personal_infra/ansible/infra/nodito/32_zfs_pool_setup_playbook.yml

- name: Setup ZFS RAID 1 Pool for Proxmox Storage
  hosts: hypervisor
  become: true

  tasks:
    - name: Verify Proxmox VE is running
      command: pveversion
      register: pve_version_check
      changed_when: false
      failed_when: pve_version_check.rc != 0

    - name: Update package cache
      apt:
        update_cache: yes
        cache_valid_time: 3600

    - name: Install ZFS utilities
      package:
        name:
          - zfsutils-linux
          - zfs-initramfs
        state: present

    - name: Load ZFS kernel module
      modprobe:
        name: zfs

    - name: Ensure ZFS module loads at boot
      lineinfile:
        path: /etc/modules
        line: zfs
        state: present

    - name: Check if ZFS pool already exists
      command: zpool list {{ zfs_pool_name }}
      register: zfs_pool_exists
      failed_when: false
      changed_when: false

    - name: Check if disks are in use
      shell: |
        for disk in {{ zfs_disk_1 }} {{ zfs_disk_2 }}; do
          if mount | grep -q "^$disk"; then
            echo "ERROR: $disk is mounted"
            exit 1
          fi
          if lsblk -n -o MOUNTPOINT "$disk" | grep -v "^$" | grep -q .; then
            echo "ERROR: $disk has mounted partitions"
            exit 1
          fi
        done
      register: disk_usage_check
      failed_when: disk_usage_check.rc != 0
      changed_when: false

    - name: Create ZFS RAID 1 pool with optimized settings
      command: >
        zpool create {{ zfs_pool_name }}
        -o ashift=12
        -O mountpoint=none
        mirror {{ zfs_disk_1 }} {{ zfs_disk_2 }}
      when: zfs_pool_exists.rc != 0
      register: zfs_pool_create_result

    - name: Check if ZFS dataset already exists
      command: zfs list {{ zfs_pool_name }}/vm-storage
      register: zfs_dataset_exists
      failed_when: false
      changed_when: false

    - name: Create ZFS dataset for Proxmox storage
      command: zfs create {{ zfs_pool_name }}/vm-storage
      when: zfs_dataset_exists.rc != 0
      register: zfs_dataset_create_result

    - name: Set ZFS dataset properties for Proxmox
      command: zfs set {{ item.property }}={{ item.value }} {{ zfs_pool_name }}/vm-storage
      loop:
        - { property: "mountpoint", value: "{{ zfs_pool_mountpoint }}" }
        - { property: "compression", value: "lz4" }
        - { property: "atime", value: "off" }
        - { property: "xattr", value: "sa" }
        - { property: "acltype", value: "posixacl" }
        - { property: "dnodesize", value: "auto" }
      when: zfs_dataset_exists.rc != 0

    - name: Set ZFS pool properties for Proxmox
      command: zpool set autotrim=off {{ zfs_pool_name }}
      when: zfs_pool_exists.rc != 0

    - name: Set ZFS pool mountpoint for Proxmox
      command: zfs set mountpoint={{ zfs_pool_mountpoint }} {{ zfs_pool_name }}
      when: zfs_pool_exists.rc == 0

    - name: Export and re-import ZFS pool for Proxmox compatibility
      shell: |
        zpool export {{ zfs_pool_name }}
        zpool import {{ zfs_pool_name }}
      when: zfs_pool_exists.rc != 0
      register: zfs_pool_import_result


    - name: Ensure ZFS services are enabled
      systemd:
        name: "{{ item }}"
        enabled: yes
        state: started
      loop:
        - zfs-import-cache
        - zfs-import-scan
        - zfs-mount
        - zfs-share
        - zfs-zed

    - name: Check if ZFS pool storage already exists in Proxmox config
      stat:
        path: /etc/pve/storage.cfg
      register: storage_cfg_file

