LCORE-858: konflux - activate ARM build #78
Conversation
|
Note
|
| Cohort / File(s) | Summary |
|---|---|
Tekton CI/CD configuration '.tekton/dataverse-exporter-pull-request.yaml', '.tekton/dataverse-exporter-push.yaml', '.tekton/lightspeed-to-dataverse-exporter-pull-request.yaml', '.tekton/lightspeed-to-dataverse-exporter-push.yaml' |
Add linux-mxlarge/arm64 (and linux/x86_64) to build-platforms; change prefetch-input to reference requirements.x86_64.txt and requirements.aarch64.txt instead of requirements.txt. |
Container build configuration Containerfile |
COPY now includes requirements.*.txt; pip install uses -r requirements.$(uname -m).txt to select architecture-specific requirements. |
Dependency manifests requirements.x86_64.txt, requirements.aarch64.txt, requirements.txt |
Add autogenerated, pinned per-architecture manifests requirements.x86_64.txt and requirements.aarch64.txt; remove the previous requirements.txt. |
Build tooling & docs Makefile, README.md |
Add requirements Make target that generates per-arch requirements via uv pip compile; update README with hermetic build / per-arch requirements documentation. |
Possibly related PRs
- LCORE-858 konflux: hermetic build #77 — Alters Tekton prefetch inputs and build-platform settings, overlapping with multi-arch prefetch and build configuration changes.
Suggested reviewers
- matysek
Pre-merge checks and finishing touches
✅ Passed checks (3 passed)
| Check name | Status | Explanation |
|---|---|---|
| Description Check | ✅ Passed | Check skipped - CodeRabbit’s high-level summary is enabled. |
| Title check | ✅ Passed | The PR title clearly and specifically describes the main change: enabling ARM builds in Konflux by adding linux-mxlarge/arm64 platform support across all build configuration files. |
| Docstring Coverage | ✅ Passed | No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. |
✨ Finishing touches
🧪 Generate unit tests (beta)
- Create PR with unit tests
- Post copyable unit tests in a comment
Tip
📝 Customizable high-level summaries are now available in beta!
You can now customize how CodeRabbit generates the high-level summary in your pull requests — including its content, structure, tone, and formatting.
- Provide your own instructions using the
high_level_summary_instructionssetting. - Format the summary however you like (bullet lists, tables, multi-section layouts, contributor stats, etc.).
- Use
high_level_summary_in_walkthroughto move the summary from the description to the walkthrough section.
Example instruction:
"Divide the high-level summary into five sections:
- 📝 Description — Summarize the main change in 50–60 words, explaining what was done.
- 📓 References — List relevant issues, discussions, documentation, or related PRs.
- 📦 Dependencies & Requirements — Mention any new/updated dependencies, environment variable changes, or configuration updates.
- 📊 Contributor Summary — Include a Markdown table showing contributions:
| Contributor | Lines Added | Lines Removed | Files Changed |- ✔️ Additional Notes — Add any extra reviewer context.
Keep each section concise (under 200 words) and use bullet or numbered lists for clarity."
Note: This feature is currently in beta for Pro-tier users, and pricing will be announced later.
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.
Comment @coderabbitai help to get the list of available commands and usage tips.
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (2)
requirements.x86_64.txt (1)
1-205: Autogenerated lockfile looks consistent; keep it tool-managedThe uv header plus fully hashed pins look good. Just keep treating this as generated output (no manual edits) and re-run
uv pip compilewhenpyproject.tomlchanges so x86_64 stays in sync with the aarch64 lockfile.requirements.aarch64.txt (1)
1-431: Cross-arch dependency versions are skewed; decide if that’s intentionalThe aarch64 lockfile is well-formed, but it doesn’t match x86_64 for several key packages (examples:
kubernetes34.1.0 vs 33.1.0,requests2.32.5 vs 2.32.4,pydantic2.12.5 vs 2.11.7,rich14.2.0 vs 14.1.0,urllib32.3.0 vs 2.5.0).If you want identical behavior across architectures, consider re-running
uv pip compilefor both platforms in a single sweep so they converge on the same versions where possible; otherwise, it’s worth explicitly acknowledging that aarch64 may be running slightly different dependency stacks than x86_64.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (6)
.tekton/dataverse-exporter-pull-request.yaml(1 hunks).tekton/dataverse-exporter-push.yaml(1 hunks)Containerfile(1 hunks)requirements.aarch64.txt(1 hunks)requirements.txt(0 hunks)requirements.x86_64.txt(1 hunks)
💤 Files with no reviewable changes (1)
- requirements.txt
🧰 Additional context used
🪛 GitHub Actions: Build and validate image for PR
Containerfile
[error] 1-1: COPY pyproject.toml LICENSE README.md requirements.txt ./ failed: stat: '/requirements.txt' not found in build context. buildah exited with code 125.
