An order goes from the app to the API to the order service, which looks up stock. The app gives up after 1 second. The order service has 60 worker threads, waits up to 1.5 seconds per stock call, and retries a timed-out call twice. At 10 seconds the stock service browns out: half of its answers take 2 seconds.
From 15 to 60 seconds, answer at least 85% of orders within the app's 1 second.
Orders answered within 1 s ≥ 85.0%not measured yet
Service map
Call chain · pausedSelect a component to inspect
SYSTEM BOUNDARY1 swait1500ms×3Appbudget 1 sCLIENTAPIno deadlineSERVICEOrders60 threadsSERVICEStock20 msDATA STORE
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.
Service map
Call chain · pausedSelect a component to inspect
SYSTEM BOUNDARY1 swait1500ms×3Appbudget 1 sCLIENTAPIno deadlineSERVICEOrders60 threadsSERVICEStock20 msDATA STORE
SYSTEM BOUNDARY1 swait1500ms×3Appbudget 1 sCLIENTAPIno deadlineSERVICEOrders60 threadsSERVICEStock20 msDATA STORE
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
Orders per second
200/s
Press Start
00:0001:00
Answered in time–
Failed–
Latency
1.0 s
Press Start
00:0001:00
p50–
p99–
Order service threads
100%
Press Start
00:0001:00
Busy–
Wasted after the deadline–
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.