Jenkins build is still unstable: beam_Release_NightlySnapshot #353

2017-03-10 Thread Apache Jenkins Server
See

[VOTE] Release 0.6.0, release candidate #2

2017-03-10 Thread Ahmet Altay
Hi everyone, Please review and vote on the release candidate #2 for the version 0.6.0, as follows: [ ] +1, Approve the release [ ] -1, Do not approve the release (please provide specific comments) The complete staging area is available for your review, which includes: * JIRA release notes [1],

Re: Style: how much testing for transform builder classes?

2017-03-10 Thread Kenneth Knowles
+0.5 Tests of trivial validation failures, if they check the error message, are actually tests of effective communication in the error message (example: testReadValidationFailsQueryLimitZero). These are on easily on par with functionality tests in terms of importance for users. But tests of

Re: Jenkins build became unstable: beam_Release_NightlySnapshot #352

2017-03-10 Thread Kenneth Knowles
Actually, this is not a flake and the cause is a double close, reverted in https://github.com/apache/beam/pull/2226 On Fri, Mar 10, 2017 at 12:56 PM, Kenneth Knowles wrote: > I see a failure that looks flake-like in BigQueryTornadoesIT. > > Also some non-fatal warnings in the

Re: First time PR question: Multiple issues addressed in single change

2017-03-10 Thread Ahmet Altay
Hi Ankur On Fri, Mar 10, 2017 at 6:08 PM, Ankur Chauhan wrote: > Hi all, > > This is my first go at contributing some simple changes to the BigTableIO > in beam. The main intent was to update the dependency to 0.9.5.1 from the > current 0.9.2 which would address some changes

Re: [VOTE] Release 0.6.0, release candidate #1

2017-03-10 Thread Ahmet Altay
Let's cancel the vote on RC1 and I will start a new one. I think Davor had a great summary and I will make those changes for RC2. I only have one comment, I would rather keep a single file format (.zip) instead of using both file formats (.zip, .tar.gz). I think this is better for having a single

Re: [VOTE] Release 0.6.0, release candidate #1

2017-03-10 Thread Kenneth Knowles
> * Also I got this minor warning on the Maven build: > > > > [WARNING] Some problems were encountered while building the effective > model > > for org.apache.beam:beam-runners-google-cloud-dataflow-java:jar:0.6.0 > > [WARNING] 'dependencies.dependency.(groupId:artifactId:type:classifier)' > >

Re: Jenkins build became unstable: beam_Release_NightlySnapshot #352

2017-03-10 Thread Kenneth Knowles
I see a failure that looks flake-like in BigQueryTornadoesIT. Also some non-fatal warnings in the javadoc for MongoDbGridFsIO that we should still fix. On Thu, Mar 9, 2017 at 11:44 PM, Apache Jenkins Server < jenk...@builds.apache.org> wrote: > See

Style: how much testing for transform builder classes?

2017-03-10 Thread Eugene Kirpichov
Hello, I've seen a pattern in a couple of different transforms (IOs) where we, I think, spend an excessive amount of code unit-testing the trivial builder methods. E.g. a significant part of

Re: [VOTE] Release 0.6.0, release candidate #1

2017-03-10 Thread Aljoscha Krettek
Sorry for the BEAM-1674 problem. We just discovered this by chance because Kenneth added a more thorough Stateful DoFn test. I have a fix for it in this PR: https://github.com/apache/beam/pull/2217 I'm afraid we have to cancel the release yet again because this is a real bug that people can run