Migration to Grafana Cloud, nuremberg-a reinstalled, london-a reinsta… (#93)

* Migration to Grafana Cloud, nuremberg-a reinstalled, london-a reinstalled

* dns config for cockpit

ansible/deploy.yml

@@ -11,7 +11,7 @@
 #
 # Prerequisites:
 #   - Target host has SSH access via Tailscale
-#   - Target host has a base OS installed (Debian/Alpine/FreeBSD)
+#   - Target host has a base OS installed (Debian/FreeBSD)
 #   - ansible-galaxy install -r requirements.yml
 
 # ──────────────────────────────────────────────
@@ -33,42 +33,6 @@
   roles:
     - role: docker
 
-# ──────────────────────────────────────────────
-# Stage 3: Monitoring agent — all hosts
-# ──────────────────────────────────────────────
-- name: "Stage 3: Node exporter"
-  hosts: all
-  tags: [monitoring, node_exporter]
-  roles:
-    - role: node_exporter
-
-# ──────────────────────────────────────────────
-# Stage 3b: systemd_exporter — Linux hosts with systemd metrics
-# ──────────────────────────────────────────────
-- name: "Stage 3b: systemd_exporter"
-  hosts: systemd_exporter_hosts
-  tags: [monitoring, systemd_exporter]
-  roles:
-    - role: systemd_exporter
-
-# ──────────────────────────────────────────────
-# Stage 3c: Alloy — all hosts (log shipping agent)
-# ──────────────────────────────────────────────
-- name: "Stage 3c: Alloy"
-  hosts: alloy_hosts
-  tags: [monitoring, alloy]
-  roles:
-    - role: alloy
-
-# ──────────────────────────────────────────────
-# Stage 3d: Loki — london-a (log aggregation server)
-# ──────────────────────────────────────────────
-- name: "Stage 3d: Loki"
-  hosts: london-a
-  tags: [monitoring, loki]
-  roles:
-    - role: loki
-
 # ──────────────────────────────────────────────
 # Stage 4: Per-host services
 # ──────────────────────────────────────────────
@@ -89,12 +53,10 @@
   roles:
     - role: docker_services
 
-# nuremberg-a: Firewall
+# nuremberg-a: Mail server
-- name: "Stage 4c: Firewall (nuremberg-a)"
+- name: "Stage 4c: Mail server (nuremberg-a)"
   hosts: nuremberg-a
   tags: [services, mail]
-  roles:
-    - role: firewall_alpine
 
 # london-b: Media stack + backups
 - name: "Stage 4d: Media stack + backups (london-b)"
@@ -112,99 +74,29 @@
     - role: systemd_services
     - role: mariadb
 
-# london-a: Monitoring stack (FreeBSD — Prometheus, Grafana)
+# london-a: Cockpit VM host (Debian)
-# Note: london-a uses FreeBSD; monitoring roles handle this via conditionals.
+- name: "Stage 4f: Cockpit VM host (london-a)"
-- name: "Stage 4e: Monitoring stack (london-a)"
   hosts: london-a
-  tags: [services, monitoring]
+  tags: [services, cockpit]
   tasks:
-    - name: Check for Prometheus config
+    - name: Install cockpit and cockpit-machines
-      delegate_to: localhost
+      ansible.builtin.apt:
-      ansible.builtin.stat:
+        name:
-        path: "{{ playbook_dir }}/services/prometheus/prometheus.yml"
+          - cockpit
-      register: prometheus_config
+          - cockpit-machines
+        state: present
+        update_cache: true
 
-    - name: Deploy Prometheus config
+    - name: Enable and start cockpit
-      ansible.builtin.copy:
-        src: "{{ playbook_dir }}/services/prometheus/prometheus.yml"
-        dest: /usr/local/etc/prometheus.yml
-        mode: '0644'
-        backup: true
-      when: prometheus_config.stat.exists
-      notify: Restart prometheus
-
-    - name: Deploy Prometheus alerting rules
-      ansible.builtin.copy:
-        src: "{{ playbook_dir }}/services/prometheus/rules/"
-        dest: /usr/local/etc/prometheus/rules/
-        mode: '0644'
-      failed_when: false
-      notify: Restart prometheus
-
-    - name: Ensure unified_alerting section exists in Grafana config
-      ansible.builtin.lineinfile:
-        path: /usr/local/etc/grafana/grafana.ini
-        regexp: '^\[unified_alerting\]'
-        line: '[unified_alerting]'
-      notify: Restart grafana
-
-    - name: Allow provenance status change in Grafana
-      ansible.builtin.lineinfile:
-        path: /usr/local/etc/grafana/grafana.ini
-        regexp: '^allow_prov_status_change'
-        insertafter: '^\[unified_alerting\]'
-        line: 'allow_prov_status_change = true'
-      notify: Restart grafana
-
-    - name: Deploy Grafana dashboards
-      ansible.posix.synchronize:
-        src: "{{ playbook_dir }}/services/grafana/dashboards/"
-        dest: /usr/local/etc/grafana/dashboards/
-      failed_when: false
-
-    - name: Ensure provisioning dir exists
-      ansible.builtin.file:
-        path: "{{ grafana_provisioning_dir }}"
-        state: directory
-        mode: '0755'
-
-    - name: Ensure alerting dir exists
-      ansible.builtin.file:
-        path: "{{ grafana_provisioning_dir }}/alerting"
-        state: directory
-        mode: '0755'
-
-    - name: Deploy Grafana provisioning
-      ansible.posix.synchronize:
-        src: "{{ playbook_dir }}/services/grafana/provisioning/"
-        dest: "{{ grafana_provisioning_dir }}/"
-      failed_when: false
-
-    - name: Template contact points with PagerDuty key
-      ansible.builtin.template:
-        src: "{{ playbook_dir }}/services/grafana/provisioning/alerting/contact-points.yml"
-        dest: "{{ grafana_provisioning_dir }}/alerting/contact-points.yml"
-        mode: '0640'
-        owner: root
-        group: grafana
-      no_log: true
-      notify: Restart grafana
-
-  handlers:
-    - name: Restart prometheus
       ansible.builtin.service:
-        name: prometheus
+        name: cockpit
-        state: restarted
+        state: started
-
+        enabled: true
-    - name: Restart grafana
-      ansible.builtin.service:
-        name: grafana
-        state: restarted
 
