[
https://issues.apache.org/jira/browse/HDFS-15644?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17219356#comment-17219356
]
Ahmed Hussein commented on HDFS-15644:
--------------------------------------
Sure! Moving forward, I will do PRs.
> Failed volumes can cause DNs to stop block reporting
> ----------------------------------------------------
>
> Key: HDFS-15644
> URL: https://issues.apache.org/jira/browse/HDFS-15644
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: block placement, datanode
> Reporter: Ahmed Hussein
> Assignee: Ahmed Hussein
> Priority: Major
> Labels: refactor
> Attachments: HDFS-15644.001.patch, HDFS-15644.002.patch
>
>
> [~daryn] found a corner case where remove failed volumes can cause a NPE in
> [FsDataSetImpl.getBlockReports()|https://github.com/apache/hadoop/blob/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/fsdataset/impl/FsDatasetImpl.java#L1939].
> +Scenario:+
> * Inside {{Datanode#HandleVolumeFailures()}}, removing a failed volume is a
> 2-step process.
> ** First it's removed from from the volumes list
> ** Later in time are the replicas scrubbed from the volume map
> * A concurrent thread generating blockReports may access the replicaMap
> accessing a non existing VolumeID.
> He made a fix for that and we have been using it on our clusters since
> Hadoop-2.7.
> By analyzing the code, the bug is still applicable to Trunk.
> * The path Datanode#removeVolumes() is safe because the two step process in
> {{FsDataImpl.removeVolumes()}}
> [FsDatasetImpl.java#L577|https://github.com/apache/hadoop/blob/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/fsdataset/impl/FsDatasetImpl.java#L577]
> is protected by {{datasetWriteLock}} .
> * The path Datanode#handleVolumeFailures() is not safe because the failed
> volume is removed from the list without acquiring
> {{datasetWriteLock}}.[FsVolumList#239|https://github.com/apache/hadoop/blob/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/fsdataset/impl/FsVolumeList.java#L239]
> The race condition can cause the caller of getBlockReports() to throw NPE if
> the RUR is referring to a volume that has already been removed
> [FsDatasetImpl.java#L1976|https://github.com/apache/hadoop/blob/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/fsdataset/impl/FsDatasetImpl.java#L1976].
> {code:java}
> case RUR:
> ReplicaInfo orig = b.getOriginalReplica();
> builders.get(volStorageID).add(orig);
> break;
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
