peerpod-ctrl

README

peerpod-ctrl

The PeerPod CR is used to track the cloud provider resources for a (peer)Pod; it requires the cloud InstanceID and the CloudProvider. PeerPod objects are owned by the matching Pod object. The PeerPod controller is watching PeerPod events and deleting dangling resources that were not deleted by the cloud-api-adaptor at Pod deletion time.

Description

Creation time:

With every successful VM creation for a Pod, cloud-api-adaptor will create a PeePod CR (predefined by the operator) which contains the VM instance id and cloud provider.

Owner references:

The PeerPod CR is owned by the original Pod object. Upon Pod deletion background cascading deletion gets into action and hence the Pod will be deleted first, followed by GC handling the owned PeerPod CR.

Deletion time:

Normal case: When remote hypervisor will get the stopVM request (upon Pod deletion) it will delete the pod VM instance and if it succeeds it will remove the finalizer attached to the owned PeerPod object so it can then be cleaned by the GC.

Failure case: If for any reason cloud-api-adaptor doesn’t honor the delete request or it fails to perform deletion, the finalizer is not removed. Hence, when PeerPod controller gets a delete event for the owned PeerPod object by the GC and it still has the finalizer, it will comprehend that it needs to perform the deletion of pod VM resource by itself, based on the PeerPod CR fields.

Getting Started

You’ll need a Kubernetes cluster on a supported provider to run against (e.g. you can use Libvirt for development). Note: Your controller will automatically use the current context in your kubeconfig file (i.e. whatever cluster kubectl cluster-info shows).

Running on the cluster

Make sure to install cloud-api-adaptor first

make deploy

Note: alternatively you can deploy the peerpod-ctrl along with cloud-api-adaptor installtion by setting RESOURCE_CTRL=true

Uninstall CRDs

To delete the CRDs from the cluster:

make uninstall

Undeploy controller

UnDeploy the controller to the cluster:

make undeploy

Contributing

For any changes in the CRD/controller make sure it doesn't break the k8s api calls from the cloud-api-adaptor and adapt it if needed.

How it works

This project aims to follow the Kubernetes Operator pattern

It uses Controllers which provides a reconcile function responsible for synchronizing resources untile the desired state is reached on the cluster

Test It Out

Running custom build

1. Install Instances of Custom Resources:

kubectl apply -f config/samples/

1. Build and push your image to the location specified by IMG:

make docker-build docker-push IMG=<some-registry>/peerpod-ctrl:tag

1. Deploy the controller to the cluster with the image specified by IMG:

make deploy IMG=<some-registry>/peerpod-ctrl:tag

Modifying the API definitions

If you are editing the API definitions, generate the manifests such as CRs or CRDs using:

make manifests

NOTE: Run make --help for more information on all potential make targets

More information can be found via the Kubebuilder Documentation