[ 
https://issues.apache.org/jira/browse/GIRAPH-276?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13429628#comment-13429628
 ] 

Hudson commented on GIRAPH-276:
-------------------------------

Integrated in Giraph-trunk-Commit #159 (See 
[https://builds.apache.org/job/Giraph-trunk-Commit/159/])
    GIRAPH-276. Fix broken tests in pseudo-distributed mode. Contributed by 
Alessandro Presta (Revision 1370085)

     Result = FAILURE
jghoman : 
http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1370085
Files : 
* /giraph/trunk/CHANGELOG
* /giraph/trunk/src/main/java/org/apache/giraph/comm/BasicRPCCommunications.java
* 
/giraph/trunk/src/main/java/org/apache/giraph/comm/CommunicationsInterface.java
* /giraph/trunk/src/test/java/org/apache/giraph/TestGraphPartitioner.java

                
> Fix broken tests in pseudo-distributed mode
> -------------------------------------------
>
>                 Key: GIRAPH-276
>                 URL: https://issues.apache.org/jira/browse/GIRAPH-276
>             Project: Giraph
>          Issue Type: Bug
>            Reporter: Alessandro Presta
>            Assignee: Alessandro Presta
>             Fix For: 0.2.0
>
>         Attachments: GIRAPH-276.patch
>
>
> As it was pointed out, the following tests are currently broken in 
> pseudo-distributed mode:
> 1) testBspPageRank, testBspPageRankWithAggregatorWriter
> These have been broken for a long time AFAIK, and are due to errors in 
> aggregators. Maja's patch (GIRAPH-259) fixes them, so I won't address them 
> here.
> 2) TestMutateGraphVertex
> After GIRAPH-244, Edge doesn't implement Writable anymore. Unfortunately, 
> there was a sneaky place where it was still required: when using Hadoop RPC, 
> BasicRPCCommunications#addEdge() on the server is called through a proxy, and 
> its parameters are expected to be Writable. A simple signature change fixes 
> this.
> 3) TestGraphPartitioner
> I opened GIRAPH-269 on this one. I don't know how we got to the hardcoded 
> number of expected bytes (123), but if that was obtained by running the test 
> itself, then it's not a great test.
> For now, I'm simply changing that to 120, which is what we get after the 
> serialization changes in EdgeListVertex, but I'm leaving GIRAPH-269 open 
> since we should replace this with something more robust.
> I added a TODO comment so we don't forget it.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Reply via email to