- RHEL-family machine (Fedora 35, Centos 8, RHEL 8)
- VM should suffice too. We recommend a minimum of 2vCPUs and 8GB of RAM. If using Amazon EC2 instance, we recommend a t2.large VM.
- Ensure the following tools are installed and available in the PATH:
It's convenient to just use the podman driver for minikube; it's lightweight and works great. We have some helper scripts that you will use to spawn a couple of minikube clusters to migrate between (let's call them src, dest) for crane, and we'll also set up some iptables network forwarding rules along with some in-cluster dns to ensure the clusters can route traffic to one another, and are able to resolve ingress. We also have scripts to help clean this up.
Here is a script to help you setup two minikube clusters and the appropriate iptable rules.
wget "https://raw.githubusercontent.com/konveyor/crane/main/hack/minikube-clusters-start.sh" chmod +x minikube-clusters-start.sh ./minikube-clusters-start.sh
Note - We also have had folks run through the scenarios on Mac machines, but fair warning, this path has significantly less usage.
Note - Tackle requires a more demanding Minikube instance. Details about how to configure Minikube for the Tackle scenario can be found here.
Crane specific prereqs:
The crane scenarios utilize several additional tools. Ensure the binaries are installed to your PATH so they are accessible: