hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sanjay Radia (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-3703) Decrease the datanode failure detection time
Date Thu, 26 Jul 2012 20:25:34 GMT

    [ https://issues.apache.org/jira/browse/HDFS-3703?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13423428#comment-13423428
] 

Sanjay Radia commented on HDFS-3703:
------------------------------------

HDFS-3705 and HDFS-3706 are client based.   Suresh's suggestion (the T parameter) is server
based.
If the server based approach is not aggressive enough for some applications then we can consider
augmenting with a client-based approach that passes in a parameter T for getBlocks and addBlock
operations. The advantage of this alternate client-based approach over the original client-based
approach proposed in HDFS-3705 and HDFS-3706 is that it uses the NN's view of which DNs are
running, stale, or dead rather then the client's view of which nodes are running or dead.

                
> Decrease the datanode failure detection time
> --------------------------------------------
>
>                 Key: HDFS-3703
>                 URL: https://issues.apache.org/jira/browse/HDFS-3703
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: data-node, name-node
>    Affects Versions: 1.0.3, 2.0.0-alpha
>            Reporter: nkeywal
>            Assignee: Suresh Srinivas
>
> By default, if a box dies, the datanode will be marked as dead by the namenode after
10:30 minutes. In the meantime, this datanode will still be proposed  by the nanenode to write
blocks or to read replicas. It happens as well if the datanode crashes: there is no shutdown
hooks to tell the nanemode we're not there anymore.
> It especially an issue with HBase. HBase regionserver timeout for production is often
30s. So with these configs, when a box dies HBase starts to recover after 30s and, while 10
minutes, the namenode will consider the blocks on the same box as available. Beyond the write
errors, this will trigger a lot of missed reads:
> - during the recovery, HBase needs to read the blocks used on the dead box (the ones
in the 'HBase Write-Ahead-Log')
> - after the recovery, reading these data blocks (the 'HBase region') will fail 33% of
the time with the default number of replica, slowering the data access, especially when the
errors are socket timeout (i.e. around 60s most of the time). 
> Globally, it would be ideal if HDFS settings could be under HBase settings. 
> As a side note, HBase relies on ZooKeeper to detect regionservers issues.

--
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

        

Mime
View raw message