Hi! We just started using Grafana OnCall and it really feels like home 
One of the problems we have been facing is that we manage to hit the limit for alert_groups per integration which is really strange because we only have 3 alert groups on that specific integration. Or is it possible that “OnCall alert groups” vs. “Our integration (other grafana)” alert groups isn’t the same?
On this specific instance the alerts is configured like this:
Folder_name > Eval_group > 3 x Alert_groups (latency, diff, error)
In my head this would count as 3 groups, but then I read somewhere that it count groups at a label level. So, I guess it would be something like this:
latency → 10 labels = 10 groups
3 alert_groups x 10 labels = 30 alert_groups
And still we manage to hit 100 groups, with alarms firing from 4 different labels.
Please explain to me how this is working?