 # ──────────────────────────────────────────────
-# Stage 4f: ZFS scrub scheduling — zfs_hosts
+# Stage 4g: ZFS scrub scheduling — zfs_hosts
 # ──────────────────────────────────────────────
-- name: "Stage 4f: ZFS scrub scheduling"
+- name: "Stage 4g: ZFS scrub scheduling"
   hosts: zfs_hosts
   tags: [services, zfs]
   roles:

ansible/inventory/host_vars/helsinki-a.yml

@@ -11,7 +11,6 @@ docker_services:
   - forgejo
   - bitwarden
 
-node_exporter_bind_tailscale: true
 
 common_ufw_allowed_ports:
   - { port: 80, proto: tcp, comment: "HTTP" }

ansible/inventory/host_vars/london-a.yml

@@ -1,27 +1,5 @@
 ---
-node_exporter_bind_tailscale: true
+host_role: vm-host
-host_role: monitoring
+host_description: "VM host (Cockpit + KVM)"
-host_description: "Monitoring stack (Prometheus, Grafana)"
 host_location: "London"
-prometheus_location: london
+ansible_python_interpreter: /usr/bin/python3
-ansible_python_interpreter: /usr/local/bin/python3
-grafana_provisioning_dir: /usr/local/etc/grafana/provisioning
-
-zfs_pools:
-  - zroot
-
-# ZFS scrub schedule (currently manual cron, not yet managed by ansible)
-# 0 12 * * sun zpool scrub zroot
-zfs_scrub_schedule: "0 12 * * 0"
-
-alloy_loki_url: "http://localhost:3100/loki/api/v1/push"
-
-# --- Services enabled in rc.conf ---
-
-# Core services (documented)
-# sshd, ntpd, powerd, zfs, tailscaled, grafana, prometheus, node_exporter, loki, alloy
-
-# --- Disabled/removed services ---
-# cloudflared — removed 2026-04-03 (PESO-134). Replaced by Caddy + Authelia.
-# InfluxDB, Redis, PostgreSQL, libvirtd — disabled 2026-04-02 (PESO-113).
-# Were leftover from a defunct pez_vps project. Pez approved removal.

ansible/inventory/host_vars/london-b.yml

@@ -23,13 +23,6 @@ apt_user_services:
 zfs_pools:
   - hdd
 
-node_exporter_extra_collectors:
-  - systemd
-  - processes
-  - sysctl
-  - ethtool
-  - zfs
-
 docker_daemon_extra:
   metrics-addr: "0.0.0.0:9323"
   data-root: "/hdd/docker"

ansible/inventory/host_vars/nuremberg-a.yml

@@ -8,7 +8,6 @@ ansible_python_interpreter: /usr/bin/python3
 docker_services:
   - poste-io
 
-# NOTE: Alpine host — UFW tasks are Debian-only.
-# Firewall: iptables + fail2ban managed by firewall_alpine role.
 # Mail ports (25,80,110,143,443,465,587,993,995) exposed via Docker
 # port mappings in ansible/services/poste-io/docker-compose.yml.
+# Firewall: managed by Hetzner Cloud firewall rules (Terraform).

ansible/inventory/hosts.ini

@@ -3,26 +3,16 @@
 
 [linux]
 helsinki-a   ansible_host=100.67.6.27
+london-a     ansible_host=100.90.111.19
 london-b     ansible_host=100.84.65.101
 london-c     ansible_host=100.123.72.87
 copenhagen-a ansible_host=100.89.206.60
 copenhagen-c ansible_host=100.115.45.53
-
+nuremberg-a  ansible_host=100.70.180.24
-[alpine]
-nuremberg-a  ansible_host=100.117.235.28
-
-[freebsd]
-london-a     ansible_host=100.122.219.41
 
 [zfs_hosts]
-london-a
 london-b
 
-[systemd_exporter_hosts]
-london-b
-london-c
-copenhagen-a
-
 [docker_hosts]
 helsinki-a
 london-b
@@ -30,17 +20,5 @@ london-c
 nuremberg-a
 copenhagen-a
 
