There is a PR being held up by Ignite failing a test under one build of hadoop.
The PR addresses Zeppelin+Scala handling companion objects for non-Spark scala interpreters. The PR is: https://github.com/apache/incubator-zeppelin/pull/794 The test that fails is: assertTrue(result.message().contains(sizeVal + ": Int = " + ignite.cluster().nodes().size())); fails in Hadoop 1.6 only (passes all other builds). The test is somewhat unique to Ignite, and only fails for Hadoop 1.6.0 build of Zeppelin. Also the spirit of the test is to ensure the Ignite interpreter is 'alive', a previous test makes sure that code runs successfully and that test is passing; this line of the test is testing the output of the code. This is not a configuration test, even though it looks like one. The question is, how significant is this failure (though it is certainly curious). And is it better to remove this test and allow companion objects in the Ignite-Scala interpreter or keep the test and open a new issue for someone to tackle to get companion objects working for Ignite. This PR is also holding up the companion object fix for Scalding and more importantly Flink (which is what I opened the PR for in the first place; Scalding is working fine, btw). If anyone could chime in, that would be swell- if not I'm going to default to dropping the Ignite fix and asking for a merge. tg Trevor Grant Data Scientist https://github.com/rawkintrevo http://stackexchange.com/users/3002022/rawkintrevo http://trevorgrant.org *"Fortunate is he, who is able to know the causes of things." -Virgil*
