You are viewing a plain text version of this content. The canonical link for it is here.

Posted to hdfs-dev@hadoop.apache.org by "Shashikant Banerjee (JIRA)" <ji...@apache.org> on 2019/03/26 10:05:00 UTC

[jira] [Created] (HDDS-1336) Better block allocation policy in SCM

Shashikant Banerjee created HDDS-1336:
-----------------------------------------

             Summary: Better block allocation policy in SCM
                 Key: HDDS-1336
                 URL: https://issues.apache.org/jira/browse/HDDS-1336
             Project: Hadoop Distributed Data Store
          Issue Type: Improvement
          Components: SCM
            Reporter: Shashikant Banerjee
            Assignee: Shashikant Banerjee


In case, a client hits close container exception and asks SCM to allocate block for the next write, SCM might allocate blocks from containers which are almost full and can get closed while the client write is still on. Ozone client by default tries 5 times on hitting an exception to write Data to a different block in case it hits CloseContainerException 5 times, it will give up.

The issue was found while testing with HDDS-1295.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-dev-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-dev-help@hadoop.apache.org