-[monitoring]
-london-a
-
-[alloy_hosts]
-helsinki-a
-london-b
-london-c
-copenhagen-a
-copenhagen-c
-nuremberg-a
-london-a
-
 [all:vars]
 ansible_user=root

ansible/playbooks/monitoring.yml

@@ -1,85 +0,0 @@
----
-# Deploy monitoring stack to london-a (Prometheus + Grafana).
-# Usage: ansible-playbook playbooks/monitoring.yml
-#        ansible-playbook playbooks/monitoring.yml --check --diff
-
-- name: "Monitoring stack (london-a)"
-  hosts: london-a
-  pre_tasks:
-    - name: Load secrets
-      ansible.builtin.include_vars:
-        file: "{{ playbook_dir }}/../group_vars/all/secrets.yaml"
-      no_log: true
-  roles:
-    - prometheus
-  tasks:
-    - name: Deploy Prometheus config
-      ansible.builtin.template:
-        src: "{{ playbook_dir }}/../services/prometheus/prometheus.yml.j2"
-        dest: /usr/local/etc/prometheus.yml
-        mode: '0644'
-        backup: true
-      notify: Restart prometheus
-
-    - name: Deploy Prometheus alerting rules
-      ansible.builtin.copy:
-        src: "{{ playbook_dir }}/../services/prometheus/rules/"
-        dest: /usr/local/etc/prometheus/rules/
-        mode: '0644'
-      failed_when: false
-      notify: Restart prometheus
-
-    - name: Ensure unified_alerting section exists in Grafana config
-      ansible.builtin.lineinfile:
-        path: /usr/local/etc/grafana/grafana.ini
-        regexp: '^\[unified_alerting\]'
-        line: '[unified_alerting]'
-      notify: Restart grafana
-
-    - name: Allow provenance status change in Grafana
-      ansible.builtin.lineinfile:
-        path: /usr/local/etc/grafana/grafana.ini
-        regexp: '^allow_prov_status_change'
-        insertafter: '^\[unified_alerting\]'
-        line: 'allow_prov_status_change = true'
-      notify: Restart grafana
-
-    - name: Deploy Grafana dashboards
-      ansible.posix.synchronize:
-        src: "{{ playbook_dir }}/../services/grafana/dashboards/"
-        dest: /usr/local/etc/grafana/dashboards/
-      failed_when: false
-
-    - name: Ensure provisioning dir exists
-      ansible.builtin.file:
-        path: "{{ grafana_provisioning_dir }}"
-        state: directory
-        mode: '0755'
-
-    - name: Ensure alerting dir exists
-      ansible.builtin.file:
-        path: "{{ grafana_provisioning_dir }}/alerting"
-        state: directory
-        mode: '0755'
-
-    - name: Deploy Grafana provisioning
-      ansible.posix.synchronize:
-        src: "{{ playbook_dir }}/../services/grafana/provisioning/"
-        dest: "{{ grafana_provisioning_dir }}/"
-      failed_when: false
-
-    - name: Template contact points with PagerDuty key
-      ansible.builtin.template:
-        src: "{{ playbook_dir }}/../services/grafana/provisioning/alerting/contact-points.yml"
-        dest: "{{ grafana_provisioning_dir }}/alerting/contact-points.yml"
-        mode: '0640'
-        owner: root
-        group: grafana
-      no_log: true
-      notify: Restart grafana
-
-  handlers:
-    - name: Restart grafana
-      ansible.builtin.service:
-        name: grafana
-        state: restarted

ansible/roles/alloy/defaults/main.yml

@@ -1,2 +0,0 @@
----
-alloy_loki_url: "http://{{ hostvars['london-a']['ansible_host'] }}:3100/loki/api/v1/push"

ansible/roles/alloy/handlers/main.yml

@@ -1,18 +0,0 @@
----
-- name: Restart alloy (Debian)
-  ansible.builtin.service:
-    name: alloy
-    state: restarted
-  listen: "Restart alloy (Debian)"
-
-- name: Restart alloy (Alpine)
-  ansible.builtin.service:
-    name: alloy
-    state: restarted
-  listen: "Restart alloy (Alpine)"
-
-- name: Restart alloy (FreeBSD)
-  ansible.builtin.service:
-    name: alloy
-    state: restarted
-  listen: "Restart alloy (FreeBSD)"

