[jira] Commented: (HADOOP-5780) Fix slightly confusing log from "-metaSave" on NameNode

["Suresh Srinivas (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HADOOP-5780?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12706690#action_12706690 ] 

Suresh Srinivas commented on HADOOP-5780:
-----------------------------------------

"decommisioned" has a typo. It should be "decommissioned".

Also not covered in the description - the patch includes additional useful information that identifies a block on a node as "corrupt" or "decommissioned".

> Fix slightly confusing log from "-metaSave" on NameNode
> -------------------------------------------------------
>
>                 Key: HADOOP-5780
>                 URL: https://issues.apache.org/jira/browse/HADOOP-5780
>             Project: Hadoop Core
>          Issue Type: Bug
>    Affects Versions: 0.20.0
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>            Priority: Minor
>             Fix For: 0.21.0
>
>         Attachments: HADOOP-5780.patch, HADOOP-5780.patch
>
>
> After HADOOP-4103, "-metaSave" on NameNode prints "MISSING" in the line for a block that does not have any good replicas left. 
> e.g. : {noformat}
> Metasave: Blocks waiting for replication: 1
> blk_6735500019364591152_0 (replicas: l: 0 d: 0 c: 2 e: 0 MISSING) 74.6.132.200:50010 :  192.168.0.107:50010 :
> {noformat}
> "0 MISSING" in the log might imply to the user that there is no missing block, though it implies "e(xcess replicas) : 0".
> We should fix that. Also as part of this I would like to indicate which datanodes are decommissioned or contain the corrupt replicas.

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