hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Andrew Purtell (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HBASE-1117) Region assignment on startup is broken
Date Sat, 10 Jan 2009 23:23:59 GMT

    [ https://issues.apache.org/jira/browse/HBASE-1117?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12662703#action_12662703
] 

Andrew Purtell commented on HBASE-1117:
---------------------------------------

This is not a symptom of HBASE-1119.

Regions stay in transition states, I think because the master is not allowing opens from HRS
it thinks are overloaded, yet the HRS are insistent to open. HRS leases time out. META goes
offline. 

What is going on here? 


> Region assignment on startup is broken
> --------------------------------------
>
>                 Key: HBASE-1117
>                 URL: https://issues.apache.org/jira/browse/HBASE-1117
>             Project: Hadoop HBase
>          Issue Type: Bug
>         Environment: apurtell cluster, HBase TRUNK on Hadoop 0.19.1-dev
>            Reporter: Andrew Purtell
>            Assignee: Andrew Purtell
>            Priority: Blocker
>             Fix For: 0.19.0
>
>         Attachments: hbase-hadoop-wtp.back.cluster.HMaster-sjdc-atr-dc-2.log, hbase-hadoop-wtp.back.cluster.HMaster-sjdc-atr-dc-2.log
>
>
> I can't get my cluster all the way up after upgrading to Hadoop 0.19. Master is ignoring
MSG_REPORT_PROCESS_OPEN from various regionservers, skipping regions it thinks are in transition.
Master UI is also unresponsive. 

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


Mime
View raw message