Skip to content
← Health for Laravel 3.0.0 View on GitHub

Caching

Caching

Health for Laravel caches check reports to reduce overhead from frequent probe requests.

Configuration

'cache' => [
    'enabled' => env('HEALTH_CACHE_ENABLED', true),
    'ttl'     => env('HEALTH_CACHE_TTL', 10),
    'store'   => null,
],
Option Default Description
enabled true Enable/disable report caching
ttl 10 Cache duration in seconds (integer)
store null Cache store (null = default store)

Environment Variables

HEALTH_CACHE_ENABLED=true
HEALTH_CACHE_TTL=10

When to Adjust TTL

  • Kubernetes probes polling every 10–15s: a TTL of 10s is appropriate
  • High-frequency monitoring: lower the TTL to 5s for fresher data
  • Expensive checks: raise the TTL to 30–60s to reduce load
  • Development and tests: set HEALTH_CACHE_ENABLED=false for instant feedback

What Is Cached

The runner caches the HealthReport for each endpoint type under the key health:report:<endpoint> (for example health:report:readiness). Every request that runs that endpoint's checks within the TTL window gets the same report — including the status, Prometheus and dashboard endpoints, which run the liveness and readiness checks, and the health:check command.

System metrics (CPU, memory, disk, network) are not cached; they are read on every request.

If the cache store is unavailable, the runner runs the checks and skips caching instead of failing. A broken cache backend therefore doesn't break the health endpoints — use CacheCheck to detect it.

If you replace the runner with your own implementation, caching is up to you. A decorator that delegates to HealthCheckRunner keeps it.

← Previous Architecture Next → System Metrics