Filed https://issues.apache.org/jira/browse/BEAM-11409 as blocking issue. Linked PRs in issue.
On Mon, Dec 7, 2020 at 11:42 AM Robert Burke <[email protected]> wrote: > Ack. Thank you! Let me know here for the cherry pick. > > On Mon, Dec 7, 2020, 11:31 AM Chamikara Jayalath <[email protected]> > wrote: > >> -1 >> Seems like Dataflow x-lang jobs might be broken for Java 11 users due to >> Java SDK defaulting to Java 11 containers [1] which are not released with >> Beam 2.26.0 [2]. +Emily Ye <[email protected]> is looking into this. >> >> Thanks, >> Cham >> >> [1] >> Error: Error syncing pod 0a07bd35d4a1e02006563901e421445f >> ("dataflow-kafka-taxi-20201207-10105-12071011-oij4-harness-6ss0_default(0a07bd35d4a1e02006563901e421445f)"), >> skipping: failed to "StartContainer" for "sdk-1-0" with ImagePullBackOff: >> "Back-off pulling image \"apache/beam_java11_sdk:2.26.0\"" >> >> [2] https://hub.docker.com/r/apache/beam_java11_sdk >> >> On Mon, Dec 7, 2020 at 9:41 AM Robert Burke <[email protected]> wrote: >> >>> Last Friday I re-built and re-pushed the containers that Ahmet found. >>> I'm going to redo that validation and try and get through additional >>> validation today. >>> >>> As it stands I'm back to a +0 due to Ahmets position on the python 3.7 >>> tests. (I'm not familiar with the python suites, so I defer to an expert) >>> >>> I'll send another update at the end of my day >>> >>> On Fri, Dec 4, 2020, 2:13 PM Ahmet Altay <[email protected]> wrote: >>> >>>> >>>> >>>> On Fri, Dec 4, 2020 at 2:03 PM Robert Burke <[email protected]> wrote: >>>> >>>>> Is this an issue on the dataflow side or with the quick start >>>>> instructions? >>>>> >>>> >>>> It is an issue with how the dataflow container is created. Quick start >>>> instructions are fine. >>>> >>>> >>>>> >>>>> On Fri, Dec 4, 2020, 1:43 PM Ahmet Altay <[email protected]> wrote: >>>>> >>>>>> -1 - I validated the python quickstarts with python 3.8. All worked >>>>>> except for the Dataflow batch pipeline. I will update my vote once this >>>>>> is >>>>>> resolved. >>>>>> >>>>>> Dataflow batch pipeline failed with the "RuntimeError: Beam SDK base >>>>>> version 2.26.0 does not match Dataflow Python worker version >>>>>> 2.25.0.dev. Please check Dataflow worker startup logs and make sure >>>>>> that correct version of Beam SDK is installed." error. This could be >>>>>> fixed >>>>>> by updating the relevant dataflow container and keeping this RC as is >>>>>> otherwise. >>>>>> >>>>>> This was an issue in the previous 2 releases as well. Does anyone >>>>>> have a suggestion on how we could catch this before an RC email. >>>>>> >>>>>> On Fri, Dec 4, 2020 at 10:38 AM Ahmet Altay <[email protected]> wrote: >>>>>> >>>>>>> >>>>>>> >>>>>>> On Thu, Dec 3, 2020 at 8:41 PM Robert Burke <[email protected]> >>>>>>> wrote: >>>>>>> >>>>>>>> Whoops, apologies for the title typo. Several years before we hit >>>>>>>> 2.62.0 ! >>>>>>>> >>>>>>>> Regarding this RC: >>>>>>>> >>>>>>>> -1 There's an extant issue with Python 3.7 >>>>>>>> :sdks:python:test-suites:direct:py37:directRunnerIT . discuss in >>>>>>>> https://issues.apache.org/jira/browse/BEAM-11396 >>>>>>>> >>>>>>> >>>>>>> I do not think this is a blocker. >>>>>>> >>>>>>> I looked at the issue. It looks like it is either a network or an >>>>>>> auth issue contained in the pubsub dependency used to set up the test >>>>>>> environment. The test fails to create a pubsub subscription (and this is >>>>>>> unrelated to pubsub io). Given that the same IT test runs and works on >>>>>>> other runners, we can consider this an issue with the test setup itself >>>>>>> for >>>>>>> the direct runner. >>>>>>> >>>>>>> >>>>>>>> I felt the need to finish RC tasks at least once end to end. This >>>>>>>> has been a challenging month to get anything done. On top of that >>>>>>>> among the >>>>>>>> multitude of Beam contributors, I'm one of the few who do not interact >>>>>>>> with >>>>>>>> either Python or Java regularly. On top of that, I do most of my work >>>>>>>> from >>>>>>>> a Chromebook using Crostini, which is essentially a lightweight Debian >>>>>>>> container. It doesn't have much pre-installed. >>>>>>>> Conveniently this means I get to find every single configuration >>>>>>>> assumption that's been made as part of the release process, which I've >>>>>>>> taken notes on. I'll be contributing these learnings back to the >>>>>>>> release >>>>>>>> guide in a PR in the next few weeks. Overall it feels like an >>>>>>>> unobstructed >>>>>>>> RC should only take about a day to sort out, if some parts of the >>>>>>>> process >>>>>>>> are clarified via expectation setting and other guide updates. >>>>>>>> It's in that spirit that I sent out this RC00, so validation on >>>>>>>> 2.26.0 can start on the rest of the components in the meantime. Now >>>>>>>> that >>>>>>>> I've gone through the process once, remaining RCs will be prompt. >>>>>>>> >>>>>>>> With that in mind, if there are any additional regressions from >>>>>>>> 2.25.0 nows the time to file a JIRA for triage in order to have fixes >>>>>>>> cherry picked for them. >>>>>>>> >>>>>>>> Cheers, >>>>>>>> Robert Burke >>>>>>>> 2.26.0 Release manager. >>>>>>>> >>>>>>>> >>>>>>>> On Thu, 3 Dec 2020 at 20:01, Robert Burke <[email protected]> wrote: >>>>>>>> >>>>>>>>> Hi everyone, >>>>>>>>> Please review and vote on the release candidate #00 for the >>>>>>>>> version 2.26.0, as follows: >>>>>>>>> [ ] +1, Approve the release >>>>>>>>> [ ] -1, Do not approve the release (please provide specific >>>>>>>>> comments) >>>>>>>>> >>>>>>>>> >>>>>>>>> Reviewers are encouraged to test their own use cases with the >>>>>>>>> release candidate, and vote +1 >>>>>>>>> if no issues are found. >>>>>>>>> >>>>>>>>> The complete staging area is available for your review, which >>>>>>>>> includes: >>>>>>>>> * JIRA release notes [1], >>>>>>>>> * the official Apache source release to be deployed to >>>>>>>>> dist.apache.org [2], which is signed with the key with >>>>>>>>> fingerprint A52F5C83BAE26160120EC25F3D56ACFBFB2975E1 [3], >>>>>>>>> * all artifacts to be deployed to the Maven Central Repository [4], >>>>>>>>> * source code tag "v2.26.0-RC00" [5], >>>>>>>>> * website pull request listing the release [6], publishing the API >>>>>>>>> reference manual [7], and the blog post [8]. >>>>>>>>> * Java artifacts were built with Maven 3.6.0 and OpenJDK 1.8.0_275. >>>>>>>>> * Python artifacts are deployed along with the source release to >>>>>>>>> the dist.apache.org [2]. >>>>>>>>> * Validation sheet with a tab for 2.26.0 release to help with >>>>>>>>> validation [9]. >>>>>>>>> * Docker images published to Docker Hub [10]. >>>>>>>>> >>>>>>>>> The vote will be open for at least 72 hours. It is adopted by >>>>>>>>> majority approval, with at least 3 PMC affirmative votes. >>>>>>>>> >>>>>>>>> Thanks, >>>>>>>>> Robert Burke >>>>>>>>> 2.26.0 Release Manager >>>>>>>>> >>>>>>>>> [1] >>>>>>>>> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12319527&version=12348833 >>>>>>>>> [2] https://dist.apache.org/repos/dist/dev/beam/2.26.0/ >>>>>>>>> [3] https://dist.apache.org/repos/dist/release/beam/KEYS >>>>>>>>> [4] >>>>>>>>> https://repository.apache.org/content/repositories/orgapachebeam-1143/ >>>>>>>>> [5] https://github.com/apache/beam/tree/v2.26.0-RC00 >>>>>>>>> [6] https://github.com/apache/beam/pull/13481 >>>>>>>>> [7] https://github.com/apache/beam-site/pull/609 >>>>>>>>> [8] https://github.com/apache/beam/pull/13482 >>>>>>>>> [9] >>>>>>>>> https://docs.google.com/spreadsheets/d/1qk-N5vjXvbcEk68GjbkSZTR8AGqyNUM-oLFo_ZXBpJw/edit#gid=. >>>>>>>>> .. >>>>>>>>> [10] https://hub.docker.com/search?q=apache%2Fbeam&type=image >>>>>>>>> >>>>>>>>
