Struct WatchHealth 

#[non_exhaustive]
pub struct WatchHealth {
    pub degraded: Vec<ProxyConfigSource>,
    pub has_live_notifications: bool,
    pub poll_interval: Option<Duration>,
    pub stopped: bool,
}

Liveness of change-notification routes (ProxyWatcher::health).

degraded: route is not delivering — set WatchOptions::poll_interval. is_frozen: nothing arrives on its own — no route and no poll, or a stopped thread. ProxyWatcher::poll_now still delivers until the thread stops.

Fields (Non-exhaustive)

Non-exhaustive structs could have additional fields added in future. Therefore, non-exhaustive structs cannot be constructed in external crates using the traditional Struct { .. } syntax; cannot be matched against without a wildcard ..; and struct update syntax will not work.

degraded: Vec<ProxyConfigSource>

Routes that could not be established, or that were established and later lost. An entry names the source a dead route stops reporting on, which is not always the store the reader opens; a source that is simply not configured here is nothing to watch rather than a failed watch, and is not listed. Do not expect every entry to appear in ProxyConfig::sources: notification routes are per store, so a lost kioslaverc watch is reported as Kioslaverc even on a machine whose ProxyType = 4 makes that store report KioslavercEnv. Only ever grows: a source that degrades stays listed for the life of the watcher, the same one-way shape as has_live_notifications.

has_live_notifications: bool

At least one native notification route is still live. Starts from what was established at construction and only ever falls to false, once every route is gone. A backend thread that stopped is reported by stopped instead — that is not a statement about the routes.

poll_interval: Option<Duration>

Poll interval copied from WatchOptions::poll_interval at construction: what the caller asked for, not the floor a shorter request is raised to before any backend uses it. The floor is never reported, because this field says whether polling was asked for and stopped says whether the request is still met.

stopped: bool

The backend thread has stopped, so poll_interval no longer buys anything: the polling this type reports happens on that thread. The stream ends with None once drained, but a caller that only reads health would otherwise keep waiting.

Implementations

impl WatchHealth

pub fn is_fully_live(&self) -> bool

Every route is delivering: nothing degraded, at least one live native notification, and a backend thread still running to serve them. A stopped thread is never fully live, whatever the routes reported while it ran.

pub fn is_frozen(&self) -> bool

No live native notification and no WatchOptions::poll_interval — or a stopped thread, which takes the polling with it.

Trait Implementations

impl Clone for WatchHealth

fn clone(&self) -> WatchHealth

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for WatchHealth

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

Formats the value using the given formatter.

impl Eq for WatchHealth

impl PartialEq for WatchHealth

fn eq(&self, other: &WatchHealth) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for WatchHealth