ansible/roles/alloy/tasks/main.yml

@@ -1,101 +0,0 @@
----
-# Install and configure Grafana Alloy log shipping agent.
-# Debian/Ubuntu: alloy package (included in default repos).
-# Alpine: alloy package (included in default repos).
-# FreeBSD: pkgng (grafana-alloy).
-
-# ── Debian/Ubuntu ────────────────────────────────────────────────────────────
-
-- name: Install alloy (Debian)
-  ansible.builtin.apt:
-    name: alloy
-    state: present
-  when: ansible_facts["os_family"] == "Debian"
-
-# ── Alpine ───────────────────────────────────────────────────────────────────
-
-- name: Install alloy (Alpine)
-  community.general.apk:
-    name: alloy
-    state: present
-  when: ansible_facts["os_family"] == "Alpine"
-
-- name: Fix alloy storage dir ownership (Alpine)
-  ansible.builtin.file:
-    path: /var/lib/alloy
-    state: directory
-    owner: alloy
-    group: alloy
-    recurse: true
-  when: ansible_facts["os_family"] == "Alpine"
-
-# ── FreeBSD: pkgng ────────────────────────────────────────────────────────────
-
-- name: Install alloy (FreeBSD)
-  community.general.pkgng:
-    name: alloy
-    state: present
-  when: ansible_facts["os_family"] == "FreeBSD"
-
-- name: Fix alloy storage dir ownership (FreeBSD)
-  ansible.builtin.file:
-    path: /var/alloy
-    state: directory
-    owner: nobody
-    group: nobody
-    mode: '0755'
-  when: ansible_facts["os_family"] == "FreeBSD"
-
-# ── Docker socket access ─────────────────────────────────────────────────────
-
-- name: Add alloy to docker group
-  ansible.builtin.user:
-    name: alloy
-    groups: docker
-    append: true
-  when: "'docker_hosts' in group_names"
-  notify: "Restart alloy ({{ ansible_facts['os_family'] }})"
-
-# ── Config — all OS ───────────────────────────────────────────────────────────
-
-- name: Set alloy config path fact
-  ansible.builtin.set_fact:
-    alloy_config_path: >-
-      {{ '/usr/local/etc/alloy.flow'
-         if ansible_facts['os_family'] == 'FreeBSD'
-         else '/etc/alloy/config.alloy' }}
-
-- name: Deploy alloy config
-  ansible.builtin.template:
-    src: alloy.config.alloy.j2
-    dest: "{{ alloy_config_path }}"
-    mode: '0644'
-  notify: "Restart alloy ({{ ansible_facts['os_family'] }})"
-
-# ── Service enable + start ────────────────────────────────────────────────────
-
-- name: Enable and start alloy (Debian)
-  ansible.builtin.service:
-    name: alloy
-    state: started
-    enabled: true
-  when: ansible_facts["os_family"] == "Debian"
-
-- name: Enable and start alloy (Alpine)
-  ansible.builtin.service:
-    name: alloy
-    state: started
-    enabled: true
-  when: ansible_facts["os_family"] == "Alpine"
-
-- name: Enable alloy (FreeBSD)
-  community.general.sysrc:
-    name: alloy_enable
-    value: "YES"
-  when: ansible_facts["os_family"] == "FreeBSD"
-
-- name: Start alloy (FreeBSD)
-  ansible.builtin.service:
-    name: alloy
-    state: started
-  when: ansible_facts["os_family"] == "FreeBSD"

ansible/roles/alloy/templates/alloy.config.alloy.j2

@@ -1,68 +0,0 @@
-// Ansible managed — generated from alloy.config.alloy.j2
-// Grafana Alloy log shipping agent — {{ inventory_hostname }}
-
-// ─── System logs ─────────────────────────────────────────────────────────────
-
-{% if ansible_facts['os_family'] == 'Debian' %}
-loki.source.journal "system" {
-  forward_to = [loki.write.default.receiver]
-  labels     = {"host" = "{{ inventory_hostname }}"}
-  relabel_rules = loki.relabel.journal.rules
-}
-
-loki.relabel "journal" {
-  forward_to = []
-  rule {
-    source_labels = ["__journal__systemd_unit"]
-    target_label  = "unit"
-  }
-  rule {
-    source_labels = ["__journal_priority_keyword"]
-    target_label  = "level"
-  }
-}
-{% elif ansible_facts['os_family'] == 'Alpine' %}
-local.file_match "system" {
-  path_targets = [
-    {"__path__" = "/var/log/messages", "job" = "messages", "host" = "{{ inventory_hostname }}"},
-  ]
-}
-{% elif ansible_facts['os_family'] == 'FreeBSD' %}
-local.file_match "system" {
-  path_targets = [
-    {"__path__" = "/var/log/messages",  "job" = "syslog", "host" = "{{ inventory_hostname }}"},
-    {"__path__" = "/var/log/auth.log",  "job" = "auth",   "host" = "{{ inventory_hostname }}"},
-  ]
-}
-{% endif %}
-
-{% if ansible_facts['os_family'] != 'Debian' %}
-loki.source.file "system" {
-  targets    = local.file_match.system.targets
-  forward_to = [loki.write.default.receiver]
-}
-{% endif %}
-
-{% if inventory_hostname == 'london-b' %}
-// ─── london-b app logs ────────────────────────────────────────────────────────
-
-local.file_match "apps" {
-  path_targets = [
-    {"__path__" = "/var/lib/plexmediaserver/Library/Application Support/Plex Media Server/Logs/*.log", "job" = "plex",     "host" = "london-b"},
-    {"__path__" = "/var/log/jellyfin/*.log",                                                            "job" = "jellyfin", "host" = "london-b"},
-  ]
-}
-
-loki.source.file "apps" {
-  targets    = local.file_match.apps.targets
-  forward_to = [loki.write.default.receiver]
-}
-{% endif %}
-
-// ─── Loki output ──────────────────────────────────────────────────────────────
-
-loki.write "default" {
-  endpoint {
-    url = "{{ alloy_loki_url }}"
-  }
-}

ansible/roles/loki/defaults/main.yml

@@ -1,7 +0,0 @@
----
-loki_http_listen_port: 3100
-loki_grpc_listen_port: 9096
-loki_data_dir: /var/db/loki
-loki_retention_period: 720h
-loki_ingestion_rate_mb: 4
-loki_ingestion_burst_size_mb: 6

