Project: Hadoop
Issue Type: Bug
Components: contrib/hbase
Reporter: stack
Assignee: stack
Priority: Minor
If multiple threads try to create a table, all will succeed. Create table needs to be synchronized and the test for pre-existence needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-1574) [hbase] Concurrent creates of a table
named 'X' all succeed
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HADOOP-1574:
--------------------------
Attachment: hadoop-1574-v7.patch
Version 7. Added utility to HRegionInfo to return table name out of a region name.
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable-v3.patch, duptable.patch, hadoop-1574-v4.patch, hadoop-1574-v5.patch, hadoop-1574-v7.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-1574) [hbase] Concurrent creates of a
table named 'X' all succeed
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12511947 ]
stack commented on HADOOP-1574:
-------------------------------
I don't think so. This test was catching RemoteException and undoing it
to pull out TableExistsException. Below changed it to catch
TableExistsException, which should still hold after you do your new commit.
Good on you Jim,
St.Ack
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable-v3.patch, duptable.patch, hadoop-1574-v4.patch, hadoop-1574-v5.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-1574) [hbase] Concurrent creates of a table
named 'X' all succeed
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HADOOP-1574:
--------------------------
Attachment: duptable-v2.patch
Version 2. Adds throwing new TableExistsException if attempt at creating an existing table:
[hbase] Concurrent creates of a table named 'X' all succeed
Make createTable synchronized. Fix test for table 'already-exists'.
M src/contrib/hbase/src/test/org/apache/hadoop/hbase/TestTable.java
Refactor. Add assertions we're getting back right exceptions.
M src/contrib/hbase/src/java/org/apache/hadoop/hbase/HClient.java
Javadoc edits.
(createTable,createTableAsync): Add mention of new TableExistsException.
M src/contrib/hbase/src/java/org/apache/hadoop/hbase/HMaster.java
(createTable): Fix test for existing table. Throw new TableExistsException.
(deleteTable, shutdown, disableTable, enableTable): Make synchronized.
A src/contrib/hbase/src/java/org/apache/hadoop/hbase/TableExistsException.java
Added.
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-1574) [hbase] Concurrent creates of a table
named 'X' all succeed
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HADOOP-1574:
--------------------------
Attachment: hadoop-1574-v4.patch
This version implements a Jim Kellerman suggestion that does away with synchronize on createTable method and instead keeps a 'lock' per table currently being created. Also added better testing of concurrent table creations.
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable-v3.patch, duptable.patch, hadoop-1574-v4.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-1574) [hbase] Concurrent table creates all
succeed
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HADOOP-1574:
--------------------------
Attachment: duptable.patch
First cut at a fix.
> [hbase] Concurrent table creates all succeed
> --------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable.patch
>
>
> If multiple threads try to create a table, all will succeed. Create table needs to be synchronized and the test for pre-existence needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-1574) [hbase] Concurrent creates of a
table named 'X' all succeed
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12511948 ]
stack commented on HADOOP-1574:
-------------------------------
No. This test was catching RemoteException and undoing it to pull out TableExistsException. The new patch changes the test to instead catch TableExistsException, which should still hold after you do your new commit (Thanks for asking)
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable-v3.patch, duptable.patch, hadoop-1574-v4.patch, hadoop-1574-v5.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-1574) [hbase] Concurrent creates of a table
named 'X' all succeed
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HADOOP-1574:
--------------------------
Status: In Progress (was: Patch Available)
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable-v3.patch, duptable.patch, hadoop-1574-v4.patch, hadoop-1574-v5.patch, hadoop-1574-v7.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-1574) [hbase] Concurrent creates of a
table named 'X' all succeed
Posted by "Hadoop QA (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12512682 ]
Hadoop QA commented on HADOOP-1574:
-----------------------------------
+1
http://issues.apache.org/jira/secure/attachment/12361753/hadoop-1574-v7.patch applied and successfully tested against trunk revision r555813.
Test results: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/411/testReport/
Console output: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/411/console
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable-v3.patch, duptable.patch, hadoop-1574-v4.patch, hadoop-1574-v5.patch, hadoop-1574-v7.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-1574) [hbase] Concurrent creates of a table
named 'X' all succeed
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HADOOP-1574:
--------------------------
Resolution: Fixed
Status: Resolved (was: Patch Available)
Committed with below message.
