Proxmox (homelab) state role

This ansible-role helps you deploying Proxmox proxmox_vms with ease. In my homelab I deploy or scale VM's within 1 minute.

Requirements

The following modules are required;

Collections:

• community.general

Pip modules:

• ansible
• requests
• proxmoxer

Features

• Configure proxmox servers
• Deploy VM's
• Scale deployed VM's

Usage

First steps

One of the first steps is to configure the vars defined in defaults/main.yml. There you define your default template to use etc. etc..

Inventory

The inventory should define your Proxmox hosts in the proxmox group and your proxmox_vms in the servers group. For example:

proxmox:
  hosts:
    node01:
      ansible_host: node01.local
    node02:
      ansible_host: node02.local

servers:
  hosts:
    rke2:
      vmid: 101
      node: node01  # Optional: specify which Proxmox node to use

Deployment

When you want to deploy VMs, define them in the servers hostgroup. You can optionally specify which Proxmox node to deploy to using the node variable:

servers:
  hosts:
    web01:
      vmid: 101
    db01:
      vmid: 102
    web02:
      vmid: 101
      node: node02

or when you work with children:

servers:
  children:
    servers_nestorix:
    servers_obelix:

servers_obelix:
  hosts:
    test:
      vmid: 110
      node: obelix
      # cores: 4
      # memory: 8192

servers_nestorix:
  hosts:
    rke2:
      vmid: 200
      node: nestorix

VM speficic variables

You can override the proxmox_cores and proxmox_memory variables by adding the desired specs to the specific host. Same for a different datastore or template to use:

servers:
  hosts:
    k3s01:
      vmid: 131
      node: node01          # Optional: specify which Proxmox node to use
      datastore: 'vmstore01'
      cores: 4
      memory: 4096
      template: 1011
    k3s02:
      vmid: 132
    k3s03:
      vmid: 133

Node specific (default) variables

proxmox_subnet: 10.0.0.
proxmox_gateway: 10.0.0.1
proxmox_api_host_ip: 10.0.0.10
proxmox_api_hostname: nodename
proxmox_api_user: root@pam
proxmox_api_password:
proxmox_default_storagepool:

vmid for default template:

proxmox_default_template: 2009

Default VM specifications:

proxmox_cores: 1
proxmox_memory: 2048 # in MB

Allow node configuration for scaling governor, packages and datastores:

proxmox_allow_node_configuration: true
proxmox_scaling_governor: "powersave"
proxmox_datastore_default_fs: ext4

Datastore configuration:

Example of default datastore of an existing folder/mount:

proxmox_datastores:
  - name: vmstore01
    path: /media/vmstore01
    content: backup images (defaults to images)
    retention: 'keep-all=1'
  - name: vmstore02
    path: /media/vmstore02

The retention variable is additional and can be used for configuring backup retention.

Also adding a new disk including mount configuration use the following configuration:

proxmox_datastores:
  - name: vmstore01
    path: /media/vmstore01
    dev: /dev/pve/data
    retention: 'keep-all=1'
  - name: vmstore02
    path: /media/vmstore02
    dev: /dev/nvme0n1p1

Known issues

Pre-install

When having a Nvidia GPU you need to start the installation with nomodeset:

• select install method
• type 'e'
• add: nomodeset to linux-line

Cloudinit

The templates you build must use cloudinit inside your proxmox installation and have qemu-qa installed. To be able to set hostname etc. configure qemu-qa with;

sed -i "s/.*BLACKLIST_RPC.*/#BLACKLIST_RPC=guest-file-open,guest-file-close,guest-file-read,guest-file-write,guest-file-seek,guest-file-flush,guest-exec,guest-exec-status/g" /etc/sysconfig/qemu-ga

License

gnu-gpl-v3.0