infra: full IaC state after hf-pdocker-01 rebuild (2026-03-08)
Replaces initial Nomad/Consul/Terraform skeleton with current Docker-based infrastructure across all three nodes (hf-pdocker-01, hf-pdocker-02, bart). Key changes: - CLAUDE.md: updated service distribution reflecting hf-pdocker-01 restored - docs/plans/2026-03-08-hf-pdocker-01-rebuild.md: new rebuild plan doc - compose/bart/: ai-agents, guacamole, matrix, media, openvas, smarthome, watchtower - compose/hf-pdocker-02/: traefik, monitoring, adguard, gitea, semaphore, wikijs, unifi, vaultwarden, actual-budget, mcpo, netclaw, trivy, watchtower, n8n - compose/hf-pdocker-01/: frigate, frigate-notify, farm-agent - stacks/: all Swarm stack definitions (monitoring, traefik, smarthome, matrix, guacamole, media, ai-agents, frigate, gitea, semaphore, wikijs, unifi, etc.) - .gitignore: exclude .env credential files and acme.json certs - .env.example: placeholder credential files for all services - .claude/agents/: gitea-backup and other infrastructure agent configs - scripts/: frigate media cleanup and NAS sync scripts Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
@@ -0,0 +1,163 @@
|
||||
global:
|
||||
scrape_interval: 1m
|
||||
evaluation_interval: 1m
|
||||
|
||||
alerting:
|
||||
alertmanagers:
|
||||
- static_configs:
|
||||
- targets:
|
||||
- 'alertmanager:9093'
|
||||
|
||||
rule_files:
|
||||
- "alerts.yml"
|
||||
|
||||
scrape_configs:
|
||||
# Prometheus self-monitoring
|
||||
- job_name: 'prometheus'
|
||||
scrape_interval: 1m
|
||||
static_configs:
|
||||
- targets: ['localhost:9090']
|
||||
|
||||
# Node Exporter (system metrics) - all 3 swarm nodes
|
||||
- job_name: 'node'
|
||||
static_configs:
|
||||
- targets: ['192.168.86.192:9100']
|
||||
labels:
|
||||
hostname: 'hf-pdocker-01'
|
||||
host_ip: '192.168.86.192'
|
||||
- targets: ['192.168.86.167:9100']
|
||||
labels:
|
||||
hostname: 'bart'
|
||||
host_ip: '192.168.86.167'
|
||||
- targets: ['192.168.86.100:9100']
|
||||
labels:
|
||||
hostname: 'hf-pdocker-02'
|
||||
host_ip: '192.168.86.100'
|
||||
|
||||
# Blackbox Exporter self-monitoring
|
||||
- job_name: 'blackbox'
|
||||
static_configs:
|
||||
- targets: ['blackbox-exporter:9115']
|
||||
|
||||
# Matrix Alertmanager Receiver metrics (running on hf-pdocker-02 during failover)
|
||||
- job_name: 'matrix-alertmanager-receiver'
|
||||
scrape_interval: 1m
|
||||
static_configs:
|
||||
- targets: ['192.168.86.100:9094']
|
||||
labels:
|
||||
hostname: 'hf-pdocker-02'
|
||||
|
||||
# HTTP Health Checks - hf-pdocker-01 services (native services only; failover services tracked in other jobs)
|
||||
- job_name: 'http-pdocker01'
|
||||
metrics_path: /probe
|
||||
params:
|
||||
module: [http_2xx]
|
||||
static_configs:
|
||||
- targets:
|
||||
- http://192.168.86.192:5000 # Frigate NVR (standalone, host network)
|
||||
- http://192.168.86.192:5199/health # Frigate Notify
|
||||
- http://192.168.86.192:8888/health # Farm Manager API
|
||||
- http://192.168.86.192:8889/health # Farm Agent
|
||||
relabel_configs:
|
||||
- source_labels: [__address__]
|
||||
target_label: __param_target
|
||||
- source_labels: [__param_target]
|
||||
target_label: instance
|
||||
- target_label: __address__
|
||||
replacement: blackbox-exporter:9115
|
||||
|
||||
# HTTP Health Checks - hf-pdocker-02 services (native + failover from hf-pdocker-01)
|
||||
- job_name: 'http-pdocker02'
|
||||
metrics_path: /probe
|
||||
params:
|
||||
module: [http_2xx]
|
||||
static_configs:
|
||||
- targets:
|
||||
- http://192.168.86.100:3003 # Gitea
|
||||
- http://192.168.86.100:3004 # Semaphore
