[
https://issues.apache.org/jira/browse/HADOOP-5484?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Amar Kamat updated HADOOP-5484:
-------------------------------
Attachment: HADOOP-5484-v1.0.patch
I was not able to reproduce the issue. Here is what is happening :
{code}
dir = "test-recovery-manager"
test-case1
cleanup dir
run 2 jobs with output as dir/output1 and dir/output2
testcase1-end
test-case2
cleanup dir
run 3 jobs with output as dir/output1, dir/output2 and dir/output3
testcase2-end
{code}
Here the second testcase is still seeing output1 from testcase1. Spoke to Jothi
and he says its some nfs issue.
Anyways attaching a patch that should solve the issue. The patch uses unique
output folders across tests. I dont think any tests are required for this.
> TestRecoveryManager fails wtih FileAlreadyExistsException
> ---------------------------------------------------------
>
> Key: HADOOP-5484
> URL: https://issues.apache.org/jira/browse/HADOOP-5484
> Project: Hadoop Core
> Issue Type: Bug
> Components: test
> Affects Versions: 0.21.0
> Reporter: Hairong Kuang
> Fix For: 0.21.0
>
> Attachments: HADOOP-5484-v1.0.patch
>
>
> TestRecoveryManager always fails when I run core tests in a linux redhat
> machine. It does not fail on a Mac machine.
> Testcase: testRecoveryManager took 55.842 sec
> Caused an ERROR
> Output directory file:XX/build/test/data/test-recovery-manager/output1
> already exists
> org.apache.hadoop.mapred.FileAlreadyExistsException: Output directory
> file:XX/build/test/data/test-recovery-manager/output1 already exists
> at
> org.apache.hadoop.mapred.FileOutputFormat.checkOutputSpecs(FileOutputFormat.java:111)
> at
> org.apache.hadoop.mapred.JobClient.submitJobInternal(JobClient.java:772)
> at org.apache.hadoop.mapred.JobClient.submitJob(JobClient.java:730)
> at
> org.apache.hadoop.mapred.TestRecoveryManager.testRecoveryManager(TestRecoveryManager.java:196)
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
