[
https://issues.apache.org/jira/browse/HDFS-7208?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ming Ma updated HDFS-7208:
--------------------------
Attachment: HDFS-7208.patch
Here is the initial patch based on heartbeat notification approach, the
assumption is DN will report all healthy storages in the heartbeat. This
approach is simpler than the blockReport approach which needs to have DN
persist the info to cover some failure scenarios. It also makes storage failure
detection faster.
1. NN detects failed storages during HB processing based on the delta between
DN's reported healthy storages and the storages NN has. Marked the state of
those missing storages DatanodeStorage.State.FAILED.
2. HeartbeatManager will remove blocks on those DatanodeStorage.State.FAILED
storages. This will cover some corner scenarios where new replicas might be
added to BlocksMap afterwards.
3. It also covers the case where admins reduce the number of healthy volumes on
DN and restart DN.
> NN doesn't schedule replication when a DN storage fails
> -------------------------------------------------------
>
> Key: HDFS-7208
> URL: https://issues.apache.org/jira/browse/HDFS-7208
> Project: Hadoop HDFS
> Issue Type: Bug
> Reporter: Ming Ma
> Attachments: HDFS-7208.patch
>
>
> We found the following problem. When a storage device on a DN fails, NN
> continues to believe replicas of those blocks on that storage are valid and
> doesn't schedule replication.
> A DN has 12 storage disks. So there is one blockReport for each storage. When
> a disk fails, # of blockReport from that DN is reduced from 12 to 11. Given
> dfs.datanode.failed.volumes.tolerated is configured to be > 0, NN still
> considers that DN healthy.
> 1. A disk failed. All blocks of that disk are removed from DN dataset.
>
> {noformat}
> 2014-10-04 02:11:12,626 WARN
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl: Removing
> replica BP-1748500278-xx.xx.xx.xxx-1377803467793:1121568886 on failed volume
> /data/disk6/dfs/current
> {noformat}
> 2. NN receives DatanodeProtocol.DISK_ERROR. But that isn't enough to have NN
> remove the DN and the replicas from the BlocksMap. In addition, blockReport
> doesn't provide the diff given that is done per storage.
> {noformat}
> 2014-10-04 02:11:12,681 WARN org.apache.hadoop.hdfs.server.namenode.NameNode:
> Disk error on DatanodeRegistration(xx.xx.xx.xxx,
> datanodeUuid=f3b8a30b-e715-40d6-8348-3c766f9ba9ab, infoPort=50075,
> ipcPort=50020,
> storageInfo=lv=-55;cid=CID-e3c38355-fde5-4e3a-b7ce-edacebdfa7a1;nsid=420527250;c=1410283484939):
> DataNode failed volumes:/data/disk6/dfs/current
> {noformat}
> 3. Run fsck on the file and confirm the NN's BlocksMap still has that replica.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
