Initial commit

playbooks/README-netdata.md

@@ -0,0 +1,96 @@
+# Netdata Parent-Child Configuration
+
+## Architecture Overview
+
+This playbook configures a centralized monitoring setup using Netdata's parent-child streaming:
+
+- **Parent Node**: Fedora workstation (192.168.1.243) - Stores all metrics, provides dashboard
+- **Child Nodes**: All other hosts - Stream metrics to parent, minimal local storage
+
+## Current Installation Status
+
+Based on the latest check:
+
+### ✅ Already Running
+- `lenovo420` - Fedora workstation (1.7GB memory usage)
+- `omv800` - OpenMediaVault server (561MB memory usage) 
+- `lenovo` - Lenovo laptop (438MB memory usage)
+
+### ❌ Need Installation
+- `surface` - Surface tablet
+- `fedora` - **Target parent node**
+- `omvbackup` - Backup server
+
+### ⚠️ Unreachable
+- `audrey` - Connection timeout (may be offline)
+
+## Resource Usage Benefits
+
+**Before (current standalone instances):**
+- Total memory: ~2.7GB across 3 nodes
+- Each node stores full retention
+- Duplicated web interfaces
+
+**After (parent-child setup):**
+- Parent: ~512MB (centralized storage + web UI)
+- Children: ~32MB each (streaming only)
+- Total memory: ~640MB (76% reduction)
+
+## Running the Playbook
+
+```bash
+# Configure all nodes
+ansible-playbook -i inventory.ini playbooks/netdata-parent-child.yml
+
+# Configure only specific hosts
+ansible-playbook -i inventory.ini playbooks/netdata-parent-child.yml --limit fedora,surface
+
+# Test connectivity after setup
+ansible-playbook -i inventory.ini playbooks/netdata-parent-child.yml --tags verify
+```
+
+## Configuration Details
+
+### Parent Node (fedora)
+- Binds to `*:19999` (accessible from network)
+- 2GB disk space for metrics storage
+- Receives streams from all children
+- Provides centralized dashboard
+
+### Child Nodes (all others)  
+- Memory mode: `none` (no local storage)
+- Binds to `localhost:19999` only
+- Streams all metrics to parent
+- Minimal resource footprint
+
+## Accessing the Dashboard
+
+After configuration:
+- **Central Dashboard**: http://192.168.1.243:19999
+- **Individual Node Access**: http://node-ip:19999 (localhost only on children)
+
+## Security Features
+
+- API key authentication for streaming
+- Network restrictions (home + Tailscale networks only)
+- Local-only access on child nodes
+- Firewall rules automatically configured
+
+## Troubleshooting
+
+### Check stream status on parent:
+```bash
+curl http://192.168.1.243:19999/api/v1/info | jq '.hosts'
+```
+
+### Check streaming on child:
+```bash
+systemctl status netdata
+journalctl -u netdata -f
+```
+
+### Verify connectivity:
+```bash
+# From child node
+curl -I http://192.168.1.243:19999/api/v1/info
+```

