[
https://issues.apache.org/jira/browse/HADOOP-15702?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16621852#comment-16621852
]
Steve Loughran commented on HADOOP-15702:
-----------------------------------------
let me worry about this for 3.3; I clearly have the most interest in this as my
network is slowest
what I'd like to see is
* configurable timeout (test property)
* configurable max size (test property)
* using the nanotimer to measure & print time effective bandwidth of writes,
reads, seeks. If tests are going to take time, the least we can do is print out
where that time is going
FWIW, I won't just say "run really small tests"; sometimes it is good to leave
an upload running for 2h to see what happens...its how I found out that the AWS
SDK wasn't retrying on failures of the final complete-multi-part-upload
request, as well as how it (sometimes/still?) gets mark & restore wrong on
block writes. Those are test runs which I can kick off and walk away from.
> ABFS: Increase timeout of ITestAbfsReadWriteAndSeek
> ---------------------------------------------------
>
> Key: HADOOP-15702
> URL: https://issues.apache.org/jira/browse/HADOOP-15702
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: fs/azure, test
> Affects Versions: HADOOP-15407
> Reporter: Sean Mackrory
> Assignee: Da Zhou
> Priority: Major
> Fix For: HADOOP-15407
>
>
> ITestAbfsReadWriteAndSeek.testReadAndWriteWithDifferentBufferSizesAndSeek
> fails for me all the time. Let's increase the timout limit.
> It also seems to get executed twice...
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
