[
https://issues.apache.org/jira/browse/HDFS-14316?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16797700#comment-16797700
]
Íñigo Goiri commented on HDFS-14316:
------------------------------------
I uploaded [^HDFS-14316-HDFS-13891.009.patch] with some of the changes:
* I added some text about the inconsistencies and being careful. This should
cover the files concern.
* I made the {{StringBuilder}} in a single call. BTW, can you share a link with
this PDM warning?
* Rebased on top of HDFS-1435.
Then things that I had existential questions about:
* For the loop where it does not remove any location, I think the cost is not
that high for this extra loop and leaves a better framework for later. In
addition, there is the chance that getting the create location works but then
the creation itself fails. In this case, it would make sense to remove it.
* For the directories is where I have the most doubts. What is the concern
exactly? The problem would be to create a directory which would not be present
in the other subcluster. Chmod and other operations would fail, right?
* Everything is in TestRouterAllResolver, maybe I can make it a separate class
like TestRouterFaultTolerant.
> RBF: Support unavailable subclusters for mount points with multiple
> destinations
> --------------------------------------------------------------------------------
>
> Key: HDFS-14316
> URL: https://issues.apache.org/jira/browse/HDFS-14316
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Íñigo Goiri
> Assignee: Íñigo Goiri
> Priority: Major
> Attachments: HDFS-14316-HDFS-13891.000.patch,
> HDFS-14316-HDFS-13891.001.patch, HDFS-14316-HDFS-13891.002.patch,
> HDFS-14316-HDFS-13891.003.patch, HDFS-14316-HDFS-13891.004.patch,
> HDFS-14316-HDFS-13891.005.patch, HDFS-14316-HDFS-13891.006.patch,
> HDFS-14316-HDFS-13891.007.patch, HDFS-14316-HDFS-13891.008.patch,
> HDFS-14316-HDFS-13891.009.patch
>
>
> Currently mount points with multiple destinations (e.g., HASH_ALL) fail
> writes when the destination subcluster is down. We need an option to allow
> writing in other subclusters when one is down.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
