scalehpa: Easy way to scale your workload when keda is active

What is scalehpa?

scalehpa is a tool to overcome a limitation seen in keda until version 2.8.x, where the keda controller will revert any attempt to scale the deployment to a value different from the minReplicaCount present on the scaledobject.

It's written in a way to execute the scale even if the deployment doesn't have scaledobjects defined, by calling the default kubectl scale.

Examples

Set a new size for a deployment or a statefulset when keda is installed.

Examples:
# Set a minimum number of replicas for the deployment on the default or current namespace
# This will set the minReplicas to 3 and keep HPA active
scalehpa deploy foo --replicas=3
scalehpa deploy foo -r 3

# This will set the maxReplicas to 10 and keep HPA active
scalehpa deploy foo --maxreplicas=10
scalehpa deploy foo -m 10

# This will set the minReplicas to 10, maxReplicas to 20 and keep HPA active
scalehpa deploy foo -r 10 -m 20

# scale a deployment on the default or current namespace and DISABLE HPA
# This will create a fixed set of 3 replicas, disabling autoscale
# To reenable HPA, use the --replicas option
scalehpa deploy foo --freeze=3

# scale a statefulset on a specific namespace
scalehpa sts -r 3 foo -n bar
scalehpa statefulset foo -n bar -r 3

# show this message
scalehpa -h,--help

---

Installation

Installing as a krew plugin

$ kubectl krew index add scalehpa https://github.com/sergiomacedo/scalehpa.git
$ kubectl krew install scalehpa/scalehpa
$ kubectl scalehpa 
Set a new size for a deployment or a statefulset when keda is installed.

Examples:
# Set a minimum number of replicas for the deployment on the default or current namespace
# This will set the minReplicas to 3 and keep HPA active
kubectl scalehpa deploy foo --replicas=3
kubectl scalehpa deploy foo -r 3

# This will set the maxReplicas to 10 and keep HPA active
kubectl scalehpa deploy foo --maxreplicas=10
kubectl scalehpa deploy foo -m 10

# This will set the minReplicas to 10, maxReplicas to 20 and keep HPA active
kubectl scalehpa deploy foo -r 10 -m 20

# scale a deployment on the default or current namespace and DISABLE HPA
# This will create a fixed set of 3 replicas, disabling autoscale
# To reenable HPA, use the --replicas option
kubectl scalehpa deploy foo --freeze=3

# scale a statefulset on a specific namespace
kubectl scalehpa sts -r 3 foo -n bar
kubectl scalehpa statefulset foo -n bar -r 3

# show this message
kubectl scalehpa -h,--help

Manual Installation (macOS and Linux)

Since scalehpa is written in Bash, you should be able to install them to any POSIX environment that has Bash installed.

• Download the scalehpa script.
• Either:
  • save them all to somewhere in your PATH,
  • or save them to a directory, then create symlinks to scalehpa from somewhere in your PATH, like /usr/local/bin
• Make scalehpa executable (chmod +x ...)

Example installation steps:

sudo git clone https://github.com/sergiomacedo/scalehpa /opt/scalehpa
sudo ln -s /opt/scalehpa/scalehpa /usr/local/bin/scalehpa