RunnerHealth.ts overview
Since v1.0.0
Exports Grouped by Category
Constructors
makeK8s
Signature
declare const makeK8s: (
options?: { readonly namespace?: string | undefined; readonly labelSelector?: string | undefined } | undefined
) => Effect.Effect<
{ readonly isAlive: (address: RunnerAddress) => Effect.Effect<boolean> },
never,
HttpClient.HttpClient | FileSystem.FileSystem
>
Since v1.0.0
makePing
Signature
declare const makePing: Effect.Effect<
{ readonly isAlive: (address: RunnerAddress) => Effect.Effect<boolean> },
never,
Scope.Scope | Runners.Runners
>
Since v1.0.0
layers
layerK8s
A layer which will check the Kubernetes API to see if a Runner is healthy.
The provided HttpClient will need to add the pod’s CA certificate to its trusted root certificates in order to communicate with the Kubernetes API.
The pod service account will also need to have permissions to list pods in order to use this layer.
Signature
declare const layerK8s: (
options?: { readonly namespace?: string | undefined; readonly labelSelector?: string | undefined } | undefined
) => Layer.Layer<RunnerHealth, never, HttpClient.HttpClient | FileSystem.FileSystem>
Since v1.0.0
layerNoop
A layer which will always consider a Runner healthy.
This is useful for testing.
Signature
declare const layerNoop: Layer.Layer<RunnerHealth, never, never>
Since v1.0.0
layerPing
A layer which will ping a Runner directly to check if it is healthy.
Signature
declare const layerPing: Layer.Layer<RunnerHealth, never, Runners.Runners>
Since v1.0.0
models
RunnerHealth (class)
Represents the service used to check if a Runner is healthy.
If a Runner is responsive, shards will not be re-assigned because the Runner may still be processing messages. If a Runner is not responsive, then its associated shards can and will be re-assigned to a different Runner.
Signature
declare class RunnerHealth
Since v1.0.0