[
https://issues.apache.org/jira/browse/HDFS-5053?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13763529#comment-13763529
]
Colin Patrick McCabe commented on HDFS-5053:
--------------------------------------------
Looks good overall. My first impressions:
{{Fallible#get}}: I think we should continue to wrap the exception here, so
that stack traces of unit test failures look reasonable (you can see where in
the test the failure happened). There might be a few places in the shell where
we're failing to unwrap, but that's a different issue...
config options: I would prefer something like
{{dfs.namenode.path.cache.enabled}}. The CacheManager class is an
implementation detail, so {{dfs.namenode.cachemanager.enabled}} is not the best
name.
{code}
default:
throw new AssertionError("Unknown action type: " + blkCmd.getAction());
}
{code}
I don't think this is needed, since you covered all possible values of the
enum, and in java that's enough.
We definitely need to add the caching factor to {{PathCacheDirective}} so that
people can request how many DataNodes they want. Maybe in a follow-up change.
I guess we need some association between {{BlockCollection}} and
{{PathCacheEntry}]... would avoid the need for the cacheReplication field as
well, since that should be in PCE....
> NameNode should invoke DataNode APIs to coordinate caching
> ----------------------------------------------------------
>
> Key: HDFS-5053
> URL: https://issues.apache.org/jira/browse/HDFS-5053
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: datanode, namenode
> Reporter: Colin Patrick McCabe
> Assignee: Andrew Wang
> Attachments: hdfs-5053-1.patch
>
>
> The NameNode should invoke the DataNode APIs to coordinate caching.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