HADOOP-1574 Concurrent creates of a table named 'X' all succeed
M src/contrib/hbase/src/test/org/apache/hadoop/hbase/TestTable.java
(testTable): Add checking of actual exceptions thrown and
assertions that we are getting right behavior. Add a test
that has ten clients concurrently trying to create same table.
M src/contrib/hbase/src/java/org/apache/hadoop/hbase/HClient.java
javadoc edit. Fix debug message that could give impression
table was found when it wasn't.
M src/contrib/hbase/src/java/org/apache/hadoop/hbase/HTableDescriptor.java
Added comment on table nameing.
M src/contrib/hbase/src/java/org/apache/hadoop/hbase/HMaster.java
(createTable): Refactored. Bulk moved to a private override.
Changed how check for existance is done.
M rc/contrib/hbase/src/java/org/apache/hadoop/hbase/HRegionInfo.java
(getTableNameFromRegionName): Utility method added.
A src/contrib/hbase/src/java/org/apache/hadoop/hbase/TableExistsException.java
Added.
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable-v3.patch, duptable.patch, hadoop-1574-v4.patch, hadoop-1574-v5.patch, hadoop-1574-v7.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-1574) [hbase] Concurrent creates of a
table named 'X' all succeed
Posted by "Hadoop QA (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12511691 ]
Hadoop QA commented on HADOOP-1574:
-----------------------------------
+1
http://issues.apache.org/jira/secure/attachment/12361533/duptable-v3.patch applied and successfully tested against trunk revision r555114.
Test results: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/394/testReport/
Console output: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/394/console
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable-v3.patch, duptable.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-1574) [hbase] Concurrent creates of a table
named 'X' all succeed
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HADOOP-1574:
--------------------------
Description:
If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
was:If multiple threads try to create a table, all will succeed. Create table needs to be synchronized and the test for pre-existence needs fixing.
Summary: [hbase] Concurrent creates of a table named 'X' all succeed (was: [hbase] Concurrent table creates all succeed)
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-1574) [hbase] Concurrent creates of a
table named 'X' all succeed
Posted by "Jim Kellerman (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12511939 ]
Jim Kellerman commented on HADOOP-1574:
---------------------------------------
Do you think it will need to be changed again if HADOOP-1589 is committed?
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable-v3.patch, duptable.patch, hadoop-1574-v4.patch, hadoop-1574-v5.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-1574) [hbase] Concurrent creates of a table
named 'X' all succeed
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HADOOP-1574:
--------------------------
Attachment: duptable-v3.patch
Break up createTable so can narrow the HMaster synchronization scope.
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable-v3.patch, duptable.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-1574) [hbase] Concurrent creates of a table
named 'X' all succeed
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HADOOP-1574:
--------------------------
Status: Patch Available (was: Open)
Builds and passes all tests locally. Trying against hudson.
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable-v3.patch, duptable.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-1574) [hbase] Concurrent creates of a table
named 'X' all succeed
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HADOOP-1574:
--------------------------
Status: Patch Available (was: In Progress)
Builds local and passes all tests. Trying against hudson.
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable-v3.patch, duptable.patch, hadoop-1574-v4.patch, hadoop-1574-v5.patch, hadoop-1574-v7.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
Re: [jira] Commented: (HADOOP-1574) [hbase] Concurrent creates of a table named 'X' all succeed
Posted by Nigel Daley <nd...@yahoo-inc.com>.
I resubmitted it to Hudson. If it hangs again, I'll get a thread
dump before I kill it.
Cheers,
Nige
On Jul 13, 2007, at 4:10 PM, Michael Stack wrote:
> Nigel Daley wrote:
>> I had to kill TestTableMapReduce on the patch build since it had
>> hung.
>>
>
> TestTableMapReduce is a long-running test. Its 2 minutes on a
> local machine (To be addressed). But looking at console output,
> org.apache.hadoop.fs.TestCopyFiles fails first with an Error. My
> patch builds locally fine (I just retried it).
>
> What would you suggest I do Nigel? Should I try a resubmit?
