Support Non-Query based KEDA Scalers #1790
Description
Describe the feature
KEDA has many scalers without a query parameter, including but not limited to cron, kubernetes-workload, and metrics-api. The current setPrimaryScaledObjectQueries function can only override the query scaler parameter.
Proposed solution
Flagger's autoscalerRef object needs to be more generic, for example, it can allow setting any trigger overrides with primaryScalerTriggerOverrides instead of primaryScalerQueries.
# Scaler reference
autoscalerRef:
apiVersion: keda.sh/v1alpha1
kind: ScaledObject
# ScaledObject targeting the canary deployment
name: podinfo-so
# Mapping between trigger names and the related parameters to use for the generated
# ScaledObject targeting the primary deployment. (Optional)
primaryScalerTriggerOverrides:
prom-trigger:
query: sum(rate(http_requests_total{ app="podinfo-primary" }[30s]))
cron-trigger:
desiredReplicas: "5"
# Overriding replica scaling configuration for the generated ScaledObject
# targeting the primary deployment. (Optional)
primaryScalerReplicas:
minReplicas: 2
maxReplicas: 5Any alternatives you've considered?
Currently, I avoid using these scalers with Flagger.