[
https://issues.apache.org/jira/browse/HDFS-15169?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17070372#comment-17070372
]
Ayush Saxena commented on HDFS-15169:
-------------------------------------
Thanx [~hexiaoqiao] for confirming, I tried tweaking the test myself to debug
and it wasn't working as expected so just wanted to confirm. :)
Couple of comments/doubts for the latest patch :
* Presently in the output, The path name is coming as of destination:
{code:java}
The filesystem under path '/testdirdst' is HEALTHY
{code}
It should be replaced with respective mount path. Same as we do, while throwing
exception containing paths.
* I didn't catch this :
{code:java} * Redirect the request to certain active downstream NameNode if
resolve
* target namespace otherwise redirect the requests to all active
* downstream NameNodes.
{code}
Can you explain what you are trying to do here? Why if the path can not be
resolved we need to send to all namenodes, isn't it like a FNF situation?
Otherwise, this seems outsmarting the RouterAdmin, A client should be able to
access only paths in a particular NS which Router Admin has authorized through
mount points.
* Doubt here too, what would happen in case of a failover :
{code:java}
if (ms.getState() == FederationNamenodeServiceState.ACTIVE
{code}
Will this logic be able to handle that? or is that being handled somewhere else.
* Added a new param in {{getURLArguments}}, Please update the same param in
javadoc of the method too.
* Well, I couldn't check this with multi destination mount points, But by code,
I don't think so, it is handling it any good, Any pointers? If not, Multi
destination stuff we can handle as follow up.
> RBF: Router FSCK should consider the mount table
> ------------------------------------------------
>
> Key: HDFS-15169
> URL: https://issues.apache.org/jira/browse/HDFS-15169
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: rbf
> Reporter: Akira Ajisaka
> Assignee: Xiaoqiao He
> Priority: Major
> Attachments: HDFS-15169.001.patch, HDFS-15169.002.patch,
> HDFS-15169.003.patch, HDFS-15169.004.patch, HDFS-15169.005.patch
>
>
> HDFS-13989 implemented FSCK to DFSRouter, however, it just redirects the
> requests to all the active downstream NameNodes for now. The DFSRouter should
> consider the mount table when redirecting the requests.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