>
> Thanks,
> St.Ack
>
>
>
>
>
>> On Jul 13, 2007, at 2:54 PM, Hadoop QA (JIRA) wrote:
>>
>>>
>>> [ https://issues.apache.org/jira/browse/HADOOP-1574?
>>> page=com.atlassian.jira.plugin.system.issuetabpanels:comment-
>>> tabpanel#action_12512633 ]
>>>
>>> Hadoop QA commented on HADOOP-1574:
>>> -----------------------------------
>>>
>>> -1, build or testing failed
>>>
>>> 2 attempts failed to build and test the latest attachment http://
>>> issues.apache.org/jira/secure/attachment/12361753/hadoop-1574-
>>> v7.patch against trunk revision r555813.
>>>
>>> Test results: http://lucene.zones.apache.org:8080/hudson/job/
>>> Hadoop-Patch/406/testReport/
>>> Console output: http://lucene.zones.apache.org:8080/hudson/job/
>>> Hadoop-Patch/406/console
>>>
>>> Please note that this message is automatically generated and may
>>> represent a problem with the automation system and not the patch.
>>>
>>>> [hbase] Concurrent creates of a table named 'X' all succeed
>>>> -----------------------------------------------------------
>>>>
>>>> Key: HADOOP-1574
>>>> URL: https://issues.apache.org/jira/browse/
>>>> HADOOP-1574
>>>> Project: Hadoop
>>>> Issue Type: Bug
>>>> Components: contrib/hbase
>>>> Reporter: stack
>>>> Assignee: stack
>>>> Priority: Minor
>>>> Attachments: duptable-v2.patch, duptable-v3.patch,
>>>> duptable.patch, hadoop-1574-v4.patch, hadoop-1574-v5.patch,
>>>> hadoop-1574-v7.patch
>>>>
>>>>
>>>> If multiple threads all try to create a table of a particular,
>>>> all succeed. It makes for a mess.
>>>> Create table needs to be synchronized so only one thread creates
>>>> a table and subsequent threads fail with 'table already exits'.
>>>> The test for pre-existence also needs fixing.
>>>
>>> --This message is automatically generated by JIRA.
>>> -
>>> You can reply to this email to add a comment to the issue online.
>>>
>>
>
Re: [jira] Commented: (HADOOP-1574) [hbase] Concurrent creates of
a table named 'X' all succeed
Posted by Michael Stack <st...@duboce.net>.
Nigel Daley wrote:
> I had to kill TestTableMapReduce on the patch build since it had hung.
>
TestTableMapReduce is a long-running test. Its 2 minutes on a local
machine (To be addressed). But looking at console output,
org.apache.hadoop.fs.TestCopyFiles fails first with an Error. My patch
builds locally fine (I just retried it).
What would you suggest I do Nigel? Should I try a resubmit?
Thanks,
St.Ack
> On Jul 13, 2007, at 2:54 PM, Hadoop QA (JIRA) wrote:
>
>>
>> [
>> https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12512633
>> ]
>>
>> Hadoop QA commented on HADOOP-1574:
>> -----------------------------------
>>
>> -1, build or testing failed
>>
>> 2 attempts failed to build and test the latest attachment
>> http://issues.apache.org/jira/secure/attachment/12361753/hadoop-1574-v7.patch
>> against trunk revision r555813.
>>
>> Test results:
>> http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/406/testReport/
>>
>> Console output:
>> http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/406/console
>>
>> Please note that this message is automatically generated and may
>> represent a problem with the automation system and not the patch.
>>
>>> [hbase] Concurrent creates of a table named 'X' all succeed
>>> -----------------------------------------------------------
>>>
>>> Key: HADOOP-1574
>>> URL: https://issues.apache.org/jira/browse/HADOOP-1574
>>> Project: Hadoop
>>> Issue Type: Bug
>>> Components: contrib/hbase
>>> Reporter: stack
>>> Assignee: stack
>>> Priority: Minor
>>> Attachments: duptable-v2.patch, duptable-v3.patch,
>>> duptable.patch, hadoop-1574-v4.patch, hadoop-1574-v5.patch,
>>> hadoop-1574-v7.patch
>>>
>>>
>>> If multiple threads all try to create a table of a particular, all
>>> succeed. It makes for a mess.
