---
# Prepare biscayne host for agave validator
#
# Deployment layers:
#   1. Base system    — Docker, ZFS (out of scope)
#   2. Prepare kind   — /srv/kind directory exists (ZFS dataset, out of scope)
#   3. laconic-so     — Installs kind, mounts /srv/kind → /mnt in kind node
#   4. Prepare agave  — THIS PLAYBOOK
#   5. Deploy agave   — laconic-so deploys agave-stack into kind
#
# Agave requires three things from the host that kind doesn't provide:
#
#   Invariant 1: /srv/kind/solana is XFS on a zvol (not ZFS)
#     Why: agave uses io_uring for async I/O. io_uring workers deadlock on
#     ZFS datasets (D-state in dsl_dir_tempreserve_space). XFS on a zvol
#     (block device) works fine. /srv/solana is NOT the zvol — it's a
#     directory on the ZFS dataset biscayne/DATA/srv. All data paths must
#     use /srv/kind/solana which is the actual zvol mount.
#     Persisted as: fstab entry mounting /dev/zvol/.../solana at /srv/kind/solana
#
#   Invariant 2: /srv/kind/solana/ramdisk is tmpfs (1TB)
#     Why: agave accounts must be in RAM for performance. tmpfs survives
#     process restarts but not host reboots (same as /dev/ram0 but simpler).
#     Persisted as: fstab entry (no format service needed)
#
# This playbook checks each invariant and only acts if it's not met.
# Idempotent — safe to run multiple times.
#
# Usage:
#   ansible-playbook playbooks/biscayne-prepare-agave.yml
#
- name: Configure OS-level services for agave
  hosts: all
  gather_facts: false
  become: true
  vars:
    zvol_device: /dev/zvol/biscayne/DATA/volumes/solana
    kind_solana_dir: /srv/kind/solana
    ramdisk_mount: /srv/kind/solana/ramdisk
    ramdisk_size: 1024G
    accounts_dir: /srv/kind/solana/ramdisk/accounts
    deployment_dir: /srv/deployments/agave

  tasks:
    # ---- cleanup legacy ramdisk services -----------------------------------------
    - name: Stop and disable legacy ramdisk services
      ansible.builtin.systemd:
        name: "{{ item }}"
        state: stopped
        enabled: false
      loop:
        - format-ramdisk.service
        - ramdisk-accounts.service
      failed_when: false

    - name: Remove legacy ramdisk service files
      ansible.builtin.file:
        path: "/etc/systemd/system/{{ item }}"
        state: absent
      loop:
        - format-ramdisk.service
        - ramdisk-accounts.service
      register: legacy_units_removed

    # ---- fstab entries ----------------------------------------------------------
    # /srv/solana is NOT the zvol — it's a directory on the ZFS dataset.
    # All data paths use /srv/kind/solana (the actual zvol mount).
    - name: Remove stale /srv/solana zvol fstab entry
      ansible.builtin.lineinfile:
        path: /etc/fstab
        regexp: '^\S+\s+/srv/solana\s+xfs'
        state: absent

    - name: Remove stale /srv/solana/ramdisk fstab entry
      ansible.builtin.lineinfile:
        path: /etc/fstab
        regexp: '^/dev/ram0\s+'
        state: absent

    - name: Remove stale kind rbind fstab entry
      ansible.builtin.lineinfile:
        path: /etc/fstab
        regexp: '^\S+\s+{{ kind_solana_dir }}\s+none\s+rbind'
        state: absent

    - name: Ensure zvol fstab entry
      ansible.builtin.lineinfile:
        path: /etc/fstab
        regexp: '^\S+\s+{{ kind_solana_dir }}\s'
        line: '{{ zvol_device }} {{ kind_solana_dir }} xfs defaults,nofail,x-systemd.requires=zfs-mount.service 0 0'
      register: fstab_zvol

    - name: Ensure tmpfs ramdisk fstab entry
      ansible.builtin.lineinfile:
        path: /etc/fstab
        regexp: '^\S+\s+{{ ramdisk_mount }}\s'
        line: "tmpfs {{ ramdisk_mount }} tmpfs nodev,nosuid,noexec,nodiratime,size={{ ramdisk_size }},nofail,x-systemd.requires=srv-kind-solana.mount 0 0"
      register: fstab_ramdisk

    # ---- reload systemd if anything changed --------------------------------------
    - name: Reload systemd
      ansible.builtin.systemd:
        daemon_reload: true
      when: legacy_units_removed.changed or fstab_zvol.changed or fstab_ramdisk.changed

