[
https://issues.apache.org/jira/browse/HADOOP-1349?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12495875
]
dhruba borthakur commented on HADOOP-1349:
------------------------------------------
I will look at the code more closely, but the approach sounds pretty good. Like
we discussed, this approach still cannot solve the race condition entirely. In
this approach, we ensure that the namenode has sent out the delete-block
request before attempting to allocate the same block on the same datanode, but
these requests could still get *processed* on the datanode out-of-order. This
fix reduces the race-window to a minimum.
The other side-effect is that a client than opens the file might try a bad
block replica for a longer time (because the block does not get deleted from
the blocksmap for a longer time) but this irritant should be minor at best and
can be ignored.
> Corrupted blocks get deleted but not replicated
> -----------------------------------------------
>
> Key: HADOOP-1349
> URL: https://issues.apache.org/jira/browse/HADOOP-1349
> Project: Hadoop
> Issue Type: Bug
> Components: dfs
> Reporter: Hairong Kuang
> Assigned To: Hairong Kuang
> Fix For: 0.14.0
>
> Attachments: blockInvalidate.patch
>
>
> When I test the patch to HADOOP-1345 on a two node dfs cluster, I see that
> dfs correctly delete the corrupted replica and successfully retry reading
> from the other correct replica, but the block does not get replicated. The
> block remains with only 1 replica until the next block report comes in.
> In my testcase, since the dfs cluster has only 2 datanodes, the target of
> replication is the same as the target of block invalidation. After poking
> the logs, I found out that the namenode sent the replication request before
> the block invalidation request.
> This is because the namenode does not invalidate a block well. In
> FSNamesystem.invalidateBlock, it first puts the invalidate request in a queue
> and then immediately removes the replica from its state, which triggers the
> choosing a target for the block. When requests are sent back to the target
> datanode as a reply to a heartbeat message, the replication requests have
> higher priority than the invalidate requests.
> This problem could be solved if a namenode removes an invalidated replica
> from its state only after the invalidate request is sent to the datanode.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
