At 10 seconds a flash sale opens: 1,000 units of one item, and half of all checkouts are for it. Each checkout reads the stock count, charges the card (about 150 ms), then writes the count back minus one.
From 10 to 70 seconds, sell no more units than exist, while at least 99% of checkouts and 99% of product reads succeed.
SYSTEM BOUNDARY0 req/sno timeout · no retriespool 0/30Users0 req/sCLIENTAPI0/120 threadsSERVICEPayments0 calls/sTHIRD PARTYPostgres8 ms/queryDATA STORERedishit rate 92%CACHE
All components healthy.requestretrytrial call (half-open)failed
What changed
Turn on a fault or change a defence while the run plays. Its effect shows here: the 5 seconds before against 5 seconds after it settles.
Dashboard
Requests per second
1,000/s
Press Start
00:0001:00
Succeeded–
Failed–
Latency
1.0 s
Press Start
00:0001:00
p50–
p95–
p99–
Saturation
100%
Press Start
00:0001:00
API threads–
DB pool–
Postgres CPU–
Payments calls per second
500/s
Press Start
00:0001:00
Calls–
Failing–
Cache hit rate
100%
Press Start
00:0001:00
Hit rate–
Events and logs
Nothing yet. Faults, timeouts, retries, breaker changes, shedding, cache expiry and your own changes appear here, and as markers on the charts.
Slowest recent request
Press Start to see requests.
This is a simulation. Its numbers come from a model of a typical system, not from measuring real servers. The same link and seed always give the same run. How the Chaos Lab works.