[
https://issues.apache.org/jira/browse/HDFS-11122?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15654836#comment-15654836
]
Mingliang Liu commented on HDFS-11122:
--------------------------------------
Triggering block report should help here obviously.
+1
{code}
540 } catch (IOException ie) {
541 assertTrue(ie instanceof ChecksumException);
542 }
{code}
I'd prefer catch the ChecksumException directly and name it ignored. This is
minor coding style preference difference though.
The waitFor is now using 3000 interval, which is too long? 3000 (3 seconds)
seems unnecessary if we can return fast. How about 500/1000?
> TestDFSAdmin.testReportCommand fails due to timed out
> -----------------------------------------------------
>
> Key: HDFS-11122
> URL: https://issues.apache.org/jira/browse/HDFS-11122
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: test
> Reporter: Yiqun Lin
> Assignee: Yiqun Lin
> Priority: Minor
> Attachments: HDFS-11122.001.patch, HDFS-11122.002.patch,
> HDFS-11122.003.patch
>
>
> After HDFS-11083, the test {{TestDFSAdmi}} fails sometimes dueto timed out.
> The stack
> infos(https://builds.apache.org/job/PreCommit-HDFS-Build/17484/testReport/)：
> {code}
> java.lang.Exception: test timed out after 30000 milliseconds
> at java.lang.Thread.sleep(Native Method)
> at
> org.apache.hadoop.test.GenericTestUtils.waitFor(GenericTestUtils.java:268)
> at
> org.apache.hadoop.hdfs.tools.TestDFSAdmin.testReportCommand(TestDFSAdmin.java:540)
> {code}
> The timed out is happened in {{GenericTestUtils.waitFor}}. We can make a
> improvement in the logic of waiting the corrupt blocks.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
