When Redis goes down, does your app die?
The Problem: The App Dies There was an incident that caused our application to fail because our Redis cache went down. It was in the middle of moving to a larger compute instance to handle a higher load, and since we only had one replica at the time, the cache became completely unavailable. The issue is that Laravel, by default, assumes your cache is always there. If Redis dies, the standard…
The application experience suffered significantly when the Redis cache went down, forcing the system to default to using the database directly. To prevent this, the Resilient Cache-Aside approach was implemented. This involved wrapping cache operations in a try-catch block to handle exceptions gracefully, thereby avoiding a 500 Internal Server Error.
However, this solution did not address the issue of timeouts. The Circuit Breaker pattern was found to be a more effective solution. By tracking failures over time, the circuit opens when failures reach a certain threshold, causing the app to skip the cache and go directly to the database. This prevents timeouts and improves user experience.
A PHP library called Ganesha was used to implement the Circuit Breaker pattern, creating a CircuitBreakerService that manages the circuit state and executes primary or fallback operations based on the circuit's state.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.