    - name: Check if storage name exists in Proxmox config
      shell: "grep -q '^zfspool: {{ zfs_pool_name }}' /etc/pve/storage.cfg"
      register: storage_exists_check
      failed_when: false
      changed_when: false
      when: storage_cfg_file.stat.exists

    - name: Set storage not configured when config file doesn't exist
      set_fact:
        storage_exists_check:
          rc: 1
      when: not storage_cfg_file.stat.exists

    - name: Debug storage configuration status
      debug:
        msg: |
          Config file exists: {{ storage_cfg_file.stat.exists }}
          Storage check result: {{ storage_exists_check.rc }}
          Pool exists: {{ zfs_pool_exists.rc == 0 }}
          Will add storage: {{ zfs_pool_exists.rc == 0 and storage_exists_check.rc != 0 }}

    # Registration is add-only on purpose. There used to be a "Remove existing
    # storage if it exists" task here that ran `pvesm remove` whenever the
    # storage WAS present, paired with an add that only ran when it was ABSENT.
    # The two conditions are mutually exclusive, so a real run against a
    # correctly-configured hypervisor removed the storage entry backing every VM
    # and never put it back. It would also have dropped `mountpoint /var/lib/vz`,
    # which the live entry has and which `pvesm add` below does not set.
    #
    # If the storage entry ever needs its options changed, edit
    # /etc/pve/storage.cfg or use `pvesm set` - do not re-register it from here.
    - name: Add ZFS pool storage to Proxmox using pvesm
      command: >
        pvesm add zfspool {{ zfs_pool_name }}
        --pool {{ zfs_pool_name }}
        --content rootdir,images
        --sparse 1
      when: 
        - zfs_pool_exists.rc == 0
        - storage_exists_check.rc != 0
      register: pvesm_add_result

    - name: Verify ZFS pool is healthy
      command: zpool status {{ zfs_pool_name }}
      register: final_zfs_status
      changed_when: false

    - name: Fail if ZFS pool is not healthy
      fail:
        msg: "ZFS pool {{ zfs_pool_name }} is not in a healthy state"
      when: "'ONLINE' not in final_zfs_status.stdout"

# ─────────────────────────────────────────────────────────────────────────────
# The monthly scrub.
#
# The ZFS HEALTH CHECK that used to share this play is gone: it is now the
# zfs-health check in infra/400_host_monitoring.yml, which carries the same five
# conditions - pool state, device states, resilver in progress, read/write/
# checksum errors, and errors from the last scan - but reports to Gatus like
# every other host check instead of owning its own push plumbing.
#
# The scrub itself stays here, because it is not monitoring: it is the
# maintenance that gives the health check something true to report. A pool that
# is never scrubbed has no idea whether it is healthy.
# ─────────────────────────────────────────────────────────────────────────────
- name: Schedule the monthly ZFS scrub
  hosts: hypervisor
  become: true
  vars_files:
    - ../../infra_vars.yml

  vars:
    zfs_systemd_scrub_service_name: zfs-monthly-scrub

  tasks:
    - name: Create systemd service for ZFS monthly scrub
      template:
        src: templates/zfs-monthly-scrub.service.j2
        dest: "/etc/systemd/system/{{ zfs_systemd_scrub_service_name }}.service"
        owner: root
        group: root
        mode: '0644'

    - name: Create systemd timer for monthly ZFS scrub
      template:
        src: templates/zfs-monthly-scrub.timer.j2
        dest: "/etc/systemd/system/{{ zfs_systemd_scrub_service_name }}.timer"
        owner: root
        group: root
        mode: '0644'

    - name: Enable and start the monthly scrub timer
      systemd:
        name: "{{ zfs_systemd_scrub_service_name }}.timer"
        enabled: yes
        state: started
        daemon_reload: yes

    - name: Report the scrub schedule
      debug:
        msg: >-
          Monthly scrub of {{ zfs_pool_name }}:
          last day of each month at 04:00.
          Health is reported separately by the zfs-health check
          (infra/400_host_monitoring.yml).