[
https://issues.apache.org/jira/browse/HADOOP-4556?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12644431#action_12644431
]
Raghu Angadi commented on HADOOP-4556:
--------------------------------------
The "main cause" (among others) seems to be because of incomplete clean up when
a datanode is marked dead.
Specifically, the pending requests for block deletions don't seem to be removed
when a node is marked dead.. These deletion requests might be sent after the
datanode registers back again... but the state might be very different that
time.
> Block went missing
> ------------------
>
> Key: HADOOP-4556
> URL: https://issues.apache.org/jira/browse/HADOOP-4556
> Project: Hadoop Core
> Issue Type: Bug
> Components: dfs
> Affects Versions: 0.17.2
> Reporter: Robert Chansler
> Assignee: Hairong Kuang
>
> Suspicion that all replicas of a block were marked for deletion. (Don't
> panic, investigation underway.)
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
