HI, hopefully I’ve not missed something easily Google-able
I’m trying to create a dashboard that will have a repeating row per AWS region. The region value is held in a custom variable. So something like us-east-1, us-west-2, eu-central-1. The all values setting is enabled. This works as expected.
Each of the rows have several panels, one per applications that we have. The panels are to show some metrics of a load balancer. To get the load balancer name, I have another variable which has a query calling out to CloudWatch. The query references the region var. So the intended result would be:
- us-east-1
– App A LB | App B LB - us-west-2
– App A LB | App B LB - eu-central-1
– App A LB | App B LB
The query looks like
dimension_values([[awsRegion]],AWS/ApplicationELB,RequestCountPerTarget,TargetGroup)
When the whole dashboard is per-region (so only displaying a single regions metrics), this all works fine. When using rows repeating per region, it looks like the query is only evaluated once and takes the first value of the region var. So the first row displays correctly but any other row fails because the graph query takes the region (correctly set) and the load balancer name (contains the value of the first regions load balancer). This is based from what I see in the query analyser.
Is there a way to have the query variable evaluated per-row/per-panel rather than what appears to be one time at the start?