ansible/roles/loki/handlers/main.yml

@@ -1,5 +0,0 @@
----
-- name: Restart loki
-  ansible.builtin.service:
-    name: loki
-    state: restarted

ansible/roles/loki/tasks/main.yml

@@ -1,54 +0,0 @@
----
-# Install and configure Grafana Loki on FreeBSD (london-a).
-# Co-located with Prometheus and Grafana; all three run as native FreeBSD services.
-# FreeBSD only — Loki is the log aggregation backend for Promtail on all hosts.
-
-- name: Install loki (FreeBSD)
-  community.general.pkgng:
-    name: grafana-loki
-    state: present
-  when: ansible_facts["os_family"] == "FreeBSD"
-
-- name: Ensure Loki data directory exists
-  ansible.builtin.file:
-    path: "{{ loki_data_dir }}"
-    state: directory
-    mode: '0755'
-    owner: loki
-    group: loki
-  when: ansible_facts["os_family"] == "FreeBSD"
-
-- name: Ensure Loki config directory exists
-  ansible.builtin.file:
-    path: /usr/local/etc/loki
-    state: directory
-    mode: '0755'
-  when: ansible_facts["os_family"] == "FreeBSD"
-
-- name: Deploy Loki config
-  ansible.builtin.template:
-    src: loki.yml.j2
-    dest: /usr/local/etc/loki/config.yml
-    mode: '0644'
-    owner: root
-    group: wheel
-  when: ansible_facts["os_family"] == "FreeBSD"
-  notify: Restart loki
-
-- name: Enable loki (FreeBSD)
-  community.general.sysrc:
-    name: loki_enable
-    value: "YES"
-  when: ansible_facts["os_family"] == "FreeBSD"
-
-- name: Set loki config path in rc.conf (FreeBSD)
-  community.general.sysrc:
-    name: loki_config
-    value: /usr/local/etc/loki/config.yml
-  when: ansible_facts["os_family"] == "FreeBSD"
-
-- name: Start loki (FreeBSD)
-  ansible.builtin.service:
-    name: loki
-    state: started
-  when: ansible_facts["os_family"] == "FreeBSD"

ansible/roles/loki/templates/loki.yml.j2

@@ -1,51 +0,0 @@
-# Ansible managed — generated from loki.yml.j2
-# Grafana Loki — london-a (FreeBSD)
-# Single-node, filesystem storage, {{ loki_retention_period }} retention
-
-auth_enabled: false
-
-server:
-  http_listen_port: {{ loki_http_listen_port }}
-  grpc_listen_port: {{ loki_grpc_listen_port }}
-  log_level: info
-
-common:
-  instance_addr: 127.0.0.1
-  path_prefix: {{ loki_data_dir }}
-  storage:
-    filesystem:
-      chunks_directory: {{ loki_data_dir }}/chunks
-      rules_directory: {{ loki_data_dir }}/rules
-  replication_factor: 1
-  ring:
-    kvstore:
-      store: inmemory
-
-schema_config:
-  configs:
-    - from: 2024-01-01
-      store: tsdb
-      object_store: filesystem
-      schema: v13
-      index:
-        prefix: index_
-        period: 24h
-
-limits_config:
-  retention_period: {{ loki_retention_period }}
-  ingestion_rate_mb: {{ loki_ingestion_rate_mb }}
-  ingestion_burst_size_mb: {{ loki_ingestion_burst_size_mb }}
-
-compactor:
-  working_directory: {{ loki_data_dir }}/compactor
-  compaction_interval: 10m
-  retention_enabled: true
-  retention_delete_delay: 2h
-  delete_request_store: filesystem
-
-query_range:
-  results_cache:
-    cache:
-      embedded_cache:
-        enabled: true
-        max_size_mb: 100

ansible/roles/node_exporter/defaults/main.yml

@@ -1,8 +0,0 @@
----
-# When true, bind node_exporter to the Tailscale IP (ansible_host) only.
-# Use on public-facing hosts to avoid exposing metrics on 0.0.0.0.
-node_exporter_bind_tailscale: false
-
-# Extra collectors to enable beyond the defaults.
-# Each entry is a collector name (e.g. "systemd", "processes").
-node_exporter_extra_collectors: []

ansible/roles/node_exporter/handlers/main.yml

@@ -1,14 +0,0 @@
----
-- name: Reload systemd
-  ansible.builtin.systemd:
-    daemon_reload: true
-
-- name: Restart node-exporter (Debian)
-  ansible.builtin.service:
-    name: prometheus-node-exporter
-    state: restarted
-
-- name: Restart node_exporter (FreeBSD)
-  ansible.builtin.service:
-    name: node_exporter
-    state: restarted

