Closed
Description
Is your feature request related to a problem? Please describe.
In our current helm chart deployment, we provide overrides for some of our sidecar containers:
sidecars:
provisioner:
resources:
requests:
cpu: 50m
memory: 64Mi
limits:
memory: 512Mi
attacher:
resources:
requests:
cpu: 20m
livenessProbe:
resources:
requests:
cpu: 20m
memory: 32Mi
limits:
memory: 256Mi
However the current EBS add-on configuration schema only allows for setting controller & node resource overrides. The others only expose additionalArgs:
"sidecars": {
"additionalProperties": false,
"properties": {
"attacher": {
"additionalProperties": false,
"properties": {
"additionalArgs": {
"default": [],
"description": "Additional arguments passed to the attacher container",
"items": {
"type": "string"
},
"type": "array"
}
},
"type": "object"
},
"livenessProbe": {
"additionalProperties": false,
"properties": {
"additionalArgs": {
"default": [],
"description": "Additional arguments passed to the livenessprobe container",
"items": {
"type": "string"
},
"type": "array"
}
},
"type": "object"
},
"provisioner": {
"additionalProperties": false,
"properties": {
"additionalArgs": {
"default": [],
"description": "Additional arguments passed to the provisioner container",
"items": {
"type": "string"
},
"type": "array"
}
},
"type": "object"
}
Describe the solution you'd like in detail
Please add resource properties to each of the sidecar containers so we can manage the resource requests & limits for these deployments.
Describe alternatives you've considered
We're sticking with Helm charts for now, removing these resources causes issues when our clusters take on heavy workloads.
Additional context
N/A