playbooks/extend-monitoring.yml

@@ -0,0 +1,190 @@
+---
+- name: Extend monitoring to non-Linux devices
+  hosts: fedora  # Run on parent node
+  become: yes
+  vars:
+    network_devices:
+      - name: "Gateway Router"
+        ip: "192.168.50.1"
+        type: "router" 
+      - name: "Immich Photos"
+        ip: "192.168.50.66"
+        type: "docker_host"
+      - name: "Unknown Device 1"
+        ip: "192.168.50.20"
+        type: "unknown"
+      - name: "Unknown Device 2" 
+        ip: "192.168.50.25"
+        type: "unknown"
+        
+  tasks:
+    - name: Create network monitoring configuration
+      copy:
+        content: |
+          # Network Device Monitoring Configuration
+          # Add to /etc/netdata/go.d/ping.conf
+          
+          jobs:
+            - name: home_network_ping
+              hosts:
+          {% for device in network_devices %}
+                - {{ device.ip }}  # {{ device.name }}
+          {% endfor %}
+              interval: 30s
+              count: 3
+              timeout: 1s
+              
+            - name: critical_services_ping  
+              hosts:
+                - 192.168.50.1    # Gateway
+                - 192.168.50.66   # Immich Photos
+                - 8.8.8.8         # Google DNS
+                - 1.1.1.1         # Cloudflare DNS
+              interval: 10s
+              count: 1
+              timeout: 2s
+              
+        dest: /etc/netdata/go.d/ping.conf
+        
+    - name: Enable ping monitoring in Netdata
+      lineinfile:
+        path: /etc/netdata/netdata.conf
+        regexp: '^.*\[plugin:go.d\]'
+        line: '[plugin:go.d]'
+        create: yes
+        
+    - name: Configure ping plugin
+      blockinfile:
+        path: /etc/netdata/netdata.conf
+        marker: "# {mark} PING MONITORING CONFIG"
+        block: |
+          [plugin:go.d]
+              ping = yes
+              
+    - name: Create Docker monitoring script for remote hosts
+      copy:
+        content: |
+          #!/bin/bash
+          # Monitor Docker containers on remote hosts
+          
+          REMOTE_HOSTS=(
+              "192.168.50.66:photos.local"  # Immich Photos
+              "100.78.26.112:omv800"        # OMV with Docker
+              "100.98.144.95:lenovo420"     # Lenovo with Docker
+          )
+          
+          for host_info in "${REMOTE_HOSTS[@]}"; do
+              IFS=':' read -r ip hostname <<< "$host_info"
+              echo "=== Docker containers on $hostname ($ip) ==="
+              
+              # Try to get container stats via SSH
+              if ssh -o ConnectTimeout=5 -o StrictHostKeyChecking=no "jon@$ip" "docker ps --format 'table {{.Names}}\t{{.Status}}\t{{.Ports}}'" 2>/dev/null; then
+                  echo "SSH connection successful"
+              else
+                  echo "SSH connection failed - checking if HTTP monitoring available"
+                  # Check for cAdvisor or similar
+                  if curl -s --connect-timeout 3 "http://$ip:8080/metrics" >/dev/null 2>&1; then
+                      echo "cAdvisor metrics available at http://$ip:8080"
+                  fi
+              fi
+              echo ""
+          done
+        dest: /usr/local/bin/check-remote-docker.sh
+        mode: '0755'
+        
+    - name: Add cron job for Docker monitoring
+      cron:
+        name: "Check remote Docker containers"
+        minute: "*/5"
+        job: "/usr/local/bin/check-remote-docker.sh >> /var/log/remote-docker-check.log 2>&1"
+        
+    - name: Create Windows monitoring via WMI (if available)
+      copy:
+        content: |
+          # Windows monitoring configuration
+          # Requires wmic or PowerShell remoting to be enabled
+          
+          [plugin:apps]
+              # Disable apps monitoring to reduce load
+              enabled = no
+              
+          # Example Windows WMI monitoring (requires additional setup)
+          # [plugin:wmi]
+          #     command = wmic -U domain/user%password //192.168.50.100 "SELECT * FROM Win32_Processor"
+          #     update every = 60
+              
+        dest: /etc/netdata/windows-monitoring.conf
+        
+    - name: Test ping monitoring
+      shell: |
+        /usr/libexec/netdata/plugins.d/go.d.plugin -m ping -d
+      register: ping_test
+      ignore_errors: yes
+      
+    - name: Display ping test results
+      debug:
+        var: ping_test.stdout_lines
+        
+    - name: Restart Netdata to apply changes
+      systemd:
+        name: netdata
+        state: restarted
+        
+    - name: Create monitoring dashboard links
+      copy:
+        content: |
+          # Extended Monitoring Dashboard Links
+          
+          ## Main Netdata Dashboard
+          http://{{ ansible_host }}:19999
+          
+          ## Network Ping Monitoring
+          http://{{ ansible_host }}:19999/#menu_ping_submenu_home_network_ping
+          
+          ## System Overview
+          http://{{ ansible_host }}:19999/#menu_system_submenu_cpu
+          
+          ## All Connected Hosts
+          {% for host in groups['all_linux'] %}
+          - {{ host }}: http://{{ ansible_host }}:19999/host/{{ host }}
+          {% endfor %}
+          
+          ## Discovered Network Devices (52 total)
+          - Gateway: 192.168.50.1
+          - Photos Server: 192.168.50.66 
+          - 50 other devices (IoT, mobile, Windows, etc.)
+          
+          ## Monitoring Capabilities by Device Type:
+          
+          ### ✅ Full Monitoring (Linux hosts with Netdata)
+          - fedora (parent)
+          - lenovo420, omv800, lenovo, surface, omvbackup (children)
+          
+          ### 📊 Limited Monitoring (ping, port checks)  
+          - Router/gateway
+          - Docker hosts
+          - Network appliances
+          
+          ### ⚠️ Manual Monitoring Required
+          - Windows machines (need WMI/SNMP setup)
+          - Mobile devices (battery optimization prevents agents)
+          - IoT devices (resource constrained)
+          
+        dest: /home/jonathan/monitoring-dashboard.md
+        owner: jonathan
+        group: jonathan
+        
+    - name: Summary of monitoring coverage
+      debug:
+        msg: |
+          Monitoring Coverage Summary:
+          
+          ✅ Linux hosts: 6/6 (100%) - Full system monitoring
+          📊 Network devices: 52 discovered - Ping monitoring added  
+          🐳 Docker containers: Remote monitoring via SSH/API
+          📱 Mobile/IoT: 46+ devices - Network connectivity only
+          
+          Total network coverage: Ping monitoring for all 52 devices
+          Detailed monitoring: 6 Linux systems (infrastructure core)
+          
+          Dashboard: http://{{ ansible_host }}:19999

