hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Rushabh S Shah (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-8872) Reporting of missing blocks is different in fsck and namenode ui/metasave
Date Mon, 23 May 2016 19:08:12 GMT

    [ https://issues.apache.org/jira/browse/HDFS-8872?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15296892#comment-15296892
] 

Rushabh S Shah commented on HDFS-8872:
--------------------------------------

Thanks [~mingma] for mentioning HDFS-7933.
bq. For this scenario, it is debatable if the block should be marked as missing, it isn't
uncommon for admins to decommission multiple nodes across racks, which means all 3 replica
nodes will be in decommissioning state.
I agree the block should not be marked as missing if all the replica are on DecommissionING
nodes.
But we *should* mark the block as missing if all the replicas are on DecommissionED nodes
since we can take the Decommissioned node out of rotation anytime.
We have seen multiple cases in which all the replicas are in Decommissioned nodes.
Any thoughts ?

> Reporting of missing blocks is different in fsck and namenode ui/metasave
> -------------------------------------------------------------------------
>
>                 Key: HDFS-8872
>                 URL: https://issues.apache.org/jira/browse/HDFS-8872
>             Project: Hadoop HDFS
>          Issue Type: Bug
>    Affects Versions: 2.6.0
>            Reporter: Rushabh S Shah
>            Assignee: Rushabh S Shah
>
> Namenode ui and metasave will not report a block as missing if the only replica is on
decommissioning/decomissioned node while fsck will show it as MISSING.
> Since decommissioned node can be formatted/removed anytime, we can actually lose the
block.
> Its better to alert on namenode ui if the only copy is on decomissioned/decommissioning
node.



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

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


Mime
View raw message