Running the reference, start to finish.

Verified configuration

Requirement	Minimum	Notes
OS	Fedora 44, Docker Engine (docker-ce, daemon running, user in docker group)	minikube node runs containerd with runc
RAM	64 GB total	24 GB allocated to minikube profile; rest is host headroom
Disk	1 TB	About 100 GB free under the Docker data root (/var/lib/docker)
CPU	16 vCPU recommended	Not strictly required but the stack is heavy

Container engine

The capstone runs on Docker Engine: native docker-ce on Linux, with the daemon running and your user in the docker group. Install it from docs.docker.com/engine/install, then enable it and point the CLI at it:

sudo systemctl enable --now docker      # or: sudo systemctl start docker
sudo usermod -aG docker "$USER"         # log out and back in afterwards
docker context use default
docker info

Membership in the docker group is root-equivalent on the host.

VM-based engines

The capstone scripts are supported on Fedora and RHEL hosts (bare metal or VM). A VM-based engine such as Docker Desktop can provide the engine instead. It is an option, not a requirement. Its VM must be sized to at least 28 GB of memory; set MINIKUBE_CPUS and MINIKUBE_MEMORY to fit it, as described below.

Upgrading from an earlier (rootless podman) capstone profile

An earlier release created the profile with rootless podman. Delete that profile first, then clear the setting, then run setup:

MINIKUBE_ROOTLESS=true minikube delete -p capstone 
minikube config unset rootless 
./scripts/setup-capstone-profile.sh

Fitting minikube to the engine

The profile defaults to 16 CPUs, 24 GB of memory, and an 80 GB disk. Override them before creating the profile:

export MINIKUBE_CPUS=16
export MINIKUBE_MEMORY=24g
export MINIKUBE_DISK=80g

The docker driver on Linux ignores the disk size; data lives under the engine's data root (docker info shows it), where about 100 GB free is recommended.

Required tooling

• docker and docker compose (compose is not used by the minikube path; listed as a standard host tool)
• minikube >= 1.39.0 — the profile pins Kubernetes v1.36.5
• kubectl 1.36.x
• helm
• istioctl 1.31.1 — with the full Istio distribution (not just the binary; setup-kiali.sh applies samples/addons/kiali.yaml from it)

Optional, for the Red Hat (OpenShift Local) appendix only: crc, podman, and podman compose.

Installing the Istio distribution

curl -fsSL https://github.com/istio/istio/releases/download/1.31.1/istio-1.31.1-linux-amd64.tar.gz \
    | tar xz -C ~/.local/share
ln -sfn ~/.local/share/istio-1.31.1 ~/.local/share/istio-current
cp ~/.local/share/istio-current/bin/istioctl ~/.local/bin/

Verifying the stack

Cluster status

./scripts/cluster-status.sh

Quick verification demos

1. ./demos/demo-order.sh — creates an order via REST, confirms it in Postgres
2. ./demos/demo-service.sh inventory — builds, deploys, and health-checks a service
3. ./demos/demo-observability.sh — verifies Prometheus is scraping and Grafana is provisioned