hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Vinayakumar B (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-7342) Lease Recovery doesn't happen some times
Date Mon, 24 Nov 2014 06:02:12 GMT

    [ https://issues.apache.org/jira/browse/HDFS-7342?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14222707#comment-14222707
] 

Vinayakumar B commented on HDFS-7342:
-------------------------------------

I got the reason, why [~raviprak] is facing the problem and I am not. 

You have mentioned affected version as 2.0.0-alpha, which came to help.
The possibility of making the last block COMPLETE when the penultimate block still in COMMITTED
state was before the fix of HDFS-5558 which was fixed in 2.3.0 version, though the problem
faced at that time was not infinite loop, but crash of the lease monitor thread.

After the fix, last block cannot be in COMPLETE state when others blocks are not COMPLETED,
and infinite loop never occurs.

I hope this clears the confusion. And I think, there is no change required in this Jira. In
that case, can this be closed as Duplicate?

Hi  [~yzhangal], any thoughts?

> Lease Recovery doesn't happen some times
> ----------------------------------------
>
>                 Key: HDFS-7342
>                 URL: https://issues.apache.org/jira/browse/HDFS-7342
>             Project: Hadoop HDFS
>          Issue Type: Bug
>    Affects Versions: 2.0.0-alpha
>            Reporter: Ravi Prakash
>            Assignee: Ravi Prakash
>         Attachments: HDFS-7342.1.patch, HDFS-7342.2.patch
>
>
> In some cases, LeaseManager tries to recover a lease, but is not able to. HDFS-4882 describes
a possibility of that. We should fix this



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message