[
https://issues.apache.org/jira/browse/HDFS-11856?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16033830#comment-16033830
]
John Zhuge commented on HDFS-11856:
-----------------------------------
[~vinayrpet] and [~kihwal] Will this patch help clusters with more than 3 DNs?
We saw HBase RegionServer occasional crashing with DamagedWALException after
the following pipeline recovery failure:
{noformat}
java.io.IOException: All datanodes
DatanodeInfoWithStorage[x.x.x.x:20002,DS-uuid,DISK] are bad. Aborting...
at
org.apache.hadoop.hdfs.DFSOutputStream$DataStreamer.setupPipelineForAppendOrRecovery(DFSOutputStream.java:1465)
at
org.apache.hadoop.hdfs.DFSOutputStream$DataStreamer.processDatanodeError(DFSOutputStream.java:1236)
at
org.apache.hadoop.hdfs.DFSOutputStream$DataStreamer.run(DFSOutputStream.java:721)
{noformat}
> Ability to re-add Upgrading Nodes (remote) to pipeline for future pipeline
> updates
> ----------------------------------------------------------------------------------
>
> Key: HDFS-11856
> URL: https://issues.apache.org/jira/browse/HDFS-11856
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: hdfs-client, rolling upgrades
> Affects Versions: 2.7.3
> Reporter: Vinayakumar B
> Assignee: Vinayakumar B
> Fix For: 2.9.0, 2.7.4, 3.0.0-alpha4, 2.8.2
>
> Attachments: HDFS-11856-01.patch, HDFS-11856-02.branch-2.patch,
> HDFS-11856-02.patch, HDFS-11856-branch-2-02.patch,
> HDFS-11856-branch-2.7-02.patch, HDFS-11856-branch-2.8-02.patch
>
>
> During rolling upgrade if the DN gets restarted, then it will send special
> OOB_RESTART status to all streams opened for write.
> 1. Local clients will wait for 30 seconds to datanode to come back.
> 2. Remote clients will consider these nodes as bad nodes and continue with
> pipeline recoveries and write. These restarted nodes will be considered as
> bad, and will be excluded for lifetime of stream.
> In case of small cluster, where total nodes itself is 3, each time a remote
> node restarts for upgrade, it will be excluded.
> So a stream writing to 3 nodes initial, will end-up writing to only one node
> at the end, there are no other nodes to replace.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