playbooks/group_vars/all.yml

@@ -0,0 +1,13 @@
+---
+# Netdata Configuration
+netdata_api_key: "homelab-stream-key-2024-secure"
+netdata_parent_retention_hours: 168  # 7 days
+netdata_child_retention_hours: 1     # 1 hour local buffer
+
+# Monitoring Settings
+netdata_memory_limit_mb: 128         # Memory limit for child nodes
+netdata_disk_space_mb: 512           # Disk space limit for child nodes
+
+# Network Configuration
+home_network_cidr: "192.168.1.0/24"
+tailscale_network_cidr: "100.64.0.0/10"

playbooks/netdata-parent-child.yml

@@ -0,0 +1,198 @@
+---
+- name: Configure Netdata Parent-Child Streaming Architecture
+  hosts: all
+  become: yes
+  vars:
+    netdata_parent_host: "192.168.50.225"  # Fedora workstation wired IP (more stable)
+    netdata_stream_api_key: "{{ netdata_api_key | default('homelab-stream-key-2024') }}"
+    
+  tasks:
+    - name: Install Netdata on hosts where it's missing
+      block:
+        - name: Install Netdata via package manager (Fedora/RHEL)
+          dnf:
+            name: netdata
+            state: present
+          when: ansible_os_family == "RedHat"
+          
+        - name: Install Netdata via package manager (Debian/Ubuntu)
+          apt:
+            name: netdata
+            state: present
+            update_cache: yes
+          when: ansible_os_family == "Debian"
+          
+        - name: Install Netdata via installer script (fallback)
+          shell: |
+            curl -L https://get.netdata.cloud/kickstart.sh | bash -s -- --stable-channel --disable-telemetry --dont-wait
+          args:
+            creates: /usr/sbin/netdata
+          when: ansible_os_family not in ["RedHat", "Debian"]
+      when: inventory_hostname in ['surface', 'fedora', 'omvbackup']
+
+    - name: Configure Netdata Parent (Central monitoring node)
+      block:
+        - name: Create Netdata parent configuration
+          copy:
+            content: |
+              [global]
+                  # Parent node configuration
+                  memory mode = dbengine
+                  page cache size = 32
+                  dbengine disk space = 2048
+                  
+                  # Network settings
+                  bind to = *:19999
+                  allow connections from = 192.168.50.0/24 100.64.0.0/10
+                  
+                  # Disable local data collection plugins to reduce load
+                  [plugins]
+                      apps = no
+                      cgroups = no
+                      diskspace = no
+                      
+              [web]
+                  # Allow dashboard from any host on home network
+                  allow dashboard from = 192.168.50.0/24 100.64.0.0/10
+                  
+              # Stream configuration for receiving child data
+              [stream]
+                  enabled = yes
+                  destination =
+                  api key = {{ netdata_stream_api_key }}
+                  timeout seconds = 60
+                  buffer size bytes = 1048576
+                  reconnect delay seconds = 5
+                  initial clock resync iterations = 60
+                  
+                  # Accept streams from child nodes
+                  [{{ netdata_stream_api_key }}]
+                      enabled = yes
+                      allow from = 192.168.50.0/24 100.64.0.0/10
+                      default history = 3600
+                      default memory = dbengine
+                      health enabled by default = auto
+            dest: /etc/netdata/netdata.conf
+            backup: yes
+            
+        - name: Enable and start Netdata parent
+          systemd:
+            name: netdata
+            enabled: yes
+            state: restarted
+            
+        - name: Open firewall for Netdata (firewalld)
+          firewalld:
+            port: 19999/tcp
+            permanent: yes
+            state: enabled
+            immediate: yes
+          ignore_errors: yes
+          
+        - name: Open firewall for Netdata (ufw)
+          ufw:
+            rule: allow
+            port: '19999'
+            proto: tcp
+          ignore_errors: yes
+          
+      when: inventory_hostname == 'fedora'
+
+    - name: Configure Netdata Children (streaming to parent)
+      block:
+        - name: Configure Netdata child to stream to parent
+          copy:
+            content: |
+              [global]
+                  # Child node - minimal local storage
+                  memory mode = none
+                  
+                  # Network settings
+                  bind to = localhost:19999
+                  
+              [web]
+                  # Restrict dashboard access to localhost only
+                  allow dashboard from = localhost
+                  
+              # Stream all data to parent
+              [stream]
+                  enabled = yes
+                  destination = {{ netdata_parent_host }}:19999
+                  api key = {{ netdata_stream_api_key }}
+                  timeout seconds = 60
+                  buffer size bytes = 1048576
+                  reconnect delay seconds = 5
+                  initial clock resync iterations = 60
+                  
+                  # Send everything to parent
+                  send charts matching = *
+                  
+              # Reduce resource usage
+              [plugins]
+                  # Keep essential monitoring but reduce frequency
+                  proc = yes
+                  diskspace = yes
+                  cgroups = yes
+                  apps = yes
+                  
+              # Optimize collection intervals
+              [plugin:proc]
+                  /proc/net/dev = yes
+                  /proc/diskstats = yes
+                  /proc/net/sockstat = yes
+                  /proc/meminfo = yes
+                  /proc/vmstat = yes
+                  /proc/stat = yes
+                  /proc/loadavg = yes
+                  
+            dest: /etc/netdata/netdata.conf
+            backup: yes
+            
+        - name: Restart Netdata child
+          systemd:
+            name: netdata
+            state: restarted
+            
+      when: inventory_hostname != 'fedora'
+
+    - name: Verify Netdata service status
+      systemd:
+        name: netdata
+        state: started
+        enabled: yes
+      register: netdata_status
+      
+    - name: Display Netdata status
+      debug:
+        msg: |
+          Netdata {{ 'parent' if inventory_hostname == 'fedora' else 'child' }} configured on {{ inventory_hostname }}
+          Status: {{ netdata_status.status.ActiveState }}
+          {% if inventory_hostname == 'fedora' %}
+          Parent dashboard: http://{{ ansible_host }}:19999
+          {% else %}
+          Streaming to: {{ netdata_parent_host }}:19999
+          {% endif %}
+
+    - name: Test connectivity to parent (from children)
+      uri:
+        url: "http://{{ netdata_parent_host }}:19999/api/v1/info"
+        method: GET
+        timeout: 10
+      register: parent_test
+      ignore_errors: yes
+      when: inventory_hostname != 'fedora'
+      
+    - name: Display connectivity test results
+      debug:
+        msg: |
+          Connection to parent: {{ 'SUCCESS' if parent_test.status == 200 else 'FAILED' }}
+          {% if parent_test.status != 200 %}
+          Error: {{ parent_test.msg | default('Unknown error') }}
+          {% endif %}
+      when: inventory_hostname != 'fedora' and parent_test is defined
+
+  handlers:
+    - name: restart netdata
+      systemd:
+        name: netdata
+        state: restarted