hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Zhe Zhang (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HDFS-9329) TestBootstrapStandby#testRateThrottling is flaky because fsimage size is smaller than IO buffer size
Date Mon, 02 Nov 2015 18:35:27 GMT

     [ https://issues.apache.org/jira/browse/HDFS-9329?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Zhe Zhang updated HDFS-9329:
----------------------------
       Resolution: Fixed
     Hadoop Flags: Reviewed
    Fix Version/s: 2.8.0
           Status: Resolved  (was: Patch Available)

Thanks Andrew and Xiao for the reviews! I just committed the patch to trunk and branch-2.


While committing to branch-2 I had to update the patch, both to fix a conflict and because
branch-2's version of {{testSuccessfulBaseCase}} cleans up the generated primary NN directory
after each run.

> 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
>             Fix For: 2.8.0
>
>         Attachments: HDFS-9329.00.patch, HDFS-9329.01.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)

Mime
View raw message