[
https://issues.apache.org/jira/browse/HDFS-9329?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14983124#comment-14983124
]
Xiao Chen commented on HDFS-9329:
---------------------------------
Hi [~zhz], thanks for fixing this! It makes sense and I'm mostly reading
through to learn. Good job explaining the test mechanism and using
AtomicBoolean. +1 (non-binding).
A super minor thing:
May be we can add a timeout to the test? I can't see from code that it could
time out, but adding one feels safer since we have a while loop here, just in
case other things break.. It's ok if you think this is unnecessary....
> TestBootstrapStandby#testRateThrottling is flaky because fsimage size is
> smaller than IO buffer size
> ----------------------------------------------------------------------------------------------------
>
> Key: HDFS-9329
> URL: https://issues.apache.org/jira/browse/HDFS-9329
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: test
> Affects Versions: 2.7.1
> Reporter: Zhe Zhang
> Assignee: Zhe Zhang
> Priority: Minor
> Attachments: HDFS-9329.00.patch
>
>
> {{testRateThrottling}} verifies that bootstrap transfer should timeout with a
> very small {{DFS_IMAGE_TRANSFER_BOOTSTRAP_STANDBY_RATE_KEY}} value. However,
> throttling on the image sender only happens after sending each IO buffer.
> Therefore, the test sometimes fails if the receiver receives the full fsimage
> (which is smaller than IO buffer size) before throttling begins.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
