Q: Your RDS instance is using 100% CPU. What do you do?
1. Identify the culprit — use RDS Performance Insights to find the top SQL queries consuming CPU.
#AWS #RDS & Databases #L2 #Cloud #Infrastructure
🎙️ Candidate Opening & Architectural Context
""In a previous role, our monitoring paged me for a similar incident across our AWS VPC infrastructure. When addressing this question, I walk the interviewer through our production incident runbook: isolating the blast radius, checking diagnostic logs and metrics, and applying a safe fix.""
Advertisement
🛠️ Production Runbook & Step-by-Step Resolution
1️⃣
Initial Diagnostics & Root Cause Analysis
- Identify the culprit — use RDS Performance Insights to find the top SQL queries consuming CPU.
- Look for slow queries — enable slow query log. Check for missing indexes.
- Check for lock contention — queries waiting for locks show high CPU wait.
- Scale vertically — stop the RDS instance, change instance class to a larger one (more CPU), restart.
2️⃣
Remediation & Permanent Safeguards
Execute the resolution runbook and verify workload health:
- Read replicas — offload read queries to a read replica. Add a read replica and point your app's read traffic there.
- Query optimization — add indexes for slow queries found in Performance Insights.
- Connection pool — if too many connections are opening/closing rapidly, it causes CPU overhead. Use RDS Proxy.
💡 The Senior SRE Gold Nugget (Key Architectural Takeaway)
"Pro-Tip: Identify the culprit — use RDS Performance Insights to find the top SQL queries consuming CPU.."
⚡ 60-Second Elevator Pitch Talking Points
- Identify the culprit — use RDS Performance Insights to find the top SQL queries consuming CPU.
- Look for slow queries — enable slow query log. Check for missing indexes.
- Check for lock contention — queries waiting for locks show high CPU wait.
Advertisement