Monitors worker health and crash patterns.
Legacy Optional Module
Snakepit does not call this module internally. It remains available for
compatibility and may be removed in v0.16.0 or later.
Prefer using worker lifecycle telemetry and host-application health policies for new integrations.
Tracks crashes within a rolling window and determines overall pool health. Can be used to trigger circuit breaker actions or alerting.
Usage
{:ok, hm} = HealthMonitor.start_link(
name: :my_pool_health,
pool: :default,
max_crashes: 10,
crash_window_ms: 60_000
)
HealthMonitor.record_crash(hm, "worker_1", %{reason: :segfault})
if HealthMonitor.healthy?(hm) do
# Pool is healthy
else
# Too many crashes, consider action
endSummary
Functions
Returns a specification to start this module under a supervisor.
Returns whether the pool is considered healthy.
Records a worker crash.
Starts a health monitor.
Returns comprehensive health statistics.
Returns health status for a specific worker.
Types
@type t() :: %{ pool: atom(), workers: %{required(String.t()) => worker_stats()}, crash_window_ms: pos_integer(), max_crashes: pos_integer(), total_crashes: non_neg_integer(), check_interval_ms: pos_integer(), check_timer: reference() | nil }
@type worker_stats() :: %{ crash_count: non_neg_integer(), last_crash_time: integer() | nil, crash_reasons: [term()] }
Functions
Returns a specification to start this module under a supervisor.
See Supervisor.
@spec healthy?(GenServer.server()) :: boolean()
Returns whether the pool is considered healthy.
@spec record_crash(GenServer.server(), String.t(), map()) :: :ok
Records a worker crash.
@spec start_link(keyword()) :: GenServer.on_start()
Starts a health monitor.
Options
:name- GenServer name (required):pool- Pool name to monitor (required):check_interval_ms- Health check interval (default: 30000):crash_window_ms- Rolling window for crash counting (default: 60000):max_crashes- Max crashes in window before unhealthy (default: 10)
@spec stats(GenServer.server()) :: map()
Returns comprehensive health statistics.
@spec worker_health(GenServer.server(), String.t()) :: map()
Returns health status for a specific worker.