Playbook

502/503/504 Debugger

Distinguish gateway timeouts from upstream failures with targeted checks.

Identify whether failures are 502, 503, or 504 and by which edge.

Compare upstream response times vs proxy timeouts.

Check error budget and rate limits for affected endpoints.

502: upstream failed or returned invalid response to proxy.

503: upstream unavailable, overloaded, or in maintenance.

504: upstream took too long; proxy timeout exceeded.

Scale upstream services or reduce load via rate limiting.

Increase timeouts temporarily while fixing root cause.

Rollback recent deploys or disable heavy features.

Inspect upstream timing

curl -s -o /dev/null -w "connect=%{time_connect} ttfb=%{time_starttransfer} total=%{time_total}
" https://api.example.com/health

Check upstream health

curl -i https://api.example.com/health

Nginx upstream errors

tail -n 200 /var/log/nginx/error.log | grep -E "upstream|timeout"