Struct GCSharedRuntimeManager 

pub struct GCSharedRuntimeManager {
    inner: Arc<RwLock<GCRuntimeManager>>,
    health_check_tx: Arc<Sender<()>>,
}

Shared handle to GCRuntimeManager, cheap to clone.

Spawns a background health-check task on construction that periodically calls the /health endpoint on every connected runtime and removes any that fail to respond. The task can also be triggered immediately via GCSharedRuntimeManager::trigger_health_check.

Fields

inner: Arc<RwLock<GCRuntimeManager>>

health_check_tx: Arc<Sender<()>>

Send () to trigger an immediate health check outside of the normal interval.

Implementations

impl GCSharedRuntimeManager

pub fn new( runner_mode: GCRunnerMode, runner: Option<Arc<dyn RuntimeRunner>>, config: &GCConfigDatabase, ) -> GCSharedRuntimeManager

Creates a runtime manager and starts the background health-check task.

Pass Some(Arc::new(runner)) for process or container mode, or None to disable runner-based start/stop (the manager will still track runtimes that register themselves).

pub fn trigger_health_check(&self)

Trigger an immediate health check outside the normal 5-second interval.

If a check is already queued the call is silently ignored (channel capacity is 1).

fn spawn_health_check_task(&self, rx: Receiver<()>)

Spawns the background health-check loop.

pub async fn write(&self) -> RwLockWriteGuard<'_, GCRuntimeManager>

pub async fn read(&self) -> RwLockReadGuard<'_, GCRuntimeManager>

pub async fn owned_write(&self) -> OwnedRwLockWriteGuard<GCRuntimeManager>

Useful for sending across threads.

pub async fn owned_read(&self) -> OwnedRwLockReadGuard<GCRuntimeManager>

Useful for sending across threads.

pub async fn stop_all_managed_runtimes(&self)

Stops all runtimes managed by the runner and removes them from the connected pool

pub async fn start_runtime( &self, runtime_id: &str, config_db: &GCConfigDatabase, ) -> Result<(), StartRuntimeError>

pub async fn connected_runtimes(&self) -> HashSet<String>

pub async fn managed_runtimes(&self) -> HashSet<String>

Returns the IDs of runtimes currently managed by the runner This may return runtimes that are not connected (e.g. zombie runtimes)

pub async fn start_multiple_runtimes( &self, running_runtimes: HashMap<String, bool>, )

Starts multiple runtimes in parallel. running_runtimes is a list of tuples of runtime ID and whether the runtime should be started in host network mode.

pub async fn stop_runtime( &self, runtime_id: &str, ) -> Result<(), StopRuntimeError>

Trait Implementations

impl Clone for GCSharedRuntimeManager

fn clone(&self) -> GCSharedRuntimeManager

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.