[
https://issues.apache.org/jira/browse/HDFS-397?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12731092#action_12731092
]
Raghu Angadi commented on HDFS-397:
-----------------------------------
> [...] What's your take on this?
This is expected to be only a minor reorg of code. But it is not, it does not
matter weather those unintended changes are correct or not. btw, the bug would
show up during upgrade I guess.
If the patch made sure it didn't change any semantics, either of us need not go
through or completely follow what backup namenode does.
I think a clear and simple bug in the patch is taking a lot of effort to
resolve.
It is very unrealistic to expect tests to catch every random bug and typo in a
patch (for anything I have ever worked on or known.)
I guess my main point is : there is no change in semantics expected and lets
make sure that is is the case. If there is a functional change, please make
that explicit.
Thanks for the references to patch in HDFS-311.. will take a look.
> Incorporate storage directories into EditLogFileInput/Output streams
> --------------------------------------------------------------------
>
> Key: HDFS-397
> URL: https://issues.apache.org/jira/browse/HDFS-397
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Luca Telloli
> Assignee: Luca Telloli
> Attachments: HADOOP-6001.patch, HADOOP-6001.patch, HADOOP-6001.patch,
> HDFS-397.patch
>
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