ansible/roles/node_exporter/tasks/main.yml

@@ -1,101 +0,0 @@
----
-# Install node_exporter for Prometheus monitoring.
-# Uses system packages on Linux, pkg on FreeBSD.
-# Optionally binds to Tailscale IP on public-facing hosts.
-
-# ── Cleanup old custom installs ──────────────────────────────
-- name: Stop and disable custom node_exporter service if present
-  ansible.builtin.service:
-    name: node_exporter
-    state: stopped
-    enabled: false
-  failed_when: false
-  when: ansible_facts["os_family"] == "Debian"
-
-- name: Remove custom node_exporter service file
-  ansible.builtin.file:
-    path: /etc/systemd/system/node_exporter.service
-    state: absent
-  when: ansible_facts["os_family"] == "Debian"
-  notify: Reload systemd
-
-- name: Remove custom node_exporter binary
-  ansible.builtin.file:
-    path: /usr/local/bin/node_exporter
-    state: absent
-  when: ansible_facts["os_family"] == "Debian"
-
-# ── Install ──────────────────────────────────────────────────
-- name: Install prometheus-node-exporter (Debian)
-  ansible.builtin.apt:
-    name: prometheus-node-exporter
-    state: present
-  when: ansible_facts["os_family"] == "Debian"
-
-- name: Install prometheus-node-exporter (Alpine)
-  community.general.apk:
-    name: prometheus-node-exporter
-    state: present
-  when: ansible_facts["os_family"] == "Alpine"
-
-# ── Configure (Debian) ──────────────────────────────────────
-- name: Build ARGS for prometheus-node-exporter
-  ansible.builtin.set_fact:
-    _node_exporter_args: >-
-      {{ (node_exporter_extra_collectors | map('regex_replace', '^(.*)$', '--collector.\1') | list)
-         + (['--web.listen-address=' + ansible_host + ':9100'] if node_exporter_bind_tailscale | bool else []) }}
-  when: ansible_facts["os_family"] == "Debian"
-
-- name: Configure prometheus-node-exporter ARGS (Debian)
-  ansible.builtin.lineinfile:
-    path: /etc/default/prometheus-node-exporter
-    regexp: '^ARGS='
-    line: 'ARGS="{{ _node_exporter_args | join(" ") }}"'
-  when:
-    - ansible_facts["os_family"] == "Debian"
-    - (_node_exporter_args | length > 0)
-  notify: Restart node-exporter (Debian)
-
-- name: Enable and start node-exporter (Debian)
-  ansible.builtin.service:
-    name: prometheus-node-exporter
-    state: started
-    enabled: true
-  when: ansible_facts["os_family"] == "Debian"
-
-- name: Enable and start node-exporter (Alpine)
-  ansible.builtin.service:
-    name: node-exporter
-    state: started
-    enabled: true
-  when: ansible_facts["os_family"] == "Alpine"
-
-# ── FreeBSD ──────────────────────────────────────────────────
-- name: Install node_exporter (FreeBSD)
-  community.general.pkgng:
-    name: node_exporter
-    state: present
-  when: ansible_facts["os_family"] == "FreeBSD"
-
-- name: Enable node_exporter (FreeBSD)
-  ansible.builtin.lineinfile:
-    path: /etc/rc.conf
-    regexp: '^node_exporter_enable='
-    line: 'node_exporter_enable="YES"'
-  when: ansible_facts["os_family"] == "FreeBSD"
-
-- name: Configure listen address (FreeBSD)
-  ansible.builtin.lineinfile:
-    path: /etc/rc.conf
-    regexp: '^node_exporter_listen_address='
-    line: 'node_exporter_listen_address="{{ ansible_host }}:9100"'
-  when:
-    - ansible_facts["os_family"] == "FreeBSD"
-    - node_exporter_bind_tailscale | bool
-  notify: Restart node_exporter (FreeBSD)
-
-- name: Start node_exporter (FreeBSD)
-  ansible.builtin.service:
-    name: node_exporter
-    state: started
-  when: ansible_facts["os_family"] == "FreeBSD"

ansible/roles/prometheus/defaults/main.yml

@@ -1,2 +0,0 @@
----
-prometheus_retention_time: "1y"

ansible/roles/prometheus/handlers/main.yml

@@ -1,5 +0,0 @@
----
-- name: Restart prometheus
-  ansible.builtin.service:
-    name: prometheus
-    state: restarted

ansible/roles/prometheus/tasks/main.yml

@@ -1,7 +0,0 @@
----
-- name: Set Prometheus args in rc.conf (FreeBSD)
-  community.general.sysrc:
-    name: prometheus_args
-    value: "--storage.tsdb.retention.time={{ prometheus_retention_time }}"
-  when: ansible_facts["os_family"] == "FreeBSD"
-  notify: Restart prometheus

ansible/roles/systemd_exporter/defaults/main.yml

@@ -1,7 +0,0 @@
----
-systemd_exporter_version: "0.6.0"
-systemd_exporter_listen_address: "0.0.0.0"
-systemd_exporter_listen_port: 9558
-systemd_exporter_log_level: "info"
-systemd_exporter_user: "systemd-exporter"
-systemd_exporter_group: "systemd-exporter"

ansible/roles/systemd_exporter/handlers/main.yml

@@ -1,9 +0,0 @@
----
-- name: Reload systemd
-  ansible.builtin.systemd:
-    daemon_reload: true
-
-- name: Restart systemd_exporter
-  ansible.builtin.service:
-    name: systemd_exporter
-    state: restarted

