Basic Night Time Saving

One of the easiest ways and most common use cases for the Downscaler is using it for night time saving.

To achieve this configuration we will use relative timespans.

A good time to start saving resources at night could be from 6:00 PM to 8:00 AM during weekdays and on weekends. Depending on your use case you can apply the night time saving cluster-wide or on either namespace or workload level.

At the end of this guide you will also find how to manage exclusions if you want to exclude some workloads or entire namespaces from this configuration.

Cluster-Wide

In order to apply the wanted downtime cluster-wide the GoKubeDownscaler has to know what downtime to use.

This can be done by either using an environment variable or a cli argument for the GoKubeDownscaler.

Environment Variable

If you want to add a cluster-wide downtime with environment variables you can add them inside the values.yaml file.

There you have to add it to the configMap.extraConfig field:

values.yaml

# ...

configMap:
  name: go-kube-downscaler
  extraConfig: |
    DEFAULT_DOWNTIME: "Mon-Fri 18:00-08:00 UTC, Sat-Sun 00:00-24:00 UTC"

# ...

Cli Argument

If you want to use a cli argument for the cluster-wide downtime you can add them inside the values.yaml file.

There you have to add it to the arguments field:

values.yaml

# ...

arguments:
  - --default-downtime="Mon-Fri 18:00-08:00 UTC, Sat-Sun 00:00-24:00 UTC"
# ...

After setting the variable in one of the two ways all workloads in your cluster will be scaled down on every weekday from 6:00 PM to 8:00 AM and on weekends.

Namespace Level

To set this up for all workloads in a namespace you have to annotate the namespace with the wanted time span.

This can be done using a kubectl command or by editing the namespace's yaml manifest.

kubectl

Adding the downtime with kubectl is a single command:

kubectl annotate ns my-namespace downscaler/downtime="Mon-Fri 18:00-08:00 UTC, Sat-Sun 00:00-24:00 UTC"

Edit manifest

If you want to edit the annotations manually with a yaml manifest file or by running kubectl edit ns my-namespace you have to add the following:

my-namespace.yaml

apiVersion: v1
kind: Namespace
metadata:
  name: my-namespace
  annotations:
    downscaler/downtime: "Mon-Fri 18:00-08:00 UTC, Sat-Sun 00:00-24:00 UTC"

Now all workloads in my-namespace will be scaled down on every weekday from 6:00 PM to 8:00 AM and on weekends.

Workload Level

To set this up for individual workloads you have to set an annotation on each workload you want to scale down.

This can also be done by either using a kubectl command or by editing the workload's yaml manifest.

For demo purposes we will showcase this with a deployment.

kubectl

Adding the downtime with kubectl is a single command:

kubectl annotate deployment my-deployment downscaler/downtime="Mon-Fri 18:00-08:00 UTC, Sat-Sun 00:00-24:00 UTC"

Edit manifest

If you want to edit the annotations manually with a yaml manifest file or by running kubectl edit deployment my-deployment you have to add the following:

my-deployment.yaml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: my-deployment
  annotations:
    downscaler/downtime: "Mon-Fri 18:00-08:00 UTC, Sat-Sun 00:00-24:00 UTC"
spec:
  replicas: 3
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:1.14.2
        ports:
        - containerPort: 80

Now your deployment my-deployment will be scaled down on every weekday from 6:00 PM to 8:00 AM and on weekends.

Managing Exclusions

Exclusions can be made with the same granularity using environment variables, cli arguments or annotations at namespace or workload level.

Environment Variable

If you want to add cluster-wide exclusions with environment variables you can add them inside the values.yaml file.

There you have to add it to the configMap.extraConfig field:

values.yaml

# ...

configMap:
name: go-kube-downscaler
extraConfig: |
DEFAULT_DOWNTIME: "Mon-Fri 18:00-08:00 UTC, Sat-Sun 00:00-24:00 UTC"
EXCLUDE_NAMESPACES: "^(kube-system|istio-system|karpenter|prometheus|.*prod.*)$"

# ...

Cli Argument

If you want to use a cli argument for the cluster-wide exclusions you can add them inside the values.yaml file.

There you have to add it to the arguments field:

values.yaml

# ...

arguments:
- --default-downtime="Mon-Fri 18:00-08:00 UTC, Sat-Sun 00:00-24:00 UTC"
- --exclude-namespaces="^(kube-system|istio-system|karpenter|prometheus|.*prod.*)$"
# ...

Namespace Level

If you want to exclude an entire namespace from the cluster-wide downtime, you can annotate the namespace with downscaler/exclude="true" or downscaler/exclude-until.

kubectl annotate namespace my-namespace downscaler/exclude="true"

kubectl annotate namespace your-namespace downscaler/exclude-until="2025-12-22T00:00:00+01:00"

Workload Level

If you want to exclude a specific workload from the cluster-wide downtime, you can annotate the workload with downscaler/exclude="true".

kubectl annotate deployment my-deploy downscaler/exclude="true"

kubectl annotate cronjob my-job downscaler/exclude-until="2025-12-22T00:00:00+01:00"

---

info

For more information about the scaling algorithm check out the scope documentation. You will find a detailed explanation of how the GoKubeDownscaler evaluates the downtime and decides when to scale down or up workloads.