Resource Limits for a Namespace

Resource Limits for a Namespace

We will create a new namespace and configure another stress deployment to run within. When set stress should not be able to use the previous amount of resources.

Begin by creating a new namespace called ns-with-limit and verify it exists.

Create Namespace

kubectl create namespace ns-with-limit

namespace/ns-with-limit created

Create LimitRange

Create a YAML file which limits CPU and memory usage. The kind to use is LimitRange.

vi /home/student/ns-with-limit-range.yaml

File: /home/student/ns-with-limit-range.yaml

apiVersion: v1
kind: LimitRange
metadata:
  name: ns-with-limit-range
spec:
  limits:
  - default:
      cpu: 1
      memory: 500Mi
    defaultRequest:
      cpu: 0.5
      memory: 100Mi
    type: Container

Create the LimitRange object and assign it to the newly created namespace low-usage-limit. You can use --namespace or -n to declare the namespace.

kubectl apply --namespace=ns-with-limit -f ns-with-limit-range.yaml

limitrange/ns-with-limit-range created

Verify that it works. Remember that every command needs a namespace and context to work. Defaults are used if not provided.

kubectl get LimitRange -n ns-with-limit

NAME                  CREATED AT
ns-with-limit-range   2019-10-30T09:39:37Z

Create an unlimited namespace:

kubectl create ns ns-without-limit

namespace/ns-without-limit created

Create a new deployment in the namespace.

Create a deployment inside namespace ns-without-limit

kubectl -n ns-without-limit create deployment \
	stress-without-limits \
	--image r.deso.tech/whoami/whoami

deployment.apps/stress-without-limits created

kubectl get deploy -n ns-without-limit -o json | jq '.items[].spec.template.spec.containers[]'

{
  "image": "r.deso.tech/whoami/whoami",
  "imagePullPolicy": "Always",
  "name": "whoami",
  "resources": {},
  "terminationMessagePath": "/dev/termination-log",
  "terminationMessagePolicy": "File"
}

As you can see, no limit was applied to the deployment or the pod:

kubectl get pod -n ns-without-limit -o json | jq '.items[].spec.containers[].resources'

{}

Create a deployment inside namespace ns-with-limit

kubectl -n ns-with-limit create deployment \
	stress-with-limits   \
	--image r.deso.tech/whoami/whoami

deployment.apps/stress-with-limits created

Check the limits on the deployment:

kubectl get deploy -n ns-with-limit -o json | jq '.items[].spec.template.spec.containers[]'

{
  "image": "r.deso.tech/whoami/whoami",
  "imagePullPolicy": "Always",
  "name": "whoami",
  "resources": {},
  "terminationMessagePath": "/dev/termination-log",
  "terminationMessagePolicy": "File"
}

Check the limit on the pod:

kubectl get pod -n ns-with-limit -o json | jq '.items[].spec.containers[].resources'

{
  "limits": {
    "cpu": "1",
    "memory": "500Mi"
  },
  "requests": {
    "cpu": "500m",
    "memory": "100Mi"
  }
}

Clean up

kubectl  delete ns ns-without-limit ns-with-limit

namespace "ns-without-limit" deleted
namespace "ns-with-limit" deleted