[
https://issues.apache.org/jira/browse/HDFS-4849?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13687017#comment-13687017
]
Suresh Srinivas commented on HDFS-4849:
---------------------------------------
I will run the test when I get some time today.
bq. I can wait as long as this change is in 2.1.
I do not think this is a 2.1 blocker. All the changes proposed in this patch
can be done without breaking backward compatibility. Lets not block a release
for a feature that you want to get in.
bq. What is your definition of an idempotent operation? How is it different
from "just allowing retry"?
See http://en.wikipedia.org/wiki/Idempotence. Just allowing retry is not
idempotent. For example, if you build retry cache and allow retrying
operations, it only works for retransmitted requests by sending the previous
response. A true idempotent operation should work irrespective of whether a
request is a retry at IPC level or at the application level or from different
clients. That is true for other operations that are marked as idempotent.
> Idempotent create and append operations.
> ----------------------------------------
>
> Key: HDFS-4849
> URL: https://issues.apache.org/jira/browse/HDFS-4849
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: namenode
> Affects Versions: 2.0.4-alpha
> Reporter: Konstantin Shvachko
> Assignee: Konstantin Shvachko
> Priority: Blocker
> Attachments: idempotentCreate.patch, idempotentCreate.patch,
> idempotentCreate.patch
>
>
> create, append and delete operations can be made idempotent. This will reduce
> chances for a job or other app failures when NN fails over.
--
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
