Rename Service to Endpoint (#192)
* Add clarifications in comments * #191: Rename Service to Endpoint
This commit is contained in:
@ -1,11 +1,11 @@
|
||||
# PagerDuty + Gatus Integration Benefits
|
||||
- Notify on-call responders based on alerts sent from Gatus.
|
||||
- Incidents will automatically resolve in PagerDuty when the service that caused the incident in Gatus returns to a healthy state.
|
||||
- Incidents will automatically resolve in PagerDuty when the endpoint that caused the incident in Gatus returns to a healthy state.
|
||||
|
||||
|
||||
# How it Works
|
||||
- Services that do not meet the user-specified conditions and that are configured with alerts of type `pagerduty` will trigger a new incident on the corresponding PagerDuty service when the alert's defined `failure-threshold` has been reached.
|
||||
- Once the unhealthy services have returned to a healthy state for the number of executions defined in `success-threshold`, the previously triggered incident will be automatically resolved.
|
||||
- Endpoints that do not meet the user-specified conditions and that are configured with alerts of type `pagerduty` will trigger a new incident on the corresponding PagerDuty service when the alert's defined `failure-threshold` has been reached.
|
||||
- Once the unhealthy endpoints have returned to a healthy state for the number of executions defined in `success-threshold`, the previously triggered incident will be automatically resolved.
|
||||
|
||||
|
||||
# Requirements
|
||||
@ -36,9 +36,9 @@ alerting:
|
||||
pagerduty:
|
||||
integration-key: "********************************"
|
||||
```
|
||||
You can now add alerts of type `pagerduty` in the services you've defined, like so:
|
||||
You can now add alerts of type `pagerduty` in the endpoint you've defined, like so:
|
||||
```yaml
|
||||
services:
|
||||
endpoints:
|
||||
- name: website
|
||||
interval: 30s
|
||||
url: "https://twin.sh/health"
|
||||
@ -56,12 +56,12 @@ services:
|
||||
```
|
||||
|
||||
The sample above will do the following:
|
||||
- Send a request to the `https://twin.sh/health` (`services[].url`) specified every **30s** (`services[].interval`)
|
||||
- Evaluate the conditions to determine whether the service is "healthy" or not
|
||||
- **If all conditions are not met 3 (`services[].alerts[].failure-threshold`) times in a row**: Gatus will create a new incident
|
||||
- **If, after an incident has been triggered, all conditions are met 5 (`services[].alerts[].success-threshold`) times in a row _AND_ `services[].alerts[].send-on-resolved` is set to `true`**: Gatus will resolve the triggered incident
|
||||
- Send a request to the `https://twin.sh/health` (`endpoints[].url`) specified every **30s** (`endpoints[].interval`)
|
||||
- Evaluate the conditions to determine whether the endpoint is "healthy" or not
|
||||
- **If all conditions are not met 3 (`endpoints[].alerts[].failure-threshold`) times in a row**: Gatus will create a new incident
|
||||
- **If, after an incident has been triggered, all conditions are met 5 (`endpoints[].alerts[].success-threshold`) times in a row _AND_ `endpoints[].alerts[].send-on-resolved` is set to `true`**: Gatus will resolve the triggered incident
|
||||
|
||||
It is highly recommended to set `services[].alerts[].send-on-resolved` to true for alerts of type `pagerduty`.
|
||||
It is highly recommended to set `endpoints[].alerts[].send-on-resolved` to true for alerts of type `pagerduty`.
|
||||
|
||||
|
||||
# How to Uninstall
|
||||
|
Reference in New Issue
Block a user