> On Oct. 16, 2013, 10:27 p.m., Jarek Cecho wrote:
> > Hi Abe,
> > thank you very much for working on this patch! Please accept my apologies 
> > for late review! I'm having difficulties in Hadoop 100 profile - getting 
> > NullPointerExceptions in the cluster initialization. I was wondering if 
> > this patch version is not yet ready for Hadoop 1 or whether I should dig 
> > into my environment to see what is going on.
> > 
> > 
> >

Jarcec, great points. I'm looking into what's going wrong with the Hadoop 100 
profile right now. As always, thank you for the review!


> On Oct. 16, 2013, 10:27 p.m., Jarek Cecho wrote:
> > pom.xml, line 99
> > <https://reviews.apache.org/r/10964/diff/8/?file=344782#file344782line99>
> >
> >     Can we wait until SQOOP-1218 will get committed? That will upgrade the 
> > version even further to 2.2.0 (first GA release of Hadoop 2).

Indeed.


> On Oct. 16, 2013, 10:27 p.m., Jarek Cecho wrote:
> > test/pom.xml, line 120
> > <https://reviews.apache.org/r/10964/diff/8/?file=344783#file344783line120>
> >
> >     One hour long timeout seems too much, what about 900 seconds (15 
> > minutes)?

Unfortunately, there is a test case that can take up to 30 minutes I've found. 
The PartitionerTest uses the parameterized runner and can take a while to 
finish.


On Oct. 16, 2013, 10:27 p.m., Abraham Elmahrek wrote:
> > Jarcec

Unfortunately, there is a test case that can take up to 30 minutes I've found. 
The PartitionerTest uses the parameterized runner and can take a while to 
finish.


- Abraham


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/10964/#review27093
-----------------------------------------------------------


On Aug. 26, 2013, 5:24 p.m., Abraham Elmahrek wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/10964/
> -----------------------------------------------------------
> 
> (Updated Aug. 26, 2013, 5:24 p.m.)
> 
> 
> Review request for Sqoop and Jarek Cecho.
> 
> 
> Bugs: SQOOP-927
>     https://issues.apache.org/jira/browse/SQOOP-927
> 
> 
> Repository: sqoop-sqoop2
> 
> 
> Description
> -------
> 
> commit ff7dba55a09dd7789a34136233000c625759e583
> Author: Abraham Elmahrek <[email protected]>
> Date:   Fri Apr 26 15:10:24 2013 -0700
> 
>     SQOOP-927 Sqoop2: Integration: Mapreduce specific tests should be running 
> on MiniCluster
>     
>     Handle MiniDFSCluster and MiniMRClientCluster on own.
>     
>     Set yarn.application.classpath to get over classpath errors.
>     Set to use fair scheduler.
> 
> :100644 100644 0abbb18... f09704b... M        pom.xml
> :100644 100644 6eb3184... a4c2a5b... M        
> test/src/test/java/org/apache/sqoop/integration/TomcatTestCase.java
> :100644 100644 0f48a8b... 758c885... M        
> test/src/test/java/org/apache/sqoop/integration/server/VersionTest.java
> 
> 
> Diffs
> -----
> 
>   pom.xml 513b6d0570ee6f1ec32f8c35a403becc1cfa156a 
>   test/pom.xml 8001fcec29bf5a7d9c3119b9f6ef548f678e1ebe 
>   test/src/main/java/org/apache/sqoop/test/asserts/HdfsAsserts.java 
> c17527285167104994d78f5919b7f11c0c31cd15 
>   test/src/main/java/org/apache/sqoop/test/hadoop/HadoopLocalRunner.java 
> PRE-CREATION 
>   
> test/src/main/java/org/apache/sqoop/test/hadoop/HadoopMiniClusterRunner.java 
> PRE-CREATION 
>   test/src/main/java/org/apache/sqoop/test/hadoop/HadoopRunner.java 
> PRE-CREATION 
>   test/src/main/java/org/apache/sqoop/test/hadoop/HadoopRunnerFactory.java 
> PRE-CREATION 
>   test/src/main/java/org/apache/sqoop/test/testcases/ConnectorTestCase.java 
> d10b94268babeb3dc7b1d2450776af5fd8449fd3 
>   test/src/main/java/org/apache/sqoop/test/testcases/TomcatTestCase.java 
> efdfed4e2065c0c01c16da88cf55705ab13d7bcf 
>   test/src/main/java/org/apache/sqoop/test/utils/HdfsUtils.java 
> 95dd177a11c5d822e35ca54e5d93785f0a40fbfc 
> 
> Diff: https://reviews.apache.org/r/10964/diff/
> 
> 
> Testing
> -------
> 
> Ran integration tests without and with miniclusters.
> Currently need to use both miniclusters or neither for tests to work.
> 
> 
> Thanks,
> 
> Abraham Elmahrek
> 
>

Reply via email to