JBTM-845 include infinispan store in the store quickstart by mmusgrov · Pull Request #629 · jbosstm/quickstart

mmusgrov · 2026-04-24T14:51:59Z

https://redhat.atlassian.net/browse/JBTM-845

NARAYANA_REPO=mmusgrov
NARAYANA_BRANCH=JBTM-845
NY_BRANCH=JBTM-845

I have put the PR on hold since this PR depends on jbosstm/narayana#2357 and the GH action isn't detecting the NY_BRANCH=JBTM-845 variable

marcosgopen · 2026-04-27T09:16:19Z

            }

            core.setOutput('environment_variables', JSON.stringify(extractedEnvVars));
+            echo "quickstart parser GITHUB_ENV: $GITHUB_ENV" 


I think these 2 lines are causing the parser to fail (script is for JS, run is for shell). Please replace them with:

core.info(quickstart parser extracted envs: ${JSON.stringify(extractedEnvVars)});

Or create a new step:

name: Debug runner env (optional)
if: ${{ github.event_name == 'pull_request' }}
run: |
echo "GITHUB_ENV: $GITHUB_ENV"
env

Do you have any thoughts about why GH action isn't detecting the env vars in the PR description?
I added some debug to main.yml over the weekend and the env vars are not processed.

Is the syntax NY_BRANCH=JBTM-845 correct?

I remember we had a similar problem with other projects. The variables were set in the initial parser but then not available while executing the script IIRC. I will have a deeper look

I think these 2 lines are causing the parser to fail (script is for JS, run is for shell).

Ah that explains a lot, thanks.

I removed the debug commit.

marcosgopen · 2026-04-28T09:26:28Z

@mmusgrov I asked for help to copilot about the issue and this is what it says:

Your PR body contains:

    NARAYANA_REPO=mmusgrov
    NARAYANA_BRANCH=JBTM-845
    NY_BRANCH=JBTM-845

But .github/workflows/parser.yml only extracts WILDFLY_RELEASE_VERSION from the PR description and passes only that through to the main workflow as environment variables.

So quickstart.sh runs with defaults:

    NARAYANA_REPO=${NARAYANA_REPO:-jbosstm}
    NARAYANA_BRANCH="${NARAYANA_BRANCH:-main}"
    NY_BRANCH is empty

(see quickstart.sh where those defaults are set: NARAYANA_REPO, NARAYANA_BRANCH, and optional NY_BRANCH are consumed there).
Fix: teach parser.yml to extract and forward Narayana variables

Update the possibleEnvVars list in .github/workflows/parser.yml to include:

    NARAYANA_REPO
    NARAYANA_BRANCH
    NY_BRANCH

So they get exported in .github/workflows/main.yml step “Load Dynamic Environment Variables”, and then quickstart.sh will clone/build the intended Narayana repo/branch.

Here is the concrete change:
jbosstm / quickstart / .github / workflows / parser.ymlv1

            const possibleEnvVars = [
              'WILDFLY_RELEASE_VERSION'
            ];

Change it to:
jbosstm / quickstart / .github / workflows / parser.ymlv2

            const possibleEnvVars = [
              'WILDFLY_RELEASE_VERSION',
              'NARAYANA_REPO',
              'NARAYANA_BRANCH',
              'NY_BRANCH'
            ];

marcosgopen · 2026-04-28T10:49:43Z

  # INITIALIZE LRA ENV
  export MAVEN_OPTS="-Xmx1024m -XX:MaxMetaspaceSize=512m"

-  LRA_BRANCH=main


marcosgopen · 2026-04-28T10:50:11Z

  git clone https://github.com/${NARAYANA_REPO}/lra.git -b ${LRA_BRANCH}
  [ $? = 0 ] || fatal "git clone https://github.com/${NARAYANA_REPO}/lra.git failed"
  echo "Checking if need Narayana LRA PR"
-  if [ -n "$NY_BRANCH" ]; then


I think we should keep having $NY_BRANCH here

Oh, I see, that would cause to override the lra branch with the narayana branch. So that makes sense. However the change means that we will test the quickstart with lra main branch by default instead of a fixed version

Specifically in this case the quickstart clones and builds lra from mmusgrov/lra.git main branch which is outdated: in fact I can see an error that we used to have with lra some months ago. cc @mmusgrov

NY_BRANCH is the branch being tested for the narayana repo (in my case JBTM-845). I am not testing an LRA branch of the narayana repo so I need the LRA quickstarts to run against main so I introduced a new env variable for that - the problem is that the quickstarts make use of two repos (narayana and lra) and the script was not originally written to cope with that.

