az-pipelines-debug

Diagnose and resolve Azure GPU pipeline failures using bash, az CLI, and Python.

1|Updated Sep 30, 2023
One-click install
npx skills add https://github.com/speediedan/interpretune --skill az-pipelines-debug
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: az-pipelines-debug
Source: https://github.com/speediedan/interpretune/tree/main/.github/skills/az-pipelines-debug
Command: npx skills add https://github.com/speediedan/interpretune --skill az-pipelines-debug

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill helps diagnose and resolve issues with the interpretune self-hosted Azure GPU pipeline, ensuring smooth and efficient CI/CD for GPU-intensive tasks.

Core Features & Use Cases

  • Pipeline Debugging: Troubleshoot builds that are stuck, failing, or experiencing memory issues.
  • Approval Management: Approve pending pipeline runs directly from the command line.
  • Resource Monitoring: Track agent activity, worker logs, and resource utilization (memory, swap).
  • Failure Triage: Differentiate between queueing, infrastructure, and test-specific memory failures.
  • Local Reproduction: Facilitate recreating remote failures in a local environment for faster iteration.

Quick Start

Use the az-pipelines-debug skill to check the status of build ID 12345.

Frequently Asked Questions about az-pipelines-debug

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I debug Azure Pipelines builds that are stuck or failing on self-hosted GPU agents?

Debug self-hosted Azure Pipelines GPU build failures by triaging agent queues, inspecting worker dispatch logs, and isolating phase-split test failures to resolve stuck or failing runs.

Why does my Azure DevOps pipeline experience PAT-backed approval delays?

PAT-backed approval delays in Azure DevOps pipelines occur when authentication token propagation stalls; diagnosing the queue triage and approval routing resolves the pending execution state.

How can I narrow memory-aware test fixtures to resolve Azure GPU pipeline out-of-memory issues?

Narrow memory-aware test fixtures by monitoring agent resource utilization and swap activity, isolating phase-split test memory spikes to prevent GPU pipeline out-of-memory failures.

What's the best way to reproduce Azure GPU pipeline failures locally?

Reproduce Azure GPU pipeline failures locally by extracting worker logs and resource metrics, recreating the failing phase-split test environment for faster iteration and debugging.

Do I need Azure DevOps credentials and az CLI to troubleshoot self-hosted GPU pipeline issues?

Troubleshooting self-hosted Azure GPU pipeline issues requires bash, az CLI, curl, Python 3.10+, and configured Azure DevOps credentials to access queue data and worker logs.