[
https://issues.apache.org/jira/browse/HADOOP-12334?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14961387#comment-14961387
]
Gaurav Kanade commented on HADOOP-12334:
----------------------------------------
Thanks [~cnauroth] for the clarification. I looked at the specific class of
tests you mention. These tests seem to be designed to verify that if an invalid
operation is attempted an error is indeed thrown. For e.g.
testTransientErrorOnReadThrowsException, testAccessContainerWithWrongVersion
throws exception and so on. In our scenario I do not believe the goal of the
test is to check whether an exception is indeed raised; what we are doing is
handling the exception thrown with the SERVER_BUSY error code. If anything
needs to be tested it should be that the new copy mechanism copies accurately.
I am not sure how we can use this suite of tests for this kind of verification.
Like I mentioned earlier, we have tested that the copy happens correctly by
manually imposing the mechanism (download first then copy) for rename. Or did
you have something else in mind that you wanted tested ? Let me know your
thoughts
> Change Mode Of Copy Operation of HBase WAL Archiving to bypass Azure Storage
> Throttling after retries
> -----------------------------------------------------------------------------------------------------
>
> Key: HADOOP-12334
> URL: https://issues.apache.org/jira/browse/HADOOP-12334
> Project: Hadoop Common
> Issue Type: Improvement
> Components: tools
> Reporter: Gaurav Kanade
> Assignee: Gaurav Kanade
> Attachments: HADOOP-12334.01.patch, HADOOP-12334.02.patch,
> HADOOP-12334.03.patch, HADOOP-12334.04.patch, HADOOP-12334.05.patch,
> HADOOP-12334.06.patch
>
>
> HADOOP-11693 mitigated the problem of HMaster aborting regionserver due to
> Azure Storage Throttling event during HBase WAL archival. The way this was
> achieved was by applying an intensive exponential retry when throttling
> occurred.
> As a second level of mitigation we will change the mode of copy operation if
> the operation fails even after all retries -i.e. we will do a client side
> copy of the blob and then copy it back to destination. This operation will
> not be subject to throttling and hence should provide a stronger mitigation.
> However it is more expensive, hence we do it only in the case we fail after
> all retries
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
