Alerting docs: Relocate Resolved stale alert instances section (#87965)

Relocate `Resolved stale alert instances` section
This commit is contained in:
Pepe Cano
2024-05-20 09:33:14 +02:00
committed by GitHub
parent 78d1b2a250
commit 54eab4db59
2 changed files with 6 additions and 6 deletions
@@ -228,12 +228,6 @@ If your evaluation returns an error, you can set the state on your alert rule to
| Alerting | Sets alert rule state to `Alerting`. The alert rule waits until the time set in the **For** field has finished before firing. |
| Ok | Sets alert rule state to `Normal`. |
### Resolve stale alert instances
An alert instance is considered stale if its dimension or series has disappeared from the query results entirely for two evaluation intervals.
Stale alert instances that are in the **Alerting**/**NoData**/**Error** states are automatically marked as **Resolved** and the grafana_state_reason annotation is added to the alert instance with the reason **MissingSeries**.
### Create alerts from panels
Create alerts from any panel type. This means you can reuse the queries in the panel and create alerts based on them.
@@ -44,6 +44,12 @@ Alert instances will be routed for [notifications][notifications] when they are
{{< figure src="/media/docs/alerting/alert-rule-evaluation-overview-statediagram-v2.png" max-width="750px" >}}
### Lifecycle of stale alert instances
An alert instance is considered stale if its dimension or series has disappeared from the query results entirely for two evaluation intervals.
Stale alert instances that are in the **Alerting**, **NoData**, or **Error** states transition to the **Normal** state as **Resolved**, and include the `grafana_state_reason` annotation with the value **MissingSeries**. They are routed for notifications like other resolved alert instances.
### Keep last state
The "Keep Last State" option helps mitigate temporary data source issues, preventing alerts from unintentionally firing, resolving, and re-firing.