Deploying cnBNG Control Plane in Redhat Openshift

Overview

cnBNG Control Plane can be deployed in Redhat Openshift environment starting from release 25.01.1. Current supported Openshift Container Platform (OCP) version is v4.17. In this tutorial we will learn to deploy cnBNG Control Plane in Single Node Openshift cluster.

cnBNG Control Plane can be deployed in Redhat Openshift environment in three steps:

1. Preparing Openshift Container Platform (OCP)
2. CEE application deployment for monitoring
3. cnBNG Control Plane application deployment

Prerequisite

Following are the prerequisite for this tutorial:

• Openshift Container Platform (OCP) deployed in either a VM or on a Baremetal Server
• Helm Chart Repository hosting Control Plane helm charts
• Docker Image Repository hosting docker images for Control Plane

Note: We will cover Helm Chart Repository and Docker Image Repository creation in some other tutorial.

Network

In this tutorial we will use single network for management as well as to peer with User Planes and communicate with Radius. You can also use separate networks for management, User Plane peering and Radius communication.

Step-1: Preparing OCP Environment

1. Login to local OCP cluster console as admin and follow below steps.

Node Labels

1. Add following labels to the node. These labels are required for cnBNG POD deployment/ placement. Add labels by selecting Nodes under Compute from side panel.

    smi.cisco.com/node-type=oam
    disktype=ssd
   

   

   

Creating Namespaces

1. Create two namespaces “ceeocp” and “bngocp” by selecting Namespaces under Administration from side panel

Creating Secrets

Secrets are required to pull Helm Charts and Docker Images for cnBNG Control Plane and CEE application. In this step we will create helm secret and docker image pull seceret for both namespaces “ceeocp” and “bngocp”.

1. Switch to the Developer mode from Administrator mode by selecting it from Side Panel

2. In Developer mode, select “Secrets” from side panel. Now click on “Create->From YAML” under project ceeocp.

3. Copy paste below yaml

     apiVersion: v1
     kind: Secret
     metadata:
      name: cnbng-helm-repo-secret
     stringData:
      username: your-helm-repo-username
      password: your-helm-repo-password
   

4. Click again on “Create->Image pull secret” for “ceeocp” project. Fill form data as per below information:

   Secret Name: cnbng-docker-repo-secret

   Authentication type: Image registry credentials

   Registry server address: your-docker-hub-url

   Username: your-username-for-dockerhub

   Password: your-password-to-access-dockerhub

5. Similarly create Helm secret and docker secret for bngocp project, using same yaml and form data, as was created for ceeocp project.

Creating Helm Repositories

Now we will create Helm repositories for CEE and cnBNG Control Plane applications.

1. Select “Create->Repository” by following options as highlighted below.

2. Select YAML view and copy paste below YAML in place of default YAML provided.

    apiVersion: helm.openshift.io/v1beta1
    kind: ProjectHelmChartRepository
    metadata:
      name: ceeocp
      namespace: ceeocp
    spec:
      name: ceeocp
      connectionConfig:
        url: >-      
          https://your-helm-chart-repo-url-for-cee
        basicAuthConfig:
          name: cnbng-helm-repo-secret
   

   Note Make sure you have selected “ceeocp” as the Project

3. Similarly create Repository for “bngocp” Project using below YAML.

    apiVersion: helm.openshift.io/v1beta1
    kind: ProjectHelmChartRepository
    metadata:
      name: bngocp
      namespace: bngocp
    spec:
      name: bngocp
      connectionConfig:
        url: >-
          https://your-helm-chart-repo-url-for-cnbng-control-plane
        basicAuthConfig:
          name: cnbng-helm-repo-secret
   

With this we complete preparation step for OCP.

Step-2: Deploying CEE

In this step we will deploy CEE PODs by first deploying CEE Ops Center POD and then doing necessary configurations on CEE Ops Center to deploy rest of CEE PODs.

Deploying CEE Ops Center POD

Now since we have got the helm repositories created, we can proceed to create helm release and deploy CEE Ops Center POD.

1. Click on “Create->Helm Release” to create helm release from the charts.

   

2. After release is created click on Ops Center POD chart for ceeocp.

   

3. Click on Create for YAML definitions.

   

4. Replace default YAML definition with YAML definition below and change “Release name” to “cee-ops-center-ams1”.

   

    global:
      caasModel: "openshift"
      imagePullPolicy: IfNotPresent
      imagePullSecrets:
        - name: cnbng-docker-repo-secret
      ingressHostname: your-node-ip-address.nip.io
      istio: false
      networkPolicy:
        enabled: false
      pathBasedIngress: true
      registry: your-dockerhub-url
      singleNode: true
      smiNfMonitoring: true
      smiPlatformMonitoring: true
      useVolumeClaims: true
    ops-center:
      product:
        helm:
          product: cee
          autoDeploy: false
          repository:
            url: https://your-helm-chart-repo-url-for-cee-charts
            name: ceeocp
            accessToken: your-helm-repo-username:your-helm-repo-access-password
    helm-api:
      env:
        - name: DOWNLOAD_CHART_PROXY
          value: "https://your-proxy-url"
   

5. Once the POD is deployed successfully and is in Running state, click on the POD name “ops-center-cee-ops-center-ams1”

   

6. Scroll down and verify that all four containers are in Running state.

   

Configuring CEE Ops Center

1. Click on Terminal for CEE Ops Center POD and connect to confd_cli using command bin/confd_cli -u admin

   

   Note You will be prompted to set the passoword while connecting for first time.

