Erich Blume
31d925814f
## Summary - Deploy Ollama as a new ArgoCD-managed service on ringtail's k3s cluster with GPU acceleration - Declarative model management via `models.txt` + sidecar sync script (mirrors kiwix torrent pattern) - Initial models: `qwen2.5:14b`, `deepseek-r1:14b`, `phi4:14b`, `gemma3:12b` - hostPath PV on `/mnt/storage1/ollama` for fast local model storage (200Gi) - Tailscale ingress at `ollama.ops.eblu.me` for API access from tailnet - Enable GPU time-slicing (`replicas: 2`) on nvidia-device-plugin so Frigate and Ollama share the RTX 4080 ## Deployment and Testing - [ ] Deploy nvidia-device-plugin changes first: `argocd app sync nvidia-device-plugin` - [ ] Verify GPU time-slicing: `kubectl describe node ringtail --context=k3s-ringtail` shows `nvidia.com/gpu: 2` - [ ] Sync `apps` app with `--revision feature/ollama-ringtail` - [ ] Set ollama app to branch: `argocd app set ollama --revision feature/ollama-ringtail && argocd app sync ollama` - [ ] Verify model-sync sidecar pulls models: `kubectl logs -n ollama deploy/ollama -c model-sync --context=k3s-ringtail` - [ ] Test API: `curl https://ollama.ops.eblu.me/api/tags` - [ ] Test inference: `curl https://ollama.ops.eblu.me/api/generate -d '{"model":"qwen2.5:14b","prompt":"Hello"}'` - [ ] Verify Frigate still works after GPU sharing change - [ ] After merge: `argocd app set ollama --revision main && argocd app sync ollama` Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/277
105 lines
3.5 KiB
YAML
105 lines
3.5 KiB
YAML
---
|
|
# Caddy reverse proxy configuration
|
|
# Caddy is built manually from ~/code/3rd/caddy with the Gandi DNS plugin
|
|
|
|
caddy_repo_dir: /Users/erichblume/code/3rd/caddy
|
|
caddy_binary: "{{ caddy_repo_dir }}/bin/caddy"
|
|
caddy_config_dir: /Users/erichblume/.config/caddy
|
|
caddy_data_dir: /Users/erichblume/.local/share/caddy
|
|
caddy_log_dir: /Users/erichblume/Library/Logs
|
|
|
|
# Gandi API token file (written by ansible, chmod 0600)
|
|
# Caddy reads this file for ACME DNS-01 challenges
|
|
caddy_gandi_token_file: /Users/erichblume/.config/caddy/gandi-token
|
|
|
|
# Domain configuration
|
|
caddy_domain: ops.eblu.me
|
|
|
|
# HTTPS port (443 is standard)
|
|
caddy_https_port: 443
|
|
|
|
# Services to proxy
|
|
# Format: { name: "service", host: "hostname", backend: "url" }
|
|
caddy_services:
|
|
# Indri-local services
|
|
- name: forge
|
|
host: "forge.{{ caddy_domain }}"
|
|
backend: "http://localhost:3001"
|
|
- name: registry
|
|
host: "registry.{{ caddy_domain }}"
|
|
backend: "http://localhost:5050"
|
|
- name: jellyfin
|
|
host: "jellyfin.{{ caddy_domain }}"
|
|
backend: "http://localhost:8096"
|
|
|
|
# K8s services (via Tailscale Ingress)
|
|
# Caddy proxies to existing Tailscale endpoints - traffic stays local
|
|
- name: grafana
|
|
host: "grafana.{{ caddy_domain }}"
|
|
backend: "https://grafana.tail8d86e.ts.net"
|
|
- name: argocd
|
|
host: "argocd.{{ caddy_domain }}"
|
|
backend: "https://argocd.tail8d86e.ts.net"
|
|
- name: prometheus
|
|
host: "prometheus.{{ caddy_domain }}"
|
|
backend: "https://prometheus.tail8d86e.ts.net"
|
|
- name: loki
|
|
host: "loki.{{ caddy_domain }}"
|
|
backend: "https://loki.tail8d86e.ts.net"
|
|
- name: miniflux
|
|
host: "feed.{{ caddy_domain }}"
|
|
backend: "https://feed.tail8d86e.ts.net"
|
|
- name: devpi
|
|
host: "pypi.{{ caddy_domain }}"
|
|
backend: "https://pypi.tail8d86e.ts.net"
|
|
- name: