Overview

New tool that automatically analyzes CI test failures from ReportPortal and creates Jira tickets for failed test cases.

Key Features

• One ticket per test case: Creates a single Jira ticket for each unique test (e.g., OCP-11111)
• Multi-platform aggregation: If OCP-11111 fails on AWS, GCP, and Azure, all failures appear in ONE ticket with platform breakdown
• Comprehensive failure links: Lists every failure instance with direct links to ReportPortal logs
• Platform breakdown: Shows failure count per platform (AWS: 3 failures, GCP: 2 failures, etc.)
• Version tracking: Track failures across multiple OpenShift versions (4.19, 4.20, 4.21, 4.22)
• Team configuration: YAML-based per-team configuration (jobs, platforms, thresholds)
• Server-side filtering: Efficient API queries with status and name filtering
• Configurable CLI: Adjust page size, max pages, and lookback days

Repository

GitHub: https://github.com/rrasouli/ci-failure-tracker

How to Use

Installation

git clone https://github.com/rrasouli/ci-failure-tracker.git
cd ci-failure-tracker
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

Set Environment Variables

h1. Required for all operations
export REPORTPORTAL_API_TOKEN="your-reportportal-token"

h1. Required only for creating tickets (not needed for --dry-run)
export JIRA_USER="your-jira-username-or-email"
export JIRA_API_TOKEN="your-jira-api-token"

Get your Jira API token from: https://id.atlassian.com/manage-profile/security/api-tokens

Usage Examples

h1. Dry run - analyze failures without creating tickets
./ci_failure_tracker.py --team winc --dry-run --days 7

h1. Create tickets for failures in last 7 days
./ci_failure_tracker.py --team winc --days 7

h1. Fetch more launches if some are missing
./ci_failure_tracker.py --team winc --dry-run --days 7 --max-pages 20

h1. Show all available options
./ci_failure_tracker.py --help

Configuration

Each team has a YAML configuration file in the teams/ directory:

• teams/winc.yaml - WINC team configuration (example)
• teams/template.yaml - Template for creating new team configs

To add your team:

1. Copy teams/template.yaml to teams/yourteam.yaml
2. Update the configuration with your team's job patterns, versions, and platforms
3. Run the tool with --team yourteam

Benefits

• Saves time: No more manual ticket creation for recurring CI failures
• Better visibility: See all platform failures in one place
• Consistent format: All tickets follow the same structure with links to logs
• Configurable: Each team can customize thresholds, versions, and platforms
• Reduces noise: Only creates tickets when failures exceed threshold

Documentation

Full documentation: https://github.com/rrasouli/ci-failure-tracker/blob/master/README.md

Next Steps

• Try the tool with --dry-run to see what tickets would be created
• Review the output and adjust thresholds in your team config
• Run without --dry-run to create actual tickets
• Set up periodic runs (cron job or Prow periodic) for continuous tracking

Support

For questions or issues:

• File an issue: https://github.com/rrasouli/ci-failure-tracker/issues
• Contact: Windows Containers QE team (#windows-containers)