GCP Troubleshooting & Root Cause Analysis
Debug GCP issues fast with systematic root cause analysis
What You Can Do
Systematically diagnose and resolve GCP performance, reliability, and cost issues by analyzing logs, metrics, and traces. You get structured troubleshooting workflows that pinpoint root causes—whether it's database latency, resource exhaustion, misconfigurations, or budget overruns—and deliver actionable remediation steps.
Features
Identify bottlenecks in compute, networking, and database layers using Cloud Monitoring metrics and trace data
Parse Cloud Logging to correlate errors, failures, and exceptions with specific GCP services and user impact
Analyze billing data and resource utilization to spot overspending, idle resources, and rightsizing opportunities
Walk through hypothesis testing, evidence collection, and elimination logic to find the actual cause—not symptoms
Review IAM, networking, quotas, and service settings for misconfigurations that trigger cascading failures
Cross-reference multiple signals (CPU, memory, latency, error rate) to isolate where problems originate
Get step-by-step fix recommendations ranked by impact and effort, with rollback strategies
Example Output
Example 1: Performance Issue
- Symptom: Cloud Run service responding slowly (p99 latency 8s, expected <500ms)
- Root Cause: Database connection pool exhausted; queries backing up
- Evidence: Cloud SQL CPU at 95%, active connections at max, query logs show 10+ second waits
- Fix: Increase connection pool size, optimize N+1 query patterns, add query caching
- Validation: Monitor metrics; confirm p99 drops to <600ms within 5 min
Example 2: Cost Spike
- Symptom: Monthly bill jumped 40% unexpectedly
- Root Cause: Unoptimized batch job running 24/7 on overpowered VMs
- Evidence: Compute Engine sustained use is 730 hours/month at e2-highmem-8
- Fix: Schedule job for peak hours only (6 hour window), downsize to e2-medium
- Savings: Est. $850/month
Example 3: Reliability Issue
- Symptom: Cloud Run health checks failing intermittently
- Root Cause: Insufficient memory; OOM kills during traffic spikes
- Evidence: Error Reporting shows OOM exceptions correlated with traffic bursts
- Fix: Increase memory allocation from 512MB to 1GB, set concurrency limit
- Validation: Health checks passing 100% for 24+ hours
What's Included
- SKILL.md: Complete troubleshooting framework with diagnostic decision trees, evidence collection templates, and remediation playbooks
- Diagnostic Checklist: Systematic workflow to eliminate causes: performance, reliability, cost, or configuration
- Evidence Collection Templates: Formatted queries for Cloud Logging, Cloud Monitoring dashboards, and metric exports
- Root Cause Analysis Matrix: Cross-reference symptoms to probable causes and validation steps
- Remediation Workflows: Ranked fixes by impact/effort for common GCP issues
- Validation Templates: Before/after metric comparisons to confirm fixes worked
- Incident Playbooks: Real-world runbooks for latency, errors, resource exhaustion, and budget overruns
Who It's For
- DevOps engineers — Respond to production incidents and resolve misconfigurations quickly
- Site reliability engineers (SREs) — Reduce MTTR by systematically isolating root causes
- Cloud architects — Audit existing deployments for performance and cost inefficiencies
- Backend developers — Debug application issues in production environments
- Platform engineers — Build runbooks and incident playbooks for your team
Best For
- Production incident response — Quickly isolate root causes and execute fixes during outages
- Performance optimization — Identify latency bottlenecks and improve user experience
- Cost reduction — Spot overspending patterns and rightsize resources
- Reliability improvements — Prevent recurring failures by fixing underlying misconfigurations
- Post-incident review — Analyze logs and metrics to understand what happened and why
You might also like
You'll systematically troubleshoot wireless connectivity problems by guiding structured data collection, analyzing root causes, and validating fixes. This skill walks you through diagnostic interviews, signal analysis, device troubleshooting, and performance optimization—turning vague WiFi complaints into actionable remediation steps with documented changes and recovery procedures.
by najicham
This skill analyzes your entire Azure infrastructure to identify underutilized resources, unused services, and cost inefficiencies. It generates detailed cost optimization recommendations including right-sizing strategies, reserved instance purchase analysis, and decommissioning plans with projected savings. Transform chaotic cloud spending into a lean, cost-optimized environment.
by najicham
You streamline your help desk workflow by using Claude to analyze support tickets, identify root causes, and generate professional responses. Claude helps you prioritize tickets, determine when escalation is needed, and craft clear documentation—saving time while ensuring consistent quality across your support team.
by najichamYou systematically isolate network problems by analyzing symptoms, logs, and infrastructure context through a structured diagnostic framework. Claude performs multi-layer analysis across physical, data link, network, and application layers to identify root causes and provide prioritized remediation steps. From intermittent connectivity drops to performance degradation to configuration drift, you get specific, actionable solutions with verification and rollback strategies.
by najicham
This skill turns your support ticket backlog into actionable resolution paths. You submit a ticket's description and environment details — Claude instantly categorizes severity, identifies root causes, suggests troubleshooting steps, and recommends escalation or self-service resolution. It surfaces knowledge base matches, estimates resolution time, and drafts professional customer communications.
by najicham
Analyze FDA guidance documents, structure regulatory narratives for IND/NDA/BLA submissions, and develop data-driven submission strategies aligned with current regulatory expectations. Claude identifies compliance gaps, predicts likely deficiency questions, and delivers submission-ready content formatted according to FDA conventions.
by najicham
You can systematically analyze adverse event databases to identify potential safety signals, apply statistical rigor to distinguish true signals from background noise, and generate regulatory-compliant safety assessments. Claude helps you evaluate signal strength, assess causality, and prepare documentation that meets ICH guidelines, FDA requirements, and international pharmacovigilance standards.
by najicham
You'll receive structured troubleshooting workflows that systematically diagnose VPN connectivity failures, identify performance bottlenecks, and evaluate your architecture against security best practices. Claude generates detailed diagnostic trees, configuration validation reports, and actionable remediation steps tailored to your specific VPN setup and observed symptoms.
by najicham