JBTM-3260 On publishing a new Narayana release there should be created Quarkus PR to update Narayana version

[mayankkunwar]

https://issues.redhat.com/browse/JBTM-3260

MAIN
!QA_JTA !QA_JTS_JDKORB !QA_JTS_OPENJDKORB !QA_JTS_JACORB !BLACKTIE !XTS !PERF NO_WIN !RTS !AS_TESTS !TOMCAT !JACOCO !LRA

[jbosstm-bot]

Started testing this pull request with PERF profile: http://172.17.130.4:8083/job/btny-pulls-narayana-perf/PROFILE=PERF,jdk=jdk8.latest,label=master/1192/

[jbosstm-bot]

Benchmark output (please refer to the article https://developer.jboss.org/wiki/PerformanceGatesForAcceptingPerformanceFixesInNarayana for information on our testing procedures.

If you just want to run a single benchmark then please refer to the README.md file in our benchmark repository at https://github.com/jbosstm/performance/tree/master/narayana

JMH benchmark run (with args -t 240 -r 25 -f 2 -wi 5 -i 5)

Comparison (pull request versus master)
(changes within the % range [-10.000000, 10.000000] are regarded as insignificant):

com.hp.mwtests.ts.arjuna.atomicaction.CheckedActionTest.testThreadActionData: 2031013.181800 vrs 2016314.666538 (0.000000%: no change)
com.arjuna.ats.jta.xa.performance.JTAStoreTests.jtaTest: 453031.979404 vrs 466436.968179 (0.000000%: no change)
com.hp.mwtests.ts.arjuna.performance.Performance1.twoPhaseTest: 1418827.672012 vrs 1504955.993373 (0.000000%: no change)
com.hp.mwtests.ts.arjuna.performance.Performance1.onePhaseTest: 4209615.945433 vrs 4272752.417047 (0.000000%: no change)
com.hp.mwtests.ts.arjuna.atomicaction.CheckedActionTest.testCheckedAction: 48646.575084 vrs 48226.543036 (0.000000%: no change)

For information on the hardware config used for this PR please consult the CI job artefact hwinfo.txt or the job output

If the purpose of this PR is to improve performance then there has been insufficient improvement to warrant a pass. See the previous text for the threshold (range) for passing optimization related PRs

[jbosstm-bot]

PERF profile job finished http://172.17.130.4:8083/job/btny-pulls-narayana-perf/PROFILE=PERF,jdk=jdk8.latest,label=master/1192/

[ochaloup]

Please, have you tried to run it locally? It's important as the CI won't catch any issue in the script. It will be found only during the release process.

I think we need to have something similar as for WFLY here:
Then there is this read -p question.
https://github.com/jbosstm/narayana/blob/5.10.4.Final/narayana-release-process.sh#L17
The quarkus issue is different as it's not created via https://issues.redhat.com but at quarkus github. Ie.g this is a bug here: https://github.com/quarkusio/quarkus/issues/new?assignees=&labels=kind%2Fbug&template=bug_report.md&title=
and we should add with label area/narayana and kind/enhancement

So, let's summarize what should be a nice to have as a result:

• It would be good to have an automatic creation of the Quarkus issue which asks for update of the Narayana version.
  ** WildFly only asks you to create an issue manualy (https://github.com/jbosstm/narayana/blob/5.10.4.Final/narayana-release-process.sh#L17). Here it would be nice to be automatized a bit.
  ** I'm not sure if it's possible with github API or something but if it's possible then instead of asking if the Quarkus issue was created it would be good to provide more automatized way of creating of the issue. E.g. create it and then show in the browser that the user who proceed with release may adjust it.
• Then there should be the code you provided
  ** Try it locally as the CI won't catch issues in processing.
  ** The pull request against the created issue should be done. The PR may say that it fixes that issue and if the PR is merged the issue is then automatically closed (https://help.github.com/en/github/managing-your-work-on-github/linking-a-pull-request-to-an-issue#linking-a-pull-request-to-an-issue-using-a-keyword).

[ochaloup]

I'm not sure if it's good to go with the same issue name with Quarkus as it was for WFLY

[mayankkunwar]

I am using the same issue name because when narayana-release-process.sh is called, WFLYISSUE is coming as third argument, and I was not knowing from where, when and who calls this file. Otherwise I would have used the fourth argument for QRKSISSUE.

[ochaloup]

I think this won't work as you go with abs path

[mayankkunwar]

locally this was working, but anyways it is always better to use full path, so I will change it as requested.

[ochaloup]

I assume this won't work (abs path used)

[mayankkunwar]

I will refactor this as per request

[ochaloup]

I assume there is no 5_BRANCH in quarkus repo. I think the master should be used.

[mayankkunwar]

okay, will change it to master.

[ochaloup]

@mayankkunwar ok, I'm putting this PR on hold as @xstefank pointed me out to the dependabot. Maybe we may just add the Narayana dependency to the quarkus config dependabot file (https://github.com/quarkusio/quarkus/blob/master/.dependabot/config.yml). Just I'm not sure about rules of such thing.

Please, do not proceed with the changes that I suggested and let's talk about the bot tomorrow - if it's a feasible option for us.

[mayankkunwar]

@ochaloup okay, we can have a discussion on the same before moving forward.

[mayankkunwar]

I am going to close this PR, as we tried to solve this problem with a better alternative solution, by adding narayana dependencies to quarkus dependabot config.xml.
Here is the PR link: quarkusio/quarkus#7977