Jenkins build is back to normal : beam_Release_NightlySnapshot #317

See

Re: Jenkins build became unstable: beam_PostCommit_Java_MavenInstall #2517

For the record: testAuthenticationWithBadPassword 59ms Fixed As said, I will watch for potential new failure in the future. Regards JB On 02/03/2017 05:48 AM, Jean-Baptiste Onofré wrote: New build (#2519) worked on JmsIO. Let's keep the Jira open and I will watch for failure. Thanks, Regards

Re: Jenkins build became unstable: beam_PostCommit_Java_MavenInstall #2517

New build (#2519) worked on JmsIO. Let's keep the Jira open and I will watch for failure. Thanks, Regards JB On 02/03/2017 05:26 AM, Jean-Baptiste Onofré wrote: Hmmm. Test on the PR worked. Let me take a look. Regards JB⁣​ On Feb 3, 2017, 03:27, at 03:27, Kenneth Knowles wrote: Failure is

Re: Jenkins build became unstable: beam_PostCommit_Java_MavenInstall #2517

Hmmm. Test on the PR worked. Let me take a look. Regards JB⁣​ On Feb 3, 2017, 03:27, at 03:27, Kenneth Knowles wrote: >Failure is JmsIO. This doesn't look related to the changes, but doesn't >look like a flake, so I'm not sure what to make of it. > >The test is looking for a message "User name

Re: Jenkins build became unstable: beam_PostCommit_Java_MavenInstall #2517

Failure is JmsIO. This doesn't look related to the changes, but doesn't look like a flake, so I'm not sure what to make of it. The test is looking for a message "User name [test_user] or password is invalid." which _is_, in fact in a message, but very deep in the causal chain of exceptions. Filed

[VOTE] Apache Beam, version 0.5.0, release candidate #2

Hi everyone, With JB leaving for his vacation, I'll try to push the 0.5.0 release across the finish line. Please review and vote on the release candidate #2 for the version 0.5.0, as follows: [ ] +1, Approve the release [ ] -1, Do not approve the release (please provide specific comments) The comp

Re: Jenkins build became unstable: beam_PostCommit_Java_MavenInstall #2510

The last time this occurred, it was because there's actually a small-but-not-impossible chance that the expected output windows are not saturated. We could do better by writing a manifest file as well or in the near term just reduce the number of buckets things are randomly strewn into or use a sli

Re: Jenkins build became unstable: beam_PostCommit_Java_MavenInstall #2510

Looks like a flake in WindowedWordCount -- listed change seems unrelated. On Thu, Feb 2, 2017 at 11:14 AM, Apache Jenkins Server < jenk...@builds.apache.org> wrote: > See MavenInstall/2510/changes> > > -- --- Jason Kuster Apache Beam /

JB off for 2 weeks

Hi guys, I'm in vacation tonight for 2 weeks. I will have access to my e-mail and I will try to follow to different discussions and threads. However, I can have some delay in my replies depending the tequila I will have in my blood ;) See you soon ! Regards JB -- Jean-Baptiste Onofré jbono

Re: [CANCEL][VOTE] Apache Beam, version 0.5.0, release candidate #1

It sounds good. Let's do that. Regards JB On 02/02/2017 12:12 PM, Aljoscha Krettek wrote: We should go with the option Kenn proposed instead of a "hard revert". On Thu, 2 Feb 2017 at 10:55 Jean-Baptiste Onofré wrote: Hi everyone, Due to the regression in the Flink runner (and potentially o

Re: [CANCEL][VOTE] Apache Beam, version 0.5.0, release candidate #1

We should go with the option Kenn proposed instead of a "hard revert". On Thu, 2 Feb 2017 at 10:55 Jean-Baptiste Onofré wrote: > Hi everyone, > > Due to the regression in the Flink runner (and potentially other > runners), I cancel RC#1. > > The fixes/reverts will be done on the 0.5.0 release br

[CANCEL][VOTE] Apache Beam, version 0.5.0, release candidate #1

Hi everyone, Due to the regression in the Flink runner (and potentially other runners), I cancel RC#1. The fixes/reverts will be done on the 0.5.0 release branch and a RC#2 will be done. Sorry about that. Regards JB On 01/27/2017 09:55 PM, Jean-Baptiste Onofré wrote: Hi everyone, Please

Re: [VOTE] Apache Beam, version 0.5.0, release candidate #1

Yes, I just wanted formal feedback (now we have 3 -1, but we had only 1 this morning) ;) On 02/02/2017 10:47 AM, Dan Halperin wrote: From the release guide, I was under the impression that the release manager is free to cancel once an issue is discovered. Canceling is not a release, so should n

Re: [VOTE] Apache Beam, version 0.5.0, release candidate #1

-1 (binding) Regards JB On 02/02/2017 10:47 AM, Dan Halperin wrote: From the release guide, I was under the impression that the release manager is free to cancel once an issue is discovered. Canceling is not a release, so should not require a specific number of -1s. See [0] That said, -1 to 0.

Re: [VOTE] Apache Beam, version 0.5.0, release candidate #1

>From the release guide, I was under the impression that the release manager is free to cancel once an issue is discovered. Canceling is not a release, so should not require a specific number of -1s. See [0] That said, -1 to 0.5.0-RC1 because of the regression in Flink and perhaps other runners.