Struct BackgroundJobServer 

pub struct BackgroundJobServer { /* private fields */ }

Background job server that manages job processing

Implementations

impl BackgroundJobServer

pub fn new( config: ServerConfig, storage: Arc<dyn Storage>, worker_registry: Arc<WorkerRegistry>, ) -> Self

Create a new background job server

pub fn with_middleware(self, middleware: Vec<Arc<dyn JobMiddleware>>) -> Self

Replace the middleware stack that wraps worker.execute in every worker thread. Runs in registration order — the first entry is the outermost layer.

The default stack is [TracingMiddleware]; calling this replaces it entirely. Re-add TracingMiddleware yourself if you still want structured spans around every execution.

Must be called before BackgroundJobServer::start — changes made after a running server has spawned its worker threads won’t affect already-started processors.

pub fn with_state_change_hook(self, hook: StateChangeHook) -> Self

Install a state-change hook fired after every persisted job state transition driven by the processor. See StateChangeHook for semantics — the hook runs synchronously inside process_job, so keep it non-blocking.

The hook is cloned into every per-worker JobProcessor when BackgroundJobServer::start spawns workers, so callers must install it before start().

pub fn with_retry_policy( config: ServerConfig, storage: Arc<dyn Storage>, worker_registry: Arc<WorkerRegistry>, retry_policy: RetryPolicy, ) -> Self

Create a new background job server with custom retry policy

pub async fn start(&self) -> Result<()>

Start the background job server

pub async fn stop(&self) -> Result<()>

Stop the background job server.

Cancels the shutdown token so every worker drops out of its polling loop after finishing its current job, then waits up to config.shutdown_timeout for all tasks to join. Any task still running past the timeout is aborted — those jobs will need stale-processing recovery on next startup.

pub async fn is_running(&self) -> bool

Check if the server is running

pub fn config(&self) -> &ServerConfig

Get server configuration

pub async fn schedule_recurring( &self, id: impl Into<String>, cron: impl Into<String>, method: impl Into<String>, payload: Value, queue: impl Into<String>, ) -> Result<()>

Register (or update) a recurring job template.

id uniquely identifies this template — calling again with the same id replaces the previous definition. cron is a 6-field cron expression (second minute hour day month day-of-week) parsed by the cron crate. The template is stored via [Storage::upsert_recurring_job] and the running RecurringJobPoller will materialize it into a normal [Job] the next time next_run_at is in the past.

pub async fn remove_recurring(&self, id: &str) -> Result<bool>

Remove a recurring job template by id. Returns true if a row was deleted, false if no template with that id existed.