acme-hardware-management-plugin

README

O-RAN O2 IMS hardware management plugin example

This project is an example of how to write a hardware management plugin for Red Hat's O-RAN O2 IMS implementation.

Building

To build the project make sure that you have a working version of Go 1.22 and then simply type go build in the top level directory of the project. It will generate a acme-hardware-management-plugin binary.

Running

The plugin is a Kubernetes controller that watches objects of kind NodeAllocationRequest and NodeReleaseRequest and acts accordingly.

To play with it you will need any Kubernetes cluster. Then you will need to apply the custom resource definitions for the NodeAllocationRequest and NodeReleaseRequest. They are available in in the openshift-knk/oran-o2ims project, inside the config/cdr/bases directory, so you can create them like this:

$ kubectl create -f https://raw.githubusercontent.com/openshift-kni/oran-o2ims/main/config/crd/bases/hardwaremanagement.oran.openshift.io_nodeallocationrequests.yaml
$ kubectl create -f https://raw.githubusercontent.com/openshift-kni/oran-o2ims/main/config/crd/bases/hardwaremanagement.oran.openshift.io_nodereleaserequests.yaml

Then start the controller in one window:

$ ./acme-hardware-management-plugin

This will run in the foreground and will write log messages to the output.

To simulate the O-RAN O2 IMS you can, in another window, manually create a node allocation request:

$ kubectl create -f - <<.
kind: NodeAllocationRequest
apiVersion: hardwaremanagement.oran.openshift.io/v1alpha1
metadata:
  namespace: europe
  name: cu-113d00a-22422

spec:
  cloudID: "0f34b4cf-41a0-4b49-bcb3-9f2daebfbee7"
  location: madrid
  extensions:
    "oran.acme.com/model": "BigIron X42"
    "oran.acme.com/firmwareSettings": |
       {
         "MinProcIdlePower": "C6"
       }
    "oran.acme.com/firmwareVersions": |
       {
         "BigIron UEFI": "4.11",
         "Intel(R) E810-XXVDA2": "2.50"
       }
.

Observe in the logs how the controller fulfills the request and check the result, in particular the status, where the controller will write the details of the node, in particular the BMC address and the name of the secret containing the BMC user name and password:

$ kubectl get nodeallocationrequest -n europe cu-113d00a-22422 -o yaml
$ kubectl get secret -n europe cu-113d00a-22422-bmc -o yaml

You can also create a node deallocation request:

$ kubectl create -f - <<.
kind: NodeReleaseRequest
apiVersion: hardwaremanagement.oran.openshift.io/v1alpha1
metadata:
  namespace: europe
  name: cu-113d00a-22422

spec:
  cloudID: "0f34b4cf-41a0-4b49-bcb3-9f2daebfbee7"
  nodeID: "acme-c3f64cf1-40c0-4324-b13a-40f22b0d16c5"
.

And check the result:

$ kubectl get nodereleaserequest -n europe cu-113d00a-22422