Rancher on HPE SimpliVity

# Initial deployment

# Run site.yml

Run the playbook site.yml to perform the initial admin and user cluster deployment:

$ cd ~/Rancher-on-SimpliVity
$ ansible-playbook -i hosts site.yml --vault-password-file .vault_pass

The playbook should take approximately twenty minutes, depending on your exact configuration.

Artifacts generated during the installation are stored in the folder determined by the install_dir variable, specified in the file group_vars/all/private.yml. The default location is the .svtrancher folder in your home directory.

Files generated include:

  • kube_config_rancher-cluster.yml - the kubeconfig file that provides access to the admin cluster
  • rke-up.log - the log file from the installation of the admin cluster
  • rancher-cluster.yml - the admin cluster definition file

# Playbooks overview

The following section provides an summary of the tasks carried out when running the site.yml playbook.

The main entry point site.yml is a wraper for a number of playbooks that perform the deployment in stages:

  • playbooks/prepare.yml
  • playbooks/provision.yml
  • playbooks/configure.yml
  • playbooks/ranchercluster.yml
  • playbooks/rancherserver.yml
  • playbooks/1stlogin.yml
  • playbooks/ucluster.yml
  • playbooks/csi.yml

# Prepare

The prepare stage performs some initial setup on the Ansible controller:

  • Configures proxy if required and installs the required packages on the Ansible box
  • Verifies that required files have been downloaded to staging area
  • Installs client tools such as Helm, and CLIs for Rancher, RKE, kubectl
  • Creates required artifacts in vCenter including VM folders, resource pools
  • Populates /etc/hosts

It also performs datastore configuration:

  • Connect to HPE SimpliVity
  • Create datastore if it does not already exist

# Provision

The provision stage performs the following tasks:

  • Provision and upgrade the template VM
  • Provision Support (DHCP) VM
  • Provision Loadbalancer VMs
  • Provision Rancher admin server VMs
  • Update package manager on all provisioned VMs

# Configure

The configure stage configures the admin server VMs, DHCP VM and loadbalancer VMs.

  • Create DRS anti-affinity rules for Rancher admin server VMs and load balancer VMs
  • Configure the Rancher admin server VMs
    • Install required packages, configure firewall and open ports
    • Install and configure Docker
  • Configure support node (DHCP)
    • Install required packages, configure firewall and open ports
    • Install and configure DHCP
  • Configure load balancer
    • Install required packages, configure firewall and open ports
    • Install and configure nginx

# Rancher admin cluster

The Rancher admin cluster is a Kubernetes cluster, deployed using RKE. The cluster configuration is stored in the file rancher-cluster.yml in the installation directory.

# Rancher server

The Rancher server software is installed using Helm on top of the Rancher admin cluster.

For self-signed certs, cert-manager software is installed. cert-manager is a Kubernetes add-on to automate the management and issuance of TLS certificates from various issuing sources. It ensures certificates are valid and up to date periodically, and attempt to renew certificates at an appropriate time before expiry. For more information, see https://github.com/jetstack/cert-manager.

For user-supplied certs, a Kubernetes secret is created from the supplied certs.

# First login

In this stage, a number of first-time log in operations are performed, including changing the admin password of the Rancher server and creating an API token.

# User cluster

  • Create cloud credentials
  • Create node template, from user (or the default admin) template using configuration specified in user-template variable in the group_vars/all/vars.yml file
  • Deploy the user cluster with the Rancher API, using the node template.

In the current release, the user cluster deploys with a kubernetesVersion of v1.17.2-rancher1-2.

# Enable CSI for user cluster

  • Create a datastore named csi_datastore_name of size csi_datastore_size
  • Retrieve the kubeconfig for user cluster and store it in installation directory
  • Configure vSphere CSI driver
  • Create storage class named csi_storageclass_name

← Redeployment →