turtles/exp

Experimental(Technical Preview)

This is a technical preview of the experimental features that are currently being developed in the project. These features are not yet ready for production use and are subject to change.

turtles-etcd-restore

Setting up the environment

To set up the environment, navigate to the root of the repository and run:

export RANCHER_HOSTNAME="<hostname>"
export NGROK_API_KEY="<api-key>"
export NGROK_AUTHTOKEN="<api-authtoken>"
export USE_TILT_DEV=true (default)

make dev-env

Note: setting USE_TILT_DEV environment variable to false will result in manually deploying Rancher Turtles locally instead of Tilt deployment and can be used for testing Rancher Turtles with Helm chart changes (enabling/disabling feature flags when passed as argument to Turtles helm installation command).

The Makefile target sets up the environment by executing the scripts/turtles-dev.sh script with the RANCHER_HOSTNAME argument. Under the hood, it performs the following steps:

1. Creates a kind cluster.
2. Deploys cert-manager, CAPI Operator and Rancher Turtles.
3. Deploys CAPRKE2 provider.
4. Deploys Docker provider.
5. Deploys ngrok.
6. Deploys Rancher accessible via ngrok.

Environment is prepared for cluster creation using CAPRKE2. UI is accessible via RANCHER_HOSTNAME.

Creating a cluster

To deploy an RKE2 cluster with automatic snapshots enabled:

export CONTROL_PLANE_MACHINE_COUNT=1
export WORKER_MACHINE_COUNT=1
export CLUSTER_NAME=rke2
export KUBERNETES_VERSION=v1.30.0
export RKE2_VERSION=v1.31.0+rke2r1

# Prevent replacing lb-config variables by envsubst
export address='$address'
export server='$server'

envsubst < test/e2e/data/cluster-templates/docker-rke2.yaml | kubectl apply -f -

Performing a manual snapshot

export CLUSTER_NAMESPACE=default
export CLUSTER_NAME=rke2
export ETCD_MACHINE_SNAPSHOT_NAME="manual-snapshot"
export MACHINE_NAME=$(kubectl get machines -l cluster.x-k8s.io/control-plane  -o jsonpath='{.items[0].metadata.name}')

envsubst < exp/etcdrestore/examples/etcd-snapshot.yaml | kubectl apply -f -

Performing the restore

When all machines in the cluster are ready, automatic ETCDMachineSnapshot object should appear on the management cluster soon.

kubectl get etcdmachinesnapshot -A

To perform a restore run the following command:

export CLUSTER_NAMESPACE=default
export CLUSTER_NAME=rke2
export ETCD_MACHINE_SNAPSHOT_NAME="<snapshot_name_from_the_output>"

envsubst < exp/etcdrestore/examples/etcd-restore.yaml | kubectl apply -f -

Cleanup

To clean up the environment, run the following command from the root of the repo:

make clean-dev-env