Problem Statement
For transient failures, which retry policy is recommended?
Explanation
Exponential backoff reduces coordinated spikes; jitter spreads attempts to avoid thundering herds. A cap limits tail latency and hands off to a dead-letter flow when needed.
Always log retry cause and decision for observability.
Code Solution
SolutionRead Only
delay = base * 2^n + rand(0..jitter)
Practice Sets
This question appears in the following practice sets: