hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Anu Engineer (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-11444) Ozone: Fix datanode ID handling in MiniOzoneCluster
Date Fri, 17 Mar 2017 17:01:41 GMT

    [ https://issues.apache.org/jira/browse/HDFS-11444?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15930302#comment-15930302
] 

Anu Engineer commented on HDFS-11444:
-------------------------------------

[~cheersyang] The changes look excellent. Thanks for solving this issue.

One question, I am just asking it for my own understanding and not asking you to 
change the patch.

Is it possible to update the datanode ID in the initDatanodeState.java instead of 
datanode.java, that only thought was then all changes of ozone would be self-contained 
in the datanode side. I was curious to understand if you considered this path and if so
what issues you found.

Two small nits: +1 after that.

1. Nit: DatanodeID.java: redundant comment. Please consider removing it.
{{// Consider IDs are equal only when the protobuf messages are same.}}

2. Check can path.delete amd createNewFile fail ? Function : writeDatanodeIDs
{code}
writeDatanodeIDs(Set<DatanodeID> ids,
      File path) throws IOException {
    if (path.exists()) {
      path.delete();
      path.createNewFile();
{code}

> Ozone: Fix datanode ID handling in MiniOzoneCluster
> ---------------------------------------------------
>
>                 Key: HDFS-11444
>                 URL: https://issues.apache.org/jira/browse/HDFS-11444
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: ozone
>    Affects Versions: HDFS-7240
>            Reporter: Anu Engineer
>            Assignee: Weiwei Yang
>            Priority: Minor
>             Fix For: HDFS-7240
>
>         Attachments: HDFS-11444.001.patch, HDFS-11444-HDFS-7240.001.patch, HDFS-11444-HDFS-7240.002.patch
>
>
> SCM protocol identifies datanodes via datanode ID. This datanode ID is stored in a path
specified in the OzoneConfig. In MiniOzoneCluster, we need to either send different paths
to nodes or support the ability to have multiple nodes share the same path.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org


Mime
View raw message