Before you run the test, please make sure:

1. The corresponding controller has been deployed.
2. The corresponding CRD has been created.

Recommend using deploy.sh to do prerequisites above.

Run cargo run -- <target>, where <target> is one of:

• rabbitmq (plus rabbitmq-scaling, rabbitmq-ephemeral)
• vreplicaset, vdeployment, vstatefulset
• vreplicaset-admission, vdeployment-admission, vstatefulset-admission

For example, cargo run -- rabbitmq tests the RabbitMQ controller and cargo run -- vreplicaset tests the VReplicaSet controller.

If the controller works as expected, the program will print xxx cluster is ready! e2e test passed.

The test may also return different errors based on different situations:

• Failed to get kube client: ... means kube client error. There is something wrong with k8s cluster or kube client connections.
• Failed to get CRD: ... means the corresponding CRD has not been created.
• Timeout, e2e test failed! means the state doesn't reach to expected state in 6 mins.
• Statefulset is not consistent with rabbitmq cluster spec! means the statefulset spec is not correct.
• ConfigMap is not consistent with rabbitmq cluster spec! means the configmap spec is not correct.
• Rabbitmq failed to set customized user/password! means the test for customized user/password failed.