-
Notifications
You must be signed in to change notification settings - Fork 110
Open
Description
Problem Description
The TrustyAI notebook build (jupyter-trustyai-ubi9-python-3.12) is experiencing significantly extended build times due to pandas==1.5.3 being built from source instead of using pre-built wheels.
Build Failure Details
- Build: jupyter-trustyai-ubi9-python-3.12
- Total build time: 19m 28s
- Failure point: Building pandas==1.5.3
- Platform: linux/amd64, Python 3.12
Context
This issue was discovered during PR #2145 implementation, which introduces uv pyproject.toml files across multiple notebooks. The TrustyAI notebook's pyproject.toml specifies pandas~=1.5.3 as a dependency.
Investigation Areas
- Wheel Availability: Check if pandas 1.5.3 wheels are available for Python 3.12 on linux/amd64
- Dependency Conflicts: Investigate if dependency conflicts are forcing source builds
- Index Configuration: Review uv index configuration in pyproject.toml for pandas resolution
- Version Compatibility: Assess if pandas 1.5.3 is appropriate for Python 3.12 environment
Potential Solutions
- Update pandas version to one with better wheel availability
- Add explicit index configuration for pandas wheels
- Investigate dependency pinning conflicts
- Consider platform-specific dependency specifications
Acceptance Criteria
- Identify root cause of pandas source build requirement
- Implement solution to use pre-built pandas wheels
- Verify TrustyAI notebook build completes in reasonable time (<5m)
- Document any version or configuration changes required
Context Links
- PR: RHAIENG-304: add uv pyproject.tomls for multiple notebooks and runtimes across Jupyter, RStudio, and VSCode #2145
- Build Log: https://github.com/opendatahub-io/notebooks/actions/runs/17268128656/job/49008591181?pr=2145#step:25:332
- Comment: RHAIENG-304: add uv pyproject.tomls for multiple notebooks and runtimes across Jupyter, RStudio, and VSCode #2145
Impact
Extended build times affect CI/CD pipeline performance and developer productivity. Source builds also indicate potential dependency resolution issues that could affect runtime stability.
Metadata
Metadata
Assignees
Labels
No labels
Type
Projects
Status
📋 Backlog