[
https://issues.apache.org/jira/browse/HDFS-15112?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17016499#comment-17016499
]
Ayush Saxena commented on HDFS-15112:
-------------------------------------
Thanx [~elgoiri] for the update. The overall patch LGTM
{code:java}
catch (IOException ioe) {
if (RouterRpcClient.isUnavailableException(ioe)) {
LOG.debug("Ignore unavailable exception: {}", ioe);
} else {
throw ioe;
}
}
{code}
But I think, this we shouldn't do here, may be we can discuss this in the
follow up where we handle invoke concurrent, I have hard feeling adding this.
That test failure was a bug and surfaced a genuine issue, I guess there is no
test like, if there is a non {{PathAll}} entry and a subcluster is down, write
should fail. The test {{testWriteWithFailedSubcluster}} checks in case the
mount entry is not fault tolerant, some only should fail, ideally all should
fail, that is why this test is passing, only in case if the entry is fault
tolerant, the write should be success, if there is a non available cluster.
This part as we decided, we can handle in another JIRA, but in that we have to
remove this catch block only.
other than that, v009 LGTM +1
> RBF: Do not return FileNotFoundException when a subcluster is unavailable
> --------------------------------------------------------------------------
>
> Key: HDFS-15112
> URL: https://issues.apache.org/jira/browse/HDFS-15112
> Project: Hadoop HDFS
> Issue Type: Improvement
> Reporter: Íñigo Goiri
> Assignee: Íñigo Goiri
> Priority: Major
> Attachments: HDFS-15112.000.patch, HDFS-15112.001.patch,
> HDFS-15112.002.patch, HDFS-15112.004.patch, HDFS-15112.005.patch,
> HDFS-15112.006.patch, HDFS-15112.007.patch, HDFS-15112.008.patch,
> HDFS-15112.009.patch, HDFS-15112.patch
>
>
> If we have a mount point using HASH_ALL across two subclusters and one of
> them is down, we may return FileNotFoundException while the file is just in
> the unavailable subcluster.
> We should not return FileNotFoundException but something that shows that the
> subcluster is unavailable.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
