[
https://issues.apache.org/jira/browse/HDFS-1595?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12986788#action_12986788
]
Koji Noguchi commented on HDFS-1595:
------------------------------------
bq. So this faulty node F has no problem receiving large amount of data?
This faulty node had problem sending/receiving large amount of data and failing
most of the time.
Bigger the data, higher the chances of the failures. I think smaller data
(,say less than 1MB) was going through 99% of the time.
So heartbeat, ack and so forth were probably working.
When I tried to scp some blocks out from this node for data recovery, it kept
on failing with
===========================
blk_-113193561174013799 0%
0 0.0KB/s --:-- ETA
Corrupted MAC on input.
Finished discarding for aa.bb.cc.dd
lost connection
===========================
So I believe *most* of the dfsclient write was failing when going through this
node.
And when it successfully went through (after hundreds of write attempts for
different blocks), it would then fail on all the following replications but
succeed on 'close' with 1 replica leading to this bug.
> DFSClient may incorrectly detect datanode failure
> -------------------------------------------------
>
> Key: HDFS-1595
> URL: https://issues.apache.org/jira/browse/HDFS-1595
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: data-node, hdfs client
> Affects Versions: 0.20.4
> Reporter: Tsz Wo (Nicholas), SZE
> Priority: Critical
> Attachments: hdfs-1595-idea.txt
>
>
> Suppose a source datanode S is writing to a destination datanode D in a write
> pipeline. We have an implicit assumption that _if S catches an exception
> when it is writing to D, then D is faulty and S is fine._ As a result,
> DFSClient will take out D from the pipeline, reconstruct the write pipeline
> with the remaining datanodes and then continue writing .
> However, we find a case that the faulty machine F is indeed S but not D. In
> the case we found, F has a faulty network interface (or a faulty switch port)
> in such a way that the faulty network interface works fine when sending out a
> small amount of data, say 1MB, but it fails when sending out a large amount
> of data, say 100MB. Reading is working fine for any data size.
> It is even worst if F is the first datanode in the pipeline. Consider the
> following:
> # DFSClient creates a pipeline with three datanodes. The first datanode is F.
> # F catches an IOException when writing to the second datanode. Then, F
> reports the second datanode has error.
> # DFSClient removes the second datanode from the pipeline and continue
> writing with the remaining datanode(s).
> # The pipeline now has two datanodes but (2) and (3) repeat.
> # Now, only F remains in the pipeline. DFSClient continues writing with one
> replica in F.
> # The write succeeds and DFSClient is able to *close the file successfully*.
> # The block is under replicated. The NameNode schedules replication from F
> to some other datanode D.
> # The replication fails for the same reason. D reports to the NameNode that
> the replica in F is corrupted.
> # The NameNode marks the replica in F is corrupted.
> # The block is corrupted since no replica is available.
> This is a *data loss* scenario.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