ansible/roles/systemd_exporter/tasks/main.yml

@@ -1,98 +0,0 @@
----
-# Install and configure systemd_exporter for Prometheus monitoring.
-# Downloads the binary from GitHub releases and deploys a systemd service.
-# Linux only — systemd_exporter has no FreeBSD equivalent.
-
-- name: Create systemd_exporter group
-  ansible.builtin.group:
-    name: "{{ systemd_exporter_group }}"
-    system: true
-    state: present
-
-- name: Create systemd_exporter user
-  ansible.builtin.user:
-    name: "{{ systemd_exporter_user }}"
-    group: "{{ systemd_exporter_group }}"
-    system: true
-    shell: /usr/sbin/nologin
-    create_home: false
-
-- name: Check if systemd_exporter binary exists
-  ansible.builtin.stat:
-    path: /usr/local/bin/systemd_exporter
-  register: systemd_exporter_bin
-
-- name: Get installed version
-  ansible.builtin.command: /usr/local/bin/systemd_exporter --version
-  register: systemd_exporter_installed_version
-  changed_when: false
-  failed_when: false
-  when: systemd_exporter_bin.stat.exists
-
-- name: Set architecture fact
-  ansible.builtin.set_fact:
-    systemd_exporter_arch: >-
-      {{ ansible_facts['architecture']
-         | regex_replace('x86_64', 'amd64')
-         | regex_replace('aarch64', 'arm64') }}
-
-- name: Set release and URL facts
-  ansible.builtin.set_fact:
-    systemd_exporter_release: >-
-      systemd_exporter-{{ systemd_exporter_version }}.linux-{{ systemd_exporter_arch }}
-    systemd_exporter_base_url: >-
-      https://github.com/prometheus-community/systemd_exporter
-
-- name: Download and install systemd_exporter
-  when: >-
-    not systemd_exporter_bin.stat.exists or
-    systemd_exporter_version not in
-    (systemd_exporter_installed_version.stdout | default(''))
-  block:
-    - name: Download systemd_exporter tarball
-      ansible.builtin.get_url:
-        url: >-
-          {{ systemd_exporter_base_url }}/releases/download/v{{
-          systemd_exporter_version }}/{{
-          systemd_exporter_release }}.tar.gz
-        dest: /tmp/systemd_exporter.tar.gz
-        mode: '0644'
-
-    - name: Extract systemd_exporter binary
-      ansible.builtin.unarchive:
-        src: /tmp/systemd_exporter.tar.gz
-        dest: /tmp
-        remote_src: true
-
-    - name: Install systemd_exporter binary
-      ansible.builtin.copy:
-        src: "/tmp/{{ systemd_exporter_release }}/systemd_exporter"
-        dest: /usr/local/bin/systemd_exporter
-        mode: '0755'
-        owner: root
-        group: root
-        remote_src: true
-      notify: Restart systemd_exporter
-
-    - name: Clean up tarball
-      ansible.builtin.file:
-        path: "{{ item }}"
-        state: absent
-      loop:
-        - /tmp/systemd_exporter.tar.gz
-        - "/tmp/{{ systemd_exporter_release }}"
-
-- name: Deploy systemd_exporter service file
-  ansible.builtin.template:
-    src: systemd_exporter.service.j2
-    dest: /etc/systemd/system/systemd_exporter.service
-    mode: '0644'
-  notify:
-    - Reload systemd
-    - Restart systemd_exporter
-
-- name: Enable and start systemd_exporter
-  ansible.builtin.service:
-    name: systemd_exporter
-    state: started
-    enabled: true

ansible/roles/systemd_exporter/templates/systemd_exporter.service.j2

@@ -1,31 +0,0 @@
-#
-# Ansible managed
-#
-
-[Unit]
-Description=Prometheus SystemD Exporter
-After=network-online.target
-
-[Service]
-Type=simple
-User={{ systemd_exporter_user }}
-Group={{ systemd_exporter_group }}
-ExecStart=/usr/local/bin/systemd_exporter \
-    --log.level={{ systemd_exporter_log_level }} \
-    --web.listen-address={{ systemd_exporter_listen_address }}:{{ systemd_exporter_listen_port }}
-
-SyslogIdentifier=systemd_exporter
-Restart=always
-RestartSec=1
-StartLimitInterval=0
-
-ProtectHome=yes
-NoNewPrivileges=yes
-
-ProtectSystem=strict
-ProtectControlGroups=true
-ProtectKernelModules=true
-ProtectKernelTunables=yes
-
-[Install]
-WantedBy=multi-user.target

ansible/services/caddy/Caddyfile

@@ -14,22 +14,17 @@
 
 ## LONDON-A SERVICES ##
 
-# Grafana
+# Cockpit
-grafana.pez.solutions, grafana.pez.sh {
+london-a.pez.sh {
         forward_auth localhost:9091 {
 		uri /api/authz/forward-auth
 		copy_headers Remote-User Remote-Groups Remote-Name Remote-Email
 	}
-        reverse_proxy 100.122.219.41:3000
+        reverse_proxy 100.90.111.19:9090 {
-}
+                transport http {
-
+                        tls_insecure_skip_verify
-# Prometheus
+                }
-prometheus.pez.solutions, prometheus.pez.sh {
+        }
-	forward_auth localhost:9091 {
-		uri /api/authz/forward-auth
-		copy_headers Remote-User Remote-Groups Remote-Name Remote-Email
-	}
-        reverse_proxy 100.122.219.41:9090
 }
 
 ## LONDON-B SERVICES ##

docs/hosts/london-a.md

@@ -1,15 +1,15 @@
 # london-a
 
-Dedicated monitoring server. Runs Prometheus and Grafana, nothing else.
+VM host. Runs KVM virtual machines via Cockpit.
 
