Xray help to monitor service health via health indicators.
The main concept is registering some callback functions which have to return Promise<any>. If the returned Promise
resolved than Xray assume that service is up and running correctly, otherwise if the Promise is rejected than service
is marked as down from some reason. Booth resolution and rejection result is saved and can accessed via status method.
npm install @vanioinformatika/appxray --savesetInterval(interval: number): thisSet pooling interval. Default interval value is 1000 milliseconds.
setDefaultIndicatorTimeout(timeout: number): thisSet default indicator timeout. If timeout is not specified for indicator than default timeout is applied during checking calling indicator. Default value is 1000 milliseconds.
setEventEmitter(eventEmitter: EventEmitter): thisSet event emitter to allow firing events. Available events are:
- xray.started - Fired after first service check executed.
- Event's callback sigbature is
(status: IndicatorStatus) => void.
- Event's callback sigbature is
- xray.indicator.success - Fire every time when checking service health for each indicator when it's callback
returned resolved promise.
- Event's callback signature is
(name: string, response: IndicatorResponse) => void.
- Event's callback signature is
- xray.indicator.error - Fire like xray.indicator.success but only when indicator's callback returned with rejected promise.
has(name: string): booleanCheck if xray has named indicator callback or not.
add(name: string, indicator: Indicator, options?: IndicatorOptions): thisAdd new indicator to xray with given name. During adding indicator it is possible to specify some options:
- timeout - Override default indicator timeout.
Note: Indicator with same name will be overwritten.
remove(name: string): booleanRemove named indicator from Xray. Return true if Xray had named indicator and it is successfully removed, otherwise
its return false.
start(): voidStart pooling indicators with given interval.
Note: It is possible to change timeout and add / remove indicators during pooling.
status(): IndicatorStatus
status(name: string): IndicatorResponse
status({detailed}: { detailed: true }): Map<string, IndicatorResponse>Via this method can access last indicator results.
Note: The
statusmethod returns only actual status. Previous status is always overwritten during checking.
status(): IndicatorStatusIf status is called without any argument then it will summarize indicators response and returns simple IndicatorStatus response.
If method is called before pooling is started than it will return UNKNOWN status, otherwise UP or DOWN base on
indicators response. If least one indicator is returned with error then status will result with DOWN otherwise UP.
status(name: string): IndicatorResponseIf status is called with single string argument, than it will act like getting information about single indicator. In this case method will return detailed response.
status({detailed}: { detailed: true }): Map<string, IndicatorResponse>The last overload accepts object with options. If detailed option is set to true than status is returned in Map
where keys are indicator names and values are corresponding detailed responses.
check(): Promise<Map<string, IndicatorResponse>>Execute every indicator, update status information base on indicators response. It will return promise with indicators results.