SCP Deployment

This procedure describes the steps to deploy SCP on CNE. The below steps need to be executed from a server which has access to Kubectl and helm commands.

Table 2-4 SCP Deployment

Step #	Procedure	Description
1	Search helm chart	Execute the following command to check the version of the helm chart installation. `helm search <deployment_name>`
2	Prepare custom_values.yaml file	Prepare a custom_values.yaml file with the required parameter information. Refer to SCP Configuration Parameters for more information on parameters. Refer to OCSCP YAML File for sample YAML file. You can also download sample ocscp_values.yaml file from OHC, refer to Table 2-3 for more information. Note: The user needs to update the "domain" in the custom_values.yaml file per the name of cluster (default value of domain is "svc.cluster.local"). If the cluster name is XYZ then domain must be svc.XYZ. The user needs to update the "clusterDomain" in the custom_values.yaml file per the name of cluster (default value of domain is "cluster.local"). If the cluster name is XYZ then domain must be XYZ.
3	Deploy SCP using HELM repository	Execute the following command: `helm install <helm-repo> -f <custom_values.yaml> --name <deployment_name> --namespace <namespace_name> --version <helm_version>` Where: helm-repo: repository name where the helm images, charts are stored custom_values: helm configuration file which needs to be updated based on the docker registry deployment_name and namespace_name: depends on customer configuration For example: helm install ocscp-helm-repo/ocscp -f <custom values.yaml> --name ocscp --namespace scpsvc --version <helm version>
4	Deploy SCP using HELM tgz	Execute the following command: `helm install -f <custom values.yaml> --name ocscp --namespace <namespace> <chartpath>./<chart>.tgz`
5	Check repo status	Execute `helm status <deployment_name>` to check the deployment status.
6	Check svc status	Check if all the services are deployed and running: `kubectl -n <namespace_name> get services`
7	Check pod status	Check if all the pods are up and running: `kubectl -n <namespace_name> get pods` Note: Worker and pilot status must be Running and Ready must be n/n. scpc-soothsayer status must be Running and Ready must be n/n, where n is number of containers in the pod.