    # ---- apply device mounts now if not correct ----------------------------------
    - name: Check kind zvol mount is XFS
      ansible.builtin.shell:
        cmd: >
          set -o pipefail &&
          findmnt -n -o FSTYPE {{ kind_solana_dir }} | grep -q xfs
        executable: /bin/bash
      register: kind_zvol_check
      failed_when: false
      changed_when: false

    - name: Unmount stale mounts
      ansible.builtin.shell:
        cmd: |
          umount {{ ramdisk_mount }} 2>/dev/null || true
          umount {{ kind_solana_dir }} 2>/dev/null || true
        executable: /bin/bash
      changed_when: kind_zvol_check.rc != 0
      when: kind_zvol_check.rc != 0

    - name: Mount zvol
      ansible.posix.mount:
        path: "{{ kind_solana_dir }}"
        src: "{{ zvol_device }}"
        fstype: xfs
        state: mounted
      when: kind_zvol_check.rc != 0

    - name: Check ramdisk mount is tmpfs
      ansible.builtin.shell:
        cmd: >
          set -o pipefail &&
          findmnt -n -o FSTYPE {{ ramdisk_mount }} | grep -q tmpfs
        executable: /bin/bash
      register: ramdisk_check
      failed_when: false
      changed_when: false

    - name: Mount tmpfs ramdisk
      ansible.posix.mount:
        path: "{{ ramdisk_mount }}"
        src: tmpfs
        fstype: tmpfs
        opts: "nodev,nosuid,noexec,nodiratime,size={{ ramdisk_size }}"
        state: mounted
      when: ramdisk_check.rc != 0

    - name: Create accounts directory
      ansible.builtin.file:
        path: "{{ accounts_dir }}"
        state: directory
        owner: solana
        group: solana
        mode: "0755"

    # Docker requires shared propagation on mounts it bind-mounts into
    # containers. Without this, `docker start` fails with "not a shared
    # or slave mount".
    # No ansible module supports mount propagation flags; command required.
    - name: Ensure shared propagation on kind mounts  # noqa: command-instead-of-module
      ansible.builtin.command:
        cmd: mount --make-shared {{ item }}
      loop:
        - "{{ kind_solana_dir }}"
        - "{{ kind_solana_dir }}/ramdisk"
      changed_when: false

    # ---- verify -----------------------------------------------------------------
    - name: Verify zvol is XFS
      ansible.builtin.shell:
        cmd: set -o pipefail && df -T {{ kind_solana_dir }} | grep -q xfs
        executable: /bin/bash
      changed_when: false

    - name: Verify ramdisk is tmpfs
      ansible.builtin.shell:
        cmd: set -o pipefail && df -T {{ ramdisk_mount }} | grep -q tmpfs
        executable: /bin/bash
      changed_when: false

    - name: Verify mount contents
      ansible.builtin.shell:
        cmd: >
          set -o pipefail &&
          ls {{ kind_solana_dir }}/ledger {{ kind_solana_dir }}/snapshots
          {{ ramdisk_mount }}/accounts 2>&1 | head -5
        executable: /bin/bash
      register: kind_mount_verify
      changed_when: false

    # Assert the kind node sees XFS at the PV mount paths.
    # laconic-so creates individual extraMounts per volume:
    #   /srv/kind/solana/ledger → /mnt/validator-ledger (inside kind node)
    #   /srv/kind/solana/ramdisk/accounts → /mnt/validator-accounts
    # The PV hostPaths use /mnt/<volume-name>, not /mnt/solana/<subpath>.
    - name: Read cluster-id from deployment
      ansible.builtin.shell:
        cmd: set -o pipefail && grep '^cluster-id:' {{ deployment_dir }}/deployment.yml | awk '{print $2}'
        executable: /bin/bash
      register: cluster_id_result
      changed_when: false

    - name: Check kind node filesystem visibility
      ansible.builtin.shell:
        cmd: >
          set -o pipefail &&
          docker exec {{ cluster_id_result.stdout }}-control-plane
          df -T /mnt/validator-ledger /mnt/validator-accounts
        executable: /bin/bash
      register: kind_fstype
      changed_when: false
      failed_when: false

    - name: Show status
      ansible.builtin.debug:
        msg:
          kind_mount: "{{ kind_mount_verify.stdout_lines }}"
          kind_fstype: "{{ kind_fstype.stdout_lines | default([]) }}"

- name: Configure Ashburn validator relay
  ansible.builtin.import_playbook: ashburn-relay-biscayne.yml