2. Configure following on CEE Ops Center to get all the PODs running for CEE.

    config
    pv-provisioner pv-path /var/data
    k8s name cnbng
    system mode running
    commit
   

3. Verify that the “system status percent-ready” is 100% using command “show system” before proceeding further.

    
    [unknown] cee# show system
    system uuid 3e2b0987-4f8b-40fa-a2c2-0cb2969e7882
    system status deployed true
    system status percent-ready 100.0
    sustem ons-center renosttory unknown system
    ops-center-debug status false
    system synch running true system synch pending
    
    

With this we complete deployment of CEE application.

Step-3: Deploying cnBNG Control Plane

In this step we will deploy cnBNG Control Plane (CP) PODs by first deploying cnBNG CP Ops Center POD and then doing necessary configurations on the Ops Center to deploy rest of the PODs.

Deploying cnBNG Control Plane Ops Center

The procedure to deploy cnBNG Control Plane Ops Center is similar as that of deploying CEE Ops Center.

1. Click on “Create->Helm Release” to create helm release from the charts.

   

2. After release is created click on Bng Ops Center POD chart for bngocp.

   

3. Click on Create for YAML definitions.

   

4. Replace default YAML definition with YAML definition below and change “Release name” to “bng-ops-center-ams1”.

   

    global:
      caasModel: "openshift"
      imagePullPolicy: IfNotPresent
      imagePullSecrets:
        - name: cnbng-docker-repo-secret
      ingressHostname: your-node-ip-address.nip.io
      istio: false
      networkPolicy:
        enabled: false
      pathBasedIngress: true
      registry: your-dockerhub-url
      singleNode: true
      smiNfMonitoring: true
      smiPlatformMonitoring: true
      useVolumeClaims: true
    ops-center:
      product:
        helm:
          product: bng
          autoDeploy: false
          repository:
            url: https://your-helm-chart-repo-url-for-cnbng-cp-charts
            name: bng-main
            accessToken: your-helm-repo-username:your-helm-repo-access-password
    helm-api:
      env:
        - name: DOWNLOAD_CHART_PROXY
          value: "https://your-proxy-url"
   

5. Once the POD is deployed successfully and is in Running state, click on the POD name “ops-center-bng-ops-center-ams1”

   

6. Scroll down and verify that all four containers are in Running state.

   

Configuring cnBNG Control Plane Ops Center

1. Click on Terminal for BNG Ops Center POD and connect to confd_cli using command bin/confd_cli -u admin

   

   Note You will be prompted to set the passoword while connecting for first time.

2. Configure following on BNG Ops Center and initialize it.

    config
    cdl node-type session
    cdl zookeeper replica 1
    cdl logging default-log-level error
    cdl datastore session
     slice-names [ 1 ]
     endpoint replica 1
     endpoint settings slot-timeout-ms 750
     index replica 1
     index map    1
     index write-factor 1
     slot replica 1
     slot map     1
     slot write-factor 1
     slot notification limit 300
    exit
    cdl kafka replica 1
    etcd replicas 1
    instance instance-id 1
     endpoint sm
     exit
     endpoint nodemgr
     exit
     endpoint n4-protocol
      retransmission timeout 0 max-retry 0
     exit
     endpoint dhcp
     exit
     endpoint pppoe
     exit
     endpoint radius
      vip-ip your-node-ip-address
      interface coa-nas
       sla response 140000
       vip-ip your-node-ip-address vip-port 2000
      exit
     exit
     endpoint udp-proxy
      vip-ip your-node-ip-address
     exit
    exit
    deployment
     app-name     BNG
     cluster-name cnbng
     dc-name      DC
     model        small
    exit
    k8 bng
     etcd-endpoint      etcd:2379
     datastore-endpoint datastore-ep-session:8882
     tracing
      enable
      enable-trace-percent 30
      append-messages      true
      endpoint             jaeger-collector:9411
     exit
    exit
    instances instance 1
     system-id  DC
     cluster-id cnbng
     slice-name 1
    exit
    local-instance instance 1
    k8 label protocol-layer key disktype value ssd
    exit
    k8 label oam-layer key smi.cisco.com/node-type value oam
    exit
    system mode running
    commit
   

3. Verify that the “system status percent-ready” is 83% using command “show system”. This is because following four PODs are in init state and can only be in Running state post Day-1 configuration.

    
    bng-dhcp-n0-0
    bng-nodemgr-n0-0
    bng-pppoe-n0-0
    bng-sm-n0-0
    
    

4. We can now also connect to Grafana Dashboards with the URL: https://your-node-ip-address.nip.io/ceeocp/grafana/

   

   Note: To login to Grafana use Username as “admin” and Password which you configured for CEE Ops Center login.

This completes the deployment of cnBNG CP in Redhat Openshift Container Platform (OCP).

Acknowledgements

Many thanks to Madhu for his contributions.