In order to fix this we could add
LRA_REPO=${LRA_REPO:-jbosstm}

and then replace 'git clone https://github.com/${NARAYANA_REPO}/lra.git -b ${LRA_BRANCH}'
with
git clone https://github.com/${LRA_REPO}/lra.git -b ${LRA_BRANCH}

In order to fix this we could add LRA_REPO=${LRA_REPO:-jbosstm}

and then replace 'git clone https://github.com/${NARAYANA_REPO}/lra.git -b ${LRA_BRANCH}' with git clone https://github.com/${LRA_REPO}/lra.git -b ${LRA_BRANCH}

I'll give it a go
I deleted my suggestion

Thanks
NY_LRA_BRANCH or LRA_BRANCH are both fine with me

mmusgrov · 2026-04-28T12:39:35Z

-    echo "Checkout failed: $BUILD_URL";
-    exit -1
-  fi
+  echo "Building LRA PR ${LRA_BRANCH}"


Note that I removed the check if [ -n "$LRA_BRANCH" ]; because LRA_BRANCH must be set at line 99 (LRA_BRANCH=${LRA_BRANCH:-main}. And we already check for a valid branch (git checkout $LRA_BRANCH on line 110).

I also removed the -b ${LRA_BRANCH} option from the git clone step since we go a git checkout $LRA_BRANCH after cloning.

I pushed another change, I had forgotten to update the GH Action with the new LRA_REPO and LRA_BRANCH var.

There's also a script/action somewhere that populates new PR descriptions with a definition of what vars are processed - I'll look for that and add these two new vars to it.

I made similar changes to the perf repo PR: jbosstm/performance#204

mmusgrov · 2026-04-28T15:04:35Z

Getting the PR correct was quite a saga, thanks @marcosgopen for your input and help.

marcosgopen

Thank you Mike!

mmusgrov added the Hold label Apr 24, 2026

mmusgrov mentioned this pull request Apr 24, 2026

JBTM-845 Datagrid ObjectStore jbosstm/narayana#2537

Merged

mmusgrov force-pushed the JBTM-845 branch 2 times, most recently from e2f2cbf to b828061 Compare April 27, 2026 07:40

mmusgrov changed the title ~~JBTM-845 include infinispan store in the store benchmarks~~ JBTM-845 include infinispan store in the store qucikstart Apr 27, 2026

mmusgrov changed the title ~~JBTM-845 include infinispan store in the store qucikstart~~ JBTM-845 include infinispan store in the store quickstart Apr 27, 2026

marcosgopen reviewed Apr 27, 2026

View reviewed changes

mmusgrov force-pushed the JBTM-845 branch 2 times, most recently from 8399ea3 to e915aaf Compare April 27, 2026 14:03

marcosgopen reviewed Apr 28, 2026

View reviewed changes

mmusgrov force-pushed the JBTM-845 branch from e6fefe1 to 2664565 Compare April 28, 2026 12:33

mmusgrov commented Apr 28, 2026

View reviewed changes

mmusgrov force-pushed the JBTM-845 branch from 2664565 to 411d8aa Compare April 28, 2026 12:46

mmusgrov removed the Hold label Apr 28, 2026

mmusgrov force-pushed the JBTM-845 branch from 411d8aa to 18c92a8 Compare April 28, 2026 13:51

marcosgopen reviewed Apr 28, 2026

View reviewed changes

Comment thread ArjunaJTA/object_store/run.sh Outdated

marcosgopen reviewed Apr 28, 2026

View reviewed changes

Comment thread ArjunaJTA/object_store/run.sh Outdated

mmusgrov added 2 commits April 28, 2026 15:13

JBTM-845 include infinispan store in the store quickstarts

a001155

JBTM-845 update GH Action to parse NY_BRANCH etc and build script

dc89f36

mmusgrov force-pushed the JBTM-845 branch from 18c92a8 to dc89f36 Compare April 28, 2026 14:13

marcosgopen approved these changes Apr 28, 2026

View reviewed changes

mmusgrov merged commit 9bc6503 into jbosstm:main May 1, 2026
3 checks passed

mmusgrov mentioned this pull request Jun 9, 2026

JBTM-845 fix typo in the infinispan store README #743

Merged

Uh oh!

Conversation

mmusgrov commented Apr 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

mmusgrov Apr 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

marcosgopen commented Apr 28, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

mmusgrov Apr 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

mmusgrov Apr 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

mmusgrov commented Apr 28, 2026

Uh oh!

marcosgopen left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

mmusgrov commented Apr 24, 2026 •

edited

Loading

mmusgrov Apr 27, 2026 •

edited

Loading

mmusgrov Apr 28, 2026 •

edited

Loading

mmusgrov Apr 28, 2026 •

edited

Loading