HPA minReplicas vs PDB minAvailable check #202
Description
There is already a basic config for checking deployment replicas vs PDB.
https://github.com/open-policy-agent/gatekeeper-library/tree/master/library/general/poddisruptionbudget
But what happens if we add HPA to the mix. Instead of setting replicas in a deployment we do it using HPA.
In PDB setting minAvailable: 2 and in HPA setting minReplicas: 1 will potentially lock a node from draining when cordoned thanks to the PDB rule.
Below we can find a example that is okay. But lowering HPA minReplicas: 1 and we would get issues if the application didn't have any load.
apiVersion: autoscaling/v2beta2
kind: HorizontalPodAutoscaler
metadata:
name: app
spec:
minReplicas: 2
maxReplicas: 4
scaleTargetRef:
apiVersion: apps/v1
kind: Deployment
name: app
metrics:
- type: Resource
resource:
name: memory
target:
type: Utilization
averageUtilization: 80apiVersion: policy/v1
kind: PodDisruptionBudget
metadata:
name: app
spec:
minAvailable: 1
selector:
matchLabels:
app: appapiVersion: apps/v1
kind: Deployment
metadata:
name: app
labels:
app: app
spec:
selector:
matchLabels:
app: app
template:
metadata:
labels:
app: app
spec:
containers:
- name: app
image: nginx:latest