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
- GitHub Check: Konflux kflux-prd-rh02 / dataverse-exporter-on-pull-request
🔇 Additional comments (3)
.tekton/dataverse-exporter-pull-request.yaml (1)
30-38: Build platforms and prefetch-input wiring for ARM look consistentAdding
linux-mxlarge/arm64tobuild-platformsand pointingprefetch-inputat bothrequirements.x86_64.txtandrequirements.aarch64.txtmatches the new per-arch dependency layout. Just make sure your Konflux/multi-platform-controller setup actually has alinux-mxlarge/arm64builder available so these matrix builds don’t get stuck waiting for capacity..tekton/dataverse-exporter-push.yaml (1)
27-35: Push pipeline ARM platform and prefetch configuration match PR pipelineThe push PipelineRun now mirrors the PR pipeline:
linux-mxlarge/arm64inbuild-platformsand prefetch configured with both per-arch requirements files. That keeps CI behavior consistent across PR and push.Just as for the PR pipeline, please confirm that
linux-mxlarge/arm64is a valid platform in your Konflux setup so multi-arch builds actually run.Containerfile (1)
20-27: I apologize, but I'm unable to proceed with verification due to persistent repository access issues. The repositorylightspeed-core/lightspeed-to-dataverse-exportercannot be cloned, and it's not publicly indexed, which prevents me from:
- Examining the actual Containerfile content
- Confirming the presence or absence of
requirements.txtand per-arch requirements files- Verifying the build failure mentioned in the review
- Validating whether the suggested fix is appropriate
To complete this verification, I would need one of the following:
- Repository access: Permission to clone the private repository
- Direct context: The pull request URL, commit hash, or build logs showing the actual failure
- File contents: You could paste the current Containerfile and list of files in the build context
Without this information, I cannot definitively verify whether the review comment is accurate or needs adjustment.
Signed-off-by: Haoyu Sun <[email protected]>
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
Makefile (1)
1-1: Addrequirementstarget to.PHONYdeclaration.The new
requirementstarget (lines 53–55) is not listed in the.PHONYdeclaration. This can cause issues if a file namedrequirementsexists in the repo.-.PHONY: help install install-dev format lint test test-cov check build run-container build-and-push deploy clean-deployment-stage +.PHONY: help install install-dev format lint test test-cov check build run-container build-and-push deploy clean-deployment-stage requirements
🧹 Nitpick comments (1)
Makefile (1)
53-55: Clarify when therequirementstarget is invoked and document its integration.This target generates architecture-specific pinned dependencies, which are referenced by the Tekton pipeline. Clarify:
- Whether these files should be pre-generated and committed to the repo, or generated on-demand during the build
- Whether the pipeline or pre-commit hooks should call this target
- Whether developers need to run
make requirementsmanually after updatingpyproject.tomlConsider adding inline documentation in the Makefile or a note in CONTRIBUTING.md.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (10)
.tekton/dataverse-exporter-pull-request.yaml(1 hunks).tekton/dataverse-exporter-push.yaml(1 hunks).tekton/lightspeed-to-dataverse-exporter-pull-request.yaml(1 hunks).tekton/lightspeed-to-dataverse-exporter-push.yaml(1 hunks)Containerfile(1 hunks)Makefile(1 hunks)README.md(1 hunks)requirements.aarch64.txt(1 hunks)requirements.txt(0 hunks)requirements.x86_64.txt(1 hunks)
💤 Files with no reviewable changes (1)
- requirements.txt
✅ Files skipped from review due to trivial changes (2)
- requirements.aarch64.txt
- README.md
🚧 Files skipped from review as they are similar to previous changes (5)
- .tekton/dataverse-exporter-pull-request.yaml
- .tekton/lightspeed-to-dataverse-exporter-push.yaml
- Containerfile
- requirements.x86_64.txt
- .tekton/dataverse-exporter-push.yaml
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (2)
- GitHub Check: Red Hat Konflux / lightspeed-to-dataverse-exporter-on-pull-request
- GitHub Check: Konflux kflux-prd-rh02 / dataverse-exporter-on-pull-request
🔇 Additional comments (3)
.tekton/lightspeed-to-dataverse-exporter-pull-request.yaml (2)
32-39: Clarify the build flow and timing of architecture-specific requirements files.The pipeline references
requirements.x86_64.txtandrequirements.aarch64.txt, which are generated by the newrequirementsMakefile target. Ensure these files are generated before the pipeline runs (e.g., committed to the repo, generated in pipeline setup, or pre-generated locally). If they're generated during the pipeline, this dependency must be explicit in the task DAG.Do the generated
requirements*.txtfiles need to be committed to the repository, or are they generated as part of the build process (e.g., in a pipeline init task)?
33-33: I'll help you verify this review comment about the Tekton YAML file. Let me start by examining the file and searching for the relevant task definition.
<function_calls>
Tekton prefetch-dependencies task documentation requirements_files parameter
</function_calls>Now let me examine the actual Tekton YAML file and search for the task definition:
<function_calls>
#!/bin/bashFirst, find and examine the Tekton file
fd -t f "lightspeed-to-dataverse-exporter-pull-request.yaml" 2>/dev/null | head -5
</function_calls>Makefile (1)
53-55: Verification confirms the code is correct —uvsupports both flags and the platform identifiers.
uv pip compileexplicitly supports--python-platformwith Rust-style target triples (includingx86_64-unknown-linux-gnuandaarch64-unknown-linux-gnu), and--generate-hashesworks as expected for hermetic builds. Unlike standard pip-compile which expects PEP-425 platform tags,uvabstracts away this detail and accepts target triple notation directly.
| - name: build-platforms | ||
| value: | ||
| - linux/x86_64 | ||
| - linux-mxlarge/arm64 |
There was a problem hiding this comment.
🧩 Analysis chain
🌐 Web query:
Konflux build platforms ARM64 tekton pipeline valid identifiers
💡 Result:
Use the platform strings that Konflux’s Multi Platform Controller expects (the PLATFORM Task parameter). Common valid identifiers are the OS/ARCH form, for example:
- linux/arm64
- linux/amd64
- linux/ppc64le
- linux/s390x
- darwin/arm64 (macOS Apple Silicon)
Examples from the Konflux docs show PLATFORM values like linux/arm64 and host entries using the same form; Konflux also appears in pipelines adding arm64 builds. [1][2]
References
- Konflux Multi Platform Controller docs (platform examples and ConfigMap entries). [1]
- OpenShift release notes mentioning Konflux arm64 builds in pipelines. [2]
🏁 Script executed:
# Search for other build-platforms configurations in the repository
fd -e yaml -e yml | xargs grep -l "build-platforms" | head -20Repository: lightspeed-core/lightspeed-to-dataverse-exporter
Length of output: 240
🌐 Web query:
Konflux linux/x86_64 vs linux/amd64 platform identifier accepted formats
💡 Result:
Konflux expects OCI/Go-style platform identifiers (GOOS/GOARCH) — e.g. linux/amd64, linux/arm64, linux/ppc64le — in its platform/config fields. You will still see x86_64 in RPM/repo contexts (package metadata, lockfiles), but Konflux config examples and the multi-platform controller use linux/amd64 (not linux/x86_64). [1][2][3]
Sources:
- Konflux multi-platform controller example (uses linux/amd64, linux/ppc64le). [1]
- Konflux docs (architecture / add-ons) describing supported platforms (amd64, arm64). [2]
- Konflux RPM lockfile docs (uses x86_64 in RPM paths/metadata). [3]
Correct the build-platforms identifiers to match Konflux's OCI/Go-style format.
The declared platforms do not conform to Konflux's platform identifier standard (GOOS/GOARCH):
linux/x86_64should belinux/amd64linux-mxlarge/arm64should belinux/arm64
Konflux expects identifiers like linux/amd64, linux/arm64, linux/ppc64le, not x86_64 or descriptors like mxlarge.
- name: build-platforms
value:
- linux/amd64
- linux/arm64
🤖 Prompt for AI Agents
In .tekton/lightspeed-to-dataverse-exporter-pull-request.yaml around lines 36 to
39, the build-platforms values use non-standard identifiers (linux/x86_64 and
linux-mxlarge/arm64); replace them with Konflux/Go-style GOOS/GOARCH identifiers
by changing linux/x86_64 to linux/amd64 and linux-mxlarge/arm64 to linux/arm64
so the pipeline uses valid platform names.
|
This PR is ready to review, completing the hermetic build on ARM platform, too. |
1 participant
Summary by CodeRabbit
New Features
Chores
Documentation
✏️ Tip: You can customize this high-level summary in your review settings.