troubleshoot-uwsgi

What problem does it solve?

This skill solves the problem of quickly triaging uWSGI incidents—such as worker exhaustion, harakiri storms, memory creep, stuck workers, and reload failures—using the same Netdata operator playbook logic but driven through MCP queries.

Core Features & Use Cases

• Signal-driven diagnostic tree: Guides a coding agent through an 8-domain triage flow aligned to Netdata’s uWSGI operator playbook failure patterns, mapping each symptom to the appropriate rule file and next MCP query.
• MCP-based verification: Actively queries Netdata via MCP to validate which uWSGI signal contexts are anomalous over a relevant time window, then re-checks after remediation to confirm the fix “holds.”
• Remediation-focused remediation loop: Recommends corrective actions implied by the playbook/rule file and enforces before/after signal re-verification to avoid false fixes.

Use cases: On-call incident response for a misbehaving uWSGI service, debugging elevated errors/latency/saturation, and investigating unexpected restarts or reload-related failures after deployments.

Quick Start

Use the troubleshoot-uwsgi skill when you ask: "My uWSGI workers are failing—can you diagnose the root cause using Netdata and MCP and tell me what remediation to try."