Dependency latency
Simulate unreliable network conditions when communicating with dependencies by adding latency to API calls. Test whether your users are affected when dependency response times degrade.
Latency
Dependencies
6 minutes
How this Scenario works
This Scenario adds increasing levels of network latency to the connections between your services and their dependencies, simulating degraded network conditions on the path to external services, APIs, and databases.
Why run this Scenario?
- Validate that your services handle slow dependencies without cascading failures to upstream callers.
- Test timeout, retry, and circuit breaker configurations for dependency communication.
- Identify dependencies where latency sensitivity is highest and could cause the most user-facing impact.
- Verify that dependency monitoring detects increased latency and triggers alerts before users are affected.
What to expect when you run it
When network latency to dependencies increases significantly, our services continue to respond within acceptable bounds and users experience minimal impact.
More Scenarios to try
Avoid downtime. Use Gremlin to turn failure into resilience.
Gremlin empowers you to proactively root out failure before it causes downtime. See how you can harness chaos to build resilient systems by requesting a demo of Gremlin.
.svg)