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 2010/09/25 01:04:34 UTC

[jira] Created: (HBASE-3037) When new master joins running cluster does "Received report from unknown server -- telling it to STOP_REGIONSERVER..."

When new master joins running cluster does "Received report from unknown server -- telling it to STOP_REGIONSERVER..."
----------------------------------------------------------------------------------------------------------------------

                 Key: HBASE-3037
                 URL: https://issues.apache.org/jira/browse/HBASE-3037
             Project: HBase
          Issue Type: Bug
            Reporter: stack
             Fix For: 0.90.0


Working on it.

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

[jira] Updated: (HBASE-3037) When new master joins running cluster does "Received report from unknown server -- telling it to STOP_REGIONSERVER..."

Posted by "stack (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HBASE-3037?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack updated HBASE-3037:
-------------------------

    Attachment: 3037-v2.txt

This stuff seems to be basically work up on the cluster.  I'm going to go ahead and commit.  Master should be roughly equivalent now to the old (but for all the extras new master brings). 

{code}
M src/main/java/org/apache/hadoop/hbase/master/ServerManager.java
 On regionServerReport, if we get a report from an 'unknown' regionserver,
 we used to tell it stop itself.  Now, if 'unknown' server AND
 this master did not start the cluster, its joining the cluster, then
 treat the report as a regionServerStart and register the incoming
 server rather than tell it shutdown.
M src/main/java/org/apache/hadoop/hbase/master/HMaster.java
 Pass the freshClusterStartup flag to ServerManager.
 Add more executors for opening and closing.  On cluster startup a
 master shouldn't be bottleneck clearning the server opens.
 Expose the run-balancer method so can make it available in
 ServeAdmin.
M src/main/java/org/apache/hadoop/hbase/master/AssignmentManager.java
 Minor formatting and javadoc
M src/main/java/org/apache/hadoop/hbase/ipc/HBaseRPCProtocolVersion.java
 Upped rpc version number because of new balancer addition (and because
 we didn't do it when we put in new master).
M src/main/java/org/apache/hadoop/hbase/ipc/HMasterInterface.java
 Added balancer method.
M src/main/resources/hbase-default.xml
 Change how ofter we check in from every 3 seconds to every 5 seconds.
{code}

> When new master joins running cluster does "Received report from unknown server -- telling it to STOP_REGIONSERVER..."
> ----------------------------------------------------------------------------------------------------------------------
>
>                 Key: HBASE-3037
>                 URL: https://issues.apache.org/jira/browse/HBASE-3037
>             Project: HBase
>          Issue Type: Bug
>            Reporter: stack
>             Fix For: 0.90.0
>
>         Attachments: 3037-v2.txt, 3037.txt
>
>
> Working on it.

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

[jira] Resolved: (HBASE-3037) When new master joins running cluster does "Received report from unknown server -- telling it to STOP_REGIONSERVER..."

Posted by "stack (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HBASE-3037?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack resolved HBASE-3037.
--------------------------

      Assignee: stack
    Resolution: Fixed

Committed.

> When new master joins running cluster does "Received report from unknown server -- telling it to STOP_REGIONSERVER..."
> ----------------------------------------------------------------------------------------------------------------------
>
>                 Key: HBASE-3037
>                 URL: https://issues.apache.org/jira/browse/HBASE-3037
>             Project: HBase
>          Issue Type: Bug
>            Reporter: stack
>            Assignee: stack
>             Fix For: 0.90.0
>
>         Attachments: 3037-v2.txt, 3037.txt
>
>
> Working on it.

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

[jira] Updated: (HBASE-3037) When new master joins running cluster does "Received report from unknown server -- telling it to STOP_REGIONSERVER..."

Posted by "stack (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HBASE-3037?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack updated HBASE-3037:
-------------------------

    Attachment: 3037.txt

This patch seems to basically work.  Testing it more.

> When new master joins running cluster does "Received report from unknown server -- telling it to STOP_REGIONSERVER..."
> ----------------------------------------------------------------------------------------------------------------------
>
>                 Key: HBASE-3037
>                 URL: https://issues.apache.org/jira/browse/HBASE-3037
>             Project: HBase
>          Issue Type: Bug
>            Reporter: stack
>             Fix For: 0.90.0
>
>         Attachments: 3037.txt
>
>
> Working on it.

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