When using "repeat escalation chain from beginning", the old payload is used instead of updated payload from the Alert Group.

[RajatGarga]

What went wrong?

What happened:

• We have escalation chain configured to repeat after waiting for 5 minutes.
• In those 5 minutes, the scope of our alert increased i.e from 3 instances impacted we may have 4 instances impacted.
• In the webhook configured with the corresponding escalation chain we get the payload with only the original 3 instances and the payload with 4 instances is never sent.

What did you expect to happen:

• When the scope of an alert increases, I want the escalation chain to use an updated payload with increased scope instead of the original scope of alert.

How do we reproduce it?

1. Used Prometheus push gateway to simulate this scenario.
2. Started with 1 metric sre_metric_1{test_label="val1"} 47 and then added more metrics by changing the test_label value

Grafana OnCall Version

1.3.112

Product Area

Alert Flow & Configuration

Grafana OnCall Platform?

Kubernetes

User's Browser?

No response

Anything else to add?

No response

[BhaRgav-MoRadiya]

We use outgoing webhooks to get notified when some alert group is created and we are seeing the similar issue. Alert groups are getting updated but webhooks that are getting fired are not up to date.

[prateekn]

Facing same issue with grouping of similar alerts