Testing resources for Kedify & KEDA

Basic: Sample SO and SJ

kubectl apply -f https://raw.githubusercontent.com/kedify/test-data/main/resources/minutemetrics.yaml
kubectl apply -f https://raw.githubusercontent.com/kedify/test-data/main/resources/target.yaml
kubectl apply -f https://raw.githubusercontent.com/kedify/test-data/main/resources/so.yaml
kubectl apply -f https://raw.githubusercontent.com/kedify/test-data/main/resources/sj.yaml

Cluster/TriggerAuthentications

Creates 8 TAs and 8 CTAs

kubectl apply -f https://raw.githubusercontent.com/kedify/test-data/main/resources/tas-ctas.yaml

Bulk creation:

Following command creates 5 namespaces (named test-1 - test-5), in each namespace 25 SOs and 20 SJs

curl -s https://raw.githubusercontent.com/kedify/test-data/main/resources/create_resources.sh | \
bash -s -- 5 25 20 test

Usage: ./create_resources.sh <num_namespaces> <scaled_objects_per_ns> <scaled_jobs_per_ns> [namespace_prefix]

Simulating a k8s cluster that has also some other workloads:

# create a cluster w/ 7 worker nodes
k3d cluster delete ; k3d cluster create --servers 7 --no-lb --k3s-arg "--disable=traefik,servicelb,local-storage@server:*"
kubectl wait --for=condition=Ready nodes --all --timeout=600s

# install kedify
kubectl kedify i --email foo@bar -y

# 30 pods in total
curl -s https://raw.githubusercontent.com/kedify/test-data/main/resources/create_resources.sh | \
OTHER_DEPLOYMENTS=1 \
bash -s -- 2 10 4 scale

# create 10 namespaces (t-1 - t-10) and in each 6 deployments and 1 stateful set (with pause container)
# 70 pods in total
curl -s https://raw.githubusercontent.com/kedify/test-data/main/resources/create_resources.sh | \
OTHER_DEPLOYMENTS=6 \
OTHER_STATEFUL_SETS=1 \
bash -s -- 10 0 0 t