>>> Create table needs to be synchronized so only one thread creates a
>>> table and subsequent threads fail with 'table already exits'. The
>>> test for pre-existence also needs fixing.
>>
>> --This message is automatically generated by JIRA.
>> -
>> You can reply to this email to add a comment to the issue online.
>>
>
Re: [jira] Commented: (HADOOP-1574) [hbase] Concurrent creates of a table named 'X' all succeed
Posted by Nigel Daley <nd...@yahoo-inc.com>.
I had to kill TestTableMapReduce on the patch build since it had hung.
On Jul 13, 2007, at 2:54 PM, Hadoop QA (JIRA) wrote:
>
> [ https://issues.apache.org/jira/browse/HADOOP-1574?
> page=com.atlassian.jira.plugin.system.issuetabpanels:comment-
> tabpanel#action_12512633 ]
>
> Hadoop QA commented on HADOOP-1574:
> -----------------------------------
>
> -1, build or testing failed
>
> 2 attempts failed to build and test the latest attachment http://
> issues.apache.org/jira/secure/attachment/12361753/hadoop-1574-
> v7.patch against trunk revision r555813.
>
> Test results: http://lucene.zones.apache.org:8080/hudson/job/
> Hadoop-Patch/406/testReport/
> Console output: http://lucene.zones.apache.org:8080/hudson/job/
> Hadoop-Patch/406/console
>
> Please note that this message is automatically generated and may
> represent a problem with the automation system and not the patch.
>
>> [hbase] Concurrent creates of a table named 'X' all succeed
>> -----------------------------------------------------------
>>
>> Key: HADOOP-1574
>> URL: https://issues.apache.org/jira/browse/
>> HADOOP-1574
>> Project: Hadoop
>> Issue Type: Bug
>> Components: contrib/hbase
>> Reporter: stack
>> Assignee: stack
>> Priority: Minor
>> Attachments: duptable-v2.patch, duptable-v3.patch,
>> duptable.patch, hadoop-1574-v4.patch, hadoop-1574-v5.patch,
>> hadoop-1574-v7.patch
>>
>>
>> If multiple threads all try to create a table of a particular, all
>> succeed. It makes for a mess.
>> Create table needs to be synchronized so only one thread creates a
>> table and subsequent threads fail with 'table already exits'. The
>> test for pre-existence also needs fixing.
>
> --
> This message is automatically generated by JIRA.
> -
> You can reply to this email to add a comment to the issue online.
>
[jira] Commented: (HADOOP-1574) [hbase] Concurrent creates of a
table named 'X' all succeed
Posted by "Hadoop QA (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12512633 ]
Hadoop QA commented on HADOOP-1574:
-----------------------------------
-1, build or testing failed
2 attempts failed to build and test the latest attachment http://issues.apache.org/jira/secure/attachment/12361753/hadoop-1574-v7.patch against trunk revision r555813.
Test results: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/406/testReport/
Console output: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/406/console
Please note that this message is automatically generated and may represent a problem with the automation system and not the patch.
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable-v3.patch, duptable.patch, hadoop-1574-v4.patch, hadoop-1574-v5.patch, hadoop-1574-v7.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-1574) [hbase] Concurrent creates of a table
named 'X' all succeed
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1574?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HADOOP-1574:
--------------------------
Attachment: hadoop-1574-v5.patch
Version 5 because version 4 tests no longer work since RemoteException handling changed (HADOOP-1589).
> [hbase] Concurrent creates of a table named 'X' all succeed
> -----------------------------------------------------------
>
> Key: HADOOP-1574
> URL: https://issues.apache.org/jira/browse/HADOOP-1574
> Project: Hadoop
> Issue Type: Bug
> Components: contrib/hbase
> Reporter: stack
> Assignee: stack
> Priority: Minor
> Attachments: duptable-v2.patch, duptable-v3.patch, duptable.patch, hadoop-1574-v4.patch, hadoop-1574-v5.patch
>
>
> If multiple threads all try to create a table of a particular, all succeed. It makes for a mess.
> Create table needs to be synchronized so only one thread creates a table and subsequent threads fail with 'table already exits'. The test for pre-existence also needs fixing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.