Struct DbMaintenanceTask 

pub struct DbMaintenanceTask {
    pub notify_tn: Option<TnId>,
}

Meta-database maintenance. Scheduled via cron at process start, and on demand from POST /api/admin/db-maintenance.

Fields

notify_tn: Option<TnId>

Tenant to push the outcome to when the sweep ends. None — the nightly run — stays silent: nobody asked for it, so nobody is waiting.

Not serialized, so it is not part of the task’s stored parameters. The sweep is node-wide; which tenant happened to press the button is a property of this request, not of the work. Serialized, two tenants triggering the same core.db_maintenance:manual key would produce two different parameter strings and send the scheduler down its “parameters changed” path. Identical parameters take the “already exists” branch and hand the second caller the running task’s id.

The cost is that a task rebuilt from its persisted row after a restart notifies nobody. That is the right outcome anyway: the browser that asked is long gone, and the notification is fire-and-forget.

Trait Implementations

impl Debug for DbMaintenanceTask

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for DbMaintenanceTask

fn default() -> DbMaintenanceTask

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for DbMaintenanceTask

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for DbMaintenanceTask

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl Task<Arc<AppState>> for DbMaintenanceTask

fn kind() -> &'static str

fn kind_of(&self) -> &'static str

fn build(_id: TaskId, ctx: &str) -> ClResult<Arc<dyn Task<App>>>

fn serialize(&self) -> String

fn run<'life0, 'life1, 'async_trait>( &'life0 self, app: &'life1 App, ) -> Pin<Box<dyn Future<Output = ClResult<()>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

fn on_failed<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, _state: &'life1 S, _attempts: u16, _last_error: &'life2 str, ) -> Pin<Box<dyn Future<Output = ()> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Called when the task transitions to Failed after exhausting retries (or on the very first failure when no retry policy is set). Lets the task perform irreversible cleanup — e.g. mark a related domain row as permanently failed — that should not happen on retryable failures. Default: no-op.

fn on_attempt_failed<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, _state: &'life1 S, _attempt: u16, _last_error: &'life2 str, ) -> Pin<Box<dyn Future<Output = ()> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Called after a failure that will be retried, with the zero-based index of the attempt that just failed. Lets a task tell the user once, on the first failure, instead of once per retry. Default: no-op.