You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "stack (JIRA)" <ji...@apache.org> on 2010/09/10 08:04:33 UTC

[jira] Created: (HBASE-2979) Fix failing TestMultParrallel in hudson build

Fix failing TestMultParrallel in hudson build
---------------------------------------------

                 Key: HBASE-2979
                 URL: https://issues.apache.org/jira/browse/HBASE-2979
             Project: HBase
          Issue Type: Bug
            Reporter: stack
            Assignee: stack
             Fix For: 0.90.0


Its failing w/ a while now.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Updated: (HBASE-2979) Fix failing TestMultParrallel in hudson build

Posted by "stack (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HBASE-2979?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack updated HBASE-2979:
-------------------------

    Attachment: 2979.txt

Not done yet.  Made it junit4 and HBaseTestingUtility while I was at it because its long running test -- was starting a cluster between each test.  Also moved it down into client package where it belongs.

> Fix failing TestMultParrallel in hudson build
> ---------------------------------------------
>
>                 Key: HBASE-2979
>                 URL: https://issues.apache.org/jira/browse/HBASE-2979
>             Project: HBase
>          Issue Type: Bug
>            Reporter: stack
>            Assignee: stack
>             Fix For: 0.90.0
>
>         Attachments: 2979.txt
>
>
> Its failing w/ a while now.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Resolved: (HBASE-2979) Fix failing TestMultParrallel in hudson build

Posted by "stack (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HBASE-2979?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack resolved HBASE-2979.
--------------------------

    Resolution: Fixed

Committed fix for broke test.

> Fix failing TestMultParrallel in hudson build
> ---------------------------------------------
>
>                 Key: HBASE-2979
>                 URL: https://issues.apache.org/jira/browse/HBASE-2979
>             Project: HBase
>          Issue Type: Bug
>            Reporter: stack
>            Assignee: stack
>             Fix For: 0.90.0
>
>         Attachments: 2979-v2.txt, 2979-v3.txt, 2979.txt
>
>
> Its failing w/ a while now.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Updated: (HBASE-2979) Fix failing TestMultParrallel in hudson build

Posted by "stack (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HBASE-2979?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack updated HBASE-2979:
-------------------------

    Attachment: 2979-v3.txt

Here is what I just committed.  Also adds new CatalogTracker test and a few bug fixes for CatalogTracker+AssignmentManager

{code}

M src/test/java/org/apache/hadoop/hbase/HBaseTestingUtility.java
  Make KEYS available for tests that want to mess with multiregion
  table row keys.
  (ensureSomeRegionServersAreAvailable): Added a boolean return which
  is true if we actually started a server.
D src/test/java/org/apache/hadoop/hbase/TestMultiParallel.java
  Moved into client package.
A src/test/java/org/apache/hadoop/hbase/client/TestMultiParallel.java
  Moved from o.a.h.h package.  Also converted to use juni4 & HTU.
M src/test/java/org/apache/hadoop/hbase/MiniHBaseCluster.java
  Catch regular NPE rather than let it spew over log on exit.
A src/test/java/org/apache/hadoop/hbase/catalog/TestCatalogTracker.java
  Tests for CatalogTracker
M src/main/java/org/apache/hadoop/hbase/zookeeper/ZooKeeperWatcher.java
  Removed commented out code.
M src/main/java/org/apache/hadoop/hbase/zookeeper/RootRegionTracker.java
  Javadoc.
M src/main/java/org/apache/hadoop/hbase/zookeeper/ZooKeeperNodeTracker.java
  (getNode) Return node that is being tracked.
M src/main/java/org/apache/hadoop/hbase/zookeeper/MetaNodeTracker.java
  Made this be a ZKNodeTracker; thats what it is.
M src/main/java/org/apache/hadoop/hbase/regionserver/HRegionServer.java
M src/main/java/org/apache/hadoop/hbase/regionserver/handler/OpenRegionHandler.java
M src/main/java/org/apache/hadoop/hbase/LocalHBaseCluster.java
M src/main/java/org/apache/hadoop/hbase/master/ServerManager.java
  Formatting.
M src/main/java/org/apache/hadoop/hbase/master/AssignmentManager.java
  Added better handling of zk master events (more to do).
  Fixed interesting bug where just presence of a server in assignments,
  though it was empty of regions and a crashed region, was making the
  load balancer give it regions to open (post-crash)
M src/main/java/org/apache/hadoop/hbase/master/HMaster.java
  Don't run balancer if deadservers.  Also log balance plans (for now)
M src/main/java/org/apache/hadoop/hbase/master/LoadBalancer.java
  Add toString to Plan. Helps debugging.
M src/main/java/org/apache/hadoop/hbase/catalog/CatalogTracker.java
  Javadoc.
  Fixup of waitForMeta and waitForRoot. In particular handle
  ConnectException when doing verify of region location.
  Made waitForRoot(timeout) protected (rather than public)
M src/main/java/org/apache/hadoop/hbase/catalog/MetaEditor.java
  Javadoc and shutdown access to a few methods.
M src/main/java/org/apache/hadoop/hbase/ipc/HRegionInterface.java
M src/main/java/org/apache/hadoop/hbase/client/HConnectionManager.java
  Formatting.
M src/main/java/org/apache/hadoop/hbase/client/HBaseAdmin.java
  Better exception  message.
M src/main/java/org/apache/hadoop/hbase/client/ScannerCallable.java
  Added in LOG.
M bin/rolling-restart.sh
  Fix double entry (patch applied twice)
{code}

> Fix failing TestMultParrallel in hudson build
> ---------------------------------------------
>
>                 Key: HBASE-2979
>                 URL: https://issues.apache.org/jira/browse/HBASE-2979
>             Project: HBase
>          Issue Type: Bug
>            Reporter: stack
>            Assignee: stack
>             Fix For: 0.90.0
>
>         Attachments: 2979-v2.txt, 2979-v3.txt, 2979.txt
>
>
> Its failing w/ a while now.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Updated: (HBASE-2979) Fix failing TestMultParrallel in hudson build

Posted by "stack (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HBASE-2979?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack updated HBASE-2979:
-------------------------

    Attachment: 2979-v2.txt

This test is failing because new master setup is not handling abort of server carrying .META. properly.  On opening of a bunch of regions, .META. carrier goes down in the middle causing the OPENED processing to fail because it can't update meta w/ info on just opened region.  So, need to make the RS wait on the meta to come back up.

But tools to do this are not yet in place.  Am writing a test for CatalogTracker first before I do anything else and adding in a few methods; e.g. we need to pass a Progressable to the waitForMeta so we can tickle zk letting it know that opening is still going on (Jon's idea).

> Fix failing TestMultParrallel in hudson build
> ---------------------------------------------
>
>                 Key: HBASE-2979
>                 URL: https://issues.apache.org/jira/browse/HBASE-2979
>             Project: HBase
>          Issue Type: Bug
>            Reporter: stack
>            Assignee: stack
>             Fix For: 0.90.0
>
>         Attachments: 2979-v2.txt, 2979.txt
>
>
> Its failing w/ a while now.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.