FixOps
Play now
Web Advanced About 35 min +300 XP

Every Site Behind the Edge Returns 502

Customer sites proxied through edge-fra-07 answer 502 Bad Gateway for almost every request while the edge host's CPU sits at 100%.

Take this incident no signup All incidents

Briefing

Brightgate Edge fronts a few thousand customer sites from PoPs like edge-fra-07, and.

The same graph shape appeared on every PoP inside the same minute, and our own dashboard and status page ride the same edge, so customers cannot even see that we know.

edgeproxy is still running and its liveness check still answers, yet none of its workers has spare capacity.

Half the channel is sure this is a layer 7 flood, and the origin health checks look slow enough to blame the customers' servers instead.

Inspired by a real outage: Details of the Cloudflare outage on July 2, 2019 (Cloudflare, 2019). Names and details are fictionalized.

The fix is yours to find: hints, the recovery checks and the debrief unlock inside the incident.

How it plays

  1. 01

    Get paged

    The alert fires and the clock starts. Read the page and the briefing.

  2. 02

    Investigate

    Work in a simulated shell with realistic output: logs, configs, services.

  3. 03

    Fix it

    Change the system the way you would in production. Hints are there if you get stuck.

  4. 04

    Prove it

    Automated checks verify the recovery, then the debrief explains what happened.

All 61 incidents

Your pager is ready.

Free, instant, and it works on your phone. No signup: start as a guest and save your progress later.

Take this incident no signup Get it on Google Play