 ## Overview
 
 | | |
 |---|---|
 | **Location** | London (NW9) |
-| **OS** | FreeBSD 14.3 |
+| **OS** | Debian |
-| **Tailscale IP** | 100.122.219.41 |
+| **Tailscale IP** | 100.90.111.19 |
-| **Role** | Monitoring (Prometheus + Grafana) |
+| **Role** | VM host (Cockpit + KVM) |
 
 ## Hardware
 
@@ -19,43 +19,16 @@ Dedicated monitoring server. Runs Prometheus and Grafana, nothing else.
 | Memory | 32 GB |
 | Boot disk | 1 TB |
 
-Old gaming PC, now perfectly happy as a monitoring host. Very lightly loaded — disk at ~6%.
+Old gaming PC. Reinstalled with Debian in 2026-05 after moving monitoring to Grafana Cloud.
 
 ## Services
 
 | Service | Port | Status | Notes |
 |---------|------|--------|-------|
-| Prometheus | 9090 | Active | prometheus.pez.sh |
+| Cockpit | 9090 | Active | Web UI for VM management |
-| Grafana | 3000 | Active | grafana.pez.sh |
+| cockpit-machines | — | Active | KVM/libvirt VM management via Cockpit |
-| node_exporter | 9100 | Active | Metrics exporter |
 | Tailscale | — | Active | Mesh networking |
 
-Both Prometheus and Grafana are behind Authelia (auth handled by Caddy on helsinki-a).
-
-### Unused services (audit 2026-03-30)
-
-These services are enabled in rc.conf but appear unused. Pending cleanup.
-
-| Service | Port | Finding |
-|---------|------|---------|
-| InfluxDB | 8086 (all interfaces!) | Only `_internal` database — never used. Listening on `*:8086` is also a security concern. |
-| Redis | 6379 (localhost) | Empty keyspace, no clients. |
-| PostgreSQL | 5432 (localhost) | Has `pez_vps` database from a defunct VPS management project. Data may need backup before removal. |
-| libvirtd | — | Zero VMs. Installed for the same pez_vps project. |
-
-## ZFS
-
-- Pool: `zroot`
-- Weekly scrub: `0 12 * * sun zpool scrub zroot` (root crontab, not ansible-managed yet)
-
-## Why FreeBSD
-
-This one runs FreeBSD instead of Ubuntu. For a single-purpose monitoring host it works well. No particular reason to change it — it's stable and does its job.
-
 ## Networking
 
 Connected via Cat 5 to the Ubiquiti switch alongside london-b.
-
-## Notes
-
-Prometheus scrapes all hosts over Tailscale. See [monitoring.md](../monitoring.md) for scrape targets and dashboard details.

docs/hosts/nuremberg-a.md

@@ -7,8 +7,8 @@ Dedicated mail server. One job, does it well.
 | | |
 |---|---|
 | **Location** | Hetzner Cloud (Nuremberg) |
-| **OS** | Alpine Linux |
+| **OS** | Debian |
-| **Tailscale IP** | 100.117.235.28 |
+| **Tailscale IP** | 100.70.180.24 |
 | **Role** | Mail server (poste.io) |
 | **Provider** | Hetzner Cloud VPS |
 
@@ -32,3 +32,7 @@ Mail-related DNS records are managed via Cloudflare (Terraform):
 - **SPF** for sender verification
 - **DKIM** for message signing
 - **DMARC** for policy enforcement
+
+## Firewall
+
+Managed by Hetzner Cloud firewall rules (Terraform). Mail ports are exposed via Docker port mappings in `ansible/services/poste-io/docker-compose.yml`.

terraform/hetzner_dns.tf

@@ -12,9 +12,9 @@ locals {
 
 resource "hcloud_zone_rrset" "A_helsinki_a" {
   for_each = toset([
-    "@", "apps", "auth", "bitwarden", "download", "git", "grafana", "helsinki-a",
+    "@", "apps", "auth", "bitwarden", "download", "git", "helsinki-a",
-    "jellyfin", "jellyfin-requests", "ldap", "lidarr", "music", "naveen",
+    "jellyfin", "jellyfin-requests", "ldap", "lidarr", "london-a", "music", "naveen",
-    "plex", "prometheus", "prowlarr", "radarr", "readarr", "request",
+    "plex", "prowlarr", "radarr", "readarr", "request",
     "rss", "sonarr", "soulseek", "status",
   ])
   zone    = hcloud_zone.pezsh.name
@@ -38,11 +38,11 @@ resource "hcloud_zone_rrset" "nuremberg_mail" {
 
 resource "hcloud_zone_rrset" "A_copenhagen" {
   for_each = toset(["minecraft", "wow"])
-  zone    = hcloud_zone.pezsh.name
+  zone     = hcloud_zone.pezsh.name
-  name    = each.value
+  name     = each.value
-  type    = "A"
+  type     = "A"
-  ttl     = 300
+  ttl      = 300
-  records = [{ value = local.copenhagen }]
+  records  = [{ value = local.copenhagen }]
 }
 
 resource "hcloud_zone_rrset" "CNAME_public" {