[
https://issues.apache.org/jira/browse/HDFS-12756?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16240100#comment-16240100
]
Mukul Kumar Singh commented on HDFS-12756:
------------------------------------------
Thanks for the updated patch, the patch looks good to me.
1) I also feel that ContainerCommandRequestProto#datanodeID should be an
optional field. Right now the pipeline info also sent for all the
DatanodeContainerCommands which is not required. As discussed in HDFS-12745,
Pipline information should be removed from DataNodeContainer commands. I agree
that this will be required for a Simulated Mini Ozone Cluster, but not sure if
this should be enforced in real scenarios.
Nitpicks:
1) ContainerTestHelper.java:44 unused import
2) Should we replace "pipeline.getLeader().getDatanodeUuid()" with something
like "pipeline.getLeaderDatanodeUUid"
> Ozone: Add datanodeID to heartbeat responses and container protocol
> -------------------------------------------------------------------
>
> Key: HDFS-12756
> URL: https://issues.apache.org/jira/browse/HDFS-12756
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Anu Engineer
> Assignee: Anu Engineer
> Priority: Trivial
> Attachments: HDFS-12756-HDFS-7240.001.patch,
> HDFS-12756-HDFS-7240.002.patch, HDFS-12756-HDFS-7240.003.patch
>
>
> if we have datanode ID in the HBs responses and commands send to datanode, we
> will be able to do additional sanity checking on datanode before executing
> the command. This is also very helpful in creating a MiniOzoneCluster with
> 1000s of simulated nodes. This is needed for scale based unit tests of SCM.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
