Preparing to install IBM Cloud Private with OpenShift

Before you install IBM Cloud Private with OpenShift, review the following installation requirements.

Installation requirements

The supported IBM Cloud Private version for this integration is 3.1.0.
You must have OpenShift version 3.9 installed and working in your cluster that includes registry and storage services.
You should have a pre-configured StorageClass in OpenShift that can be used for creating storage for IBM Cloud Private
Hardware requirements:
- Linux® x86_64 64-bit platform
- Master nodes: 8 Core | 32 GB RAM
  - Label the master node as compute by running the following command:
```
sudo kubectl label nodes ${cluster_name}-master.purple-chesterfield.com node-role.kubernetes.io/compute=true
```
- Worker nodes: See OpenShift requirements
Networking:
- The port number 8445 is required to be open on every node in the OS environment for the node exporter in the monitoring service. This port is configurable and 8445 is the default value.
- You must have different ports for IBM Cloud Private management ingress, as port 8080 and 8443 is used by OpenShift services.
- You must have different ports for nginx ingress, as port 80 and 443 is used by OpenShift services.
Storage:
- Set up a storage class.
- If you are using SELinux, for PersistentVolume, you might need to create the directories that are used for the local Helm repositories; and then set their security context so that they can be accessed from within the containers. On each master node, run the following commands:
```
mkdir /var/lib/icp/helmrepo
chcon -Rt svirt_sandbox_file_t /var/lib/icp/helmrepo
mkdir /var/lib/icp/mgmtrepo
chcon -Rt svirt_sandbox_file_t /var/lib/icp/mgmtrepo
```
  Note: The the mkdir commands are needed only if the directories do not exist.

For Elasticsearch, ensure that the vm.max_map_count setting is at least 262144 on all nodes:

sysctl -w vm.max_map_count=262144
echo "vm.max_map_count=262144" | sudo tee -a /etc/sysctl.conf