|
||||
- http://192.168.86.100:3002 # Wiki.js
|
||||
- http://192.168.86.100:26226 # Vaultwarden
|
||||
- http://192.168.86.100:5006 # Actual Budget
|
||||
- http://192.168.86.100:8001/openapi.json # MCPO
|
||||
- http://192.168.86.100:8889/health # Farm Agent
|
||||
- http://192.168.86.100:3001 # Grafana (FAILOVER from hf-pdocker-01)
|
||||
- http://192.168.86.100:9095 # Prometheus (FAILOVER from hf-pdocker-01)
|
||||
- http://192.168.86.100:9093 # Alertmanager (FAILOVER from hf-pdocker-01)
|
||||
- http://192.168.86.100:3000 # AdGuard Home (FAILOVER from hf-pdocker-01)
|
||||
- http://192.168.86.100:9000 # Portainer
|
||||
relabel_configs:
|
||||
- source_labels: [__address__]
|
||||
target_label: __param_target
|
||||
- source_labels: [__param_target]
|
||||
target_label: instance
|
||||
- target_label: __address__
|
||||
replacement: blackbox-exporter:9115
|
||||
|
||||
# HTTPS Health Checks - services with self-signed certs
|
||||
- job_name: 'http-pdocker02-https'
|
||||
metrics_path: /probe
|
||||
params:
|
||||
module: [http_2xx_insecure]
|
||||
static_configs:
|
||||
- targets:
|
||||
- https://192.168.86.100:8443 # Unifi Controller (self-signed cert)
|
||||
relabel_configs:
|
||||
- source_labels: [__address__]
|
||||
target_label: __param_target
|
||||
- source_labels: [__param_target]
|
||||
target_label: instance
|
||||
- target_label: __address__
|
||||
replacement: blackbox-exporter:9115
|
||||
|
||||
# HTTP Health Checks - bart services (native + failover from hf-pdocker-01)
|
||||
- job_name: 'http-bart'
|
||||
metrics_path: /probe
|
||||
params:
|
||||
module: [http_2xx]
|
||||
static_configs:
|
||||
- targets:
|
||||
- http://192.168.86.167:8096 # Jellyfin
|
||||
- http://192.168.86.167:8989 # Sonarr
|
||||
- http://192.168.86.167:7878 # Radarr
|
||||
- http://192.168.86.167:8082 # SABnzbd
|
||||
- http://192.168.86.167:5076 # NZBHydra2
|
||||
- http://192.168.86.167:5055 # Jellyseerr
|
||||
- http://192.168.86.167:9392 # OpenVAS
|
||||
- http://192.168.86.167:8889/health # Farm Agent
|
||||
- http://192.168.86.167:8123 # Home Assistant (FAILOVER from hf-pdocker-01)
|
||||
- http://192.168.86.167:6052 # ESPHome (FAILOVER from hf-pdocker-01)
|
||||
- http://192.168.86.167:8199/guacamole/ # Guacamole (FAILOVER from hf-pdocker-01)
|
||||
- http://192.168.86.167:8008 # Matrix Synapse (FAILOVER from hf-pdocker-01)
|
||||
relabel_configs:
|
||||
- source_labels: [__address__]
|
||||
target_label: __param_target
|
||||
- source_labels: [__param_target]
|
||||
target_label: instance
|
||||
- target_label: __address__
|
||||
replacement: blackbox-exporter:9115
|
||||
|
||||
# HTTPS Health Checks - External domains via Traefik
|
||||
- job_name: 'http-external'
|
||||
metrics_path: /probe
|
||||
params:
|
||||
module: [http_2xx]
|
||||
static_configs:
|
||||
- targets:
|
||||
- https://mess.hobbs.farm # Matrix (via Traefik)
|
||||
- https://solo.hobbs.farm/guacamole/ # Guacamole (via Traefik)
|
||||
- https://yoda.hobbs.farm # Home Assistant
|
||||
- https://vault.hobbs.farm # Vaultwarden
|
||||
- https://git.hobbs.farm # Gitea
|
||||
relabel_configs:
|
||||
- source_labels: [__address__]
|
||||
target_label: __param_target
|
||||
- source_labels: [__param_target]
|
||||
target_label: instance
|
||||
- target_label: __address__
|
||||
replacement: blackbox-exporter:9115
|
||||
|
||||
# Unifi Network Devices - uncomment when unifi-exporter is deployed
|
||||
# - job_name: 'unifi-devices'
|
||||
# scrape_interval: 30s
|
||||
# static_configs:
|
||||
# - targets: ['unifi-exporter:9101']
|
||||
Reference in New Issue
Block a user