Molecule testing

Requirements (Client)

Molecule and pyone are installed as Python/pip dependencies for one-deploy:

~/one-deploy$ cat requirements.txt
netaddr
molecule
pyone>=6.8.0

You can install all one-deploy requirements using the included Makefile:

~/one-deploy $ make requirements

Or directly:

~/one-deploy $ pip3 install -r requirements.txt
~/one-deploy $ ansible-galaxy collection install -r requirements.yml

Requirements (Server)

To deploy Molecule integration environments (ceph-hci, passenger-ha, ...) you will need a working OpenNebula instance with pre-configured networking and pre-installed OS images and VM templates.

[!NOTE] You can download OS images and VM templates from OpenNebula's Public Marketplace.

Download (and rename) VM templates and OS images into your OpenNebula instance:

• Ubuntu 22.04 -> ubuntu2204
• Ubuntu 20.04 -> ubuntu2004
• AlmaLinux 9 -> alma9
• AlmaLinux 8 -> alma8

Configuration

Create the .env.yml file in your one-deploy's root directory, with the following content (you can use .env.yml.sample as a template):

ONE_HOST: http://localhost:2633/RPC2
ONE_USER: oneadmin
ONE_PSWD: opennebula
ONE_TOKEN: example:example
ONE_VNET: service
ONE_SUBNET: 172.20.0.0/16
ONE_RANGE1: 172.20.86.100 4
ONE_RANGE2: 172.20.86.104 4
ONE_RANGE3: 172.20.86.108 4

• Provide OpenNebula endpoint and credentials.
• Provide your support token (required for the prometheus-ha environment).
• Customize VNET name, subnet and IP ranges.

Converging Environments

To list all available environments/tests:

molecule list

~/one-deploy $ molecule list
INFO     Running ceph-hci > list
INFO     Running passenger-ha > list
INFO     Running prometheus-ha > list
                   ╷             ╷                  ╷               ╷         ╷
  Instance Name    │ Driver Name │ Provisioner Name │ Scenario Name │ Created │ Converged
╶──────────────────┼─────────────┼──────────────────┼───────────────┼─────────┼───────────╴
  ceph-hci-f1      │ default     │ ansible          │ ceph-hci      │ false   │ false
  ceph-hci-n1      │ default     │ ansible          │ ceph-hci      │ false   │ false
  ceph-hci-n2      │ default     │ ansible          │ ceph-hci      │ false   │ false
  passenger-ha-f1  │ default     │ ansible          │ passenger-ha  │ false   │ false
  passenger-ha-f2  │ default     │ ansible          │ passenger-ha  │ false   │ false
  passenger-ha-f3  │ default     │ ansible          │ passenger-ha  │ false   │ false
  passenger-ha-f4  │ default     │ ansible          │ passenger-ha  │ false   │ false
  prometheus-ha-f1 │ default     │ ansible          │ prometheus-ha │ false   │ false
  prometheus-ha-f2 │ default     │ ansible          │ prometheus-ha │ false   │ false
  prometheus-ha-f3 │ default     │ ansible          │ prometheus-ha │ false   │ false
  prometheus-ha-f4 │ default     │ ansible          │ prometheus-ha │ false   │ false

To deploy the ceph-hci environment:

~/one-deploy $ molecule converge -s ceph-hci

If everything goes well, you should be able to log in and examine the environment:

~/one-deploy $ molecule list -s ceph-hci
INFO     Running ceph-hci > list
                ╷             ╷                  ╷               ╷         ╷
  Instance Name │ Driver Name │ Provisioner Name │ Scenario Name │ Created │ Converged
╶───────────────┼─────────────┼──────────────────┼───────────────┼─────────┼───────────╴
  ceph-hci-f1   │ default     │ ansible          │ ceph-hci      │ true    │ true
  ceph-hci-n1   │ default     │ ansible          │ ceph-hci      │ true    │ true
  ceph-hci-n2   │ default     │ ansible          │ ceph-hci      │ true    │ true

$ onevm ssh ceph-hci-f1

Destroying environments

To destroy a specific environment:

~/one-deploy $ molecule destroy -s ceph-hci

~/one-deploy $ molecule list -s ceph-hci
INFO     Running ceph-hci > list
                ╷             ╷                  ╷               ╷         ╷
  Instance Name │ Driver Name │ Provisioner Name │ Scenario Name │ Created │ Converged
╶───────────────┼─────────────┼──────────────────┼───────────────┼─────────┼───────────╴
  ceph-hci-f1   │ default     │ ansible          │ ceph-hci      │ false   │ false
  ceph-hci-n1   │ default     │ ansible          │ ceph-hci      │ false   │ false
  ceph-hci-n2   │ default     │ ansible          │ ceph-hci      │ false   │ false

To destroy all available environments:

~/one-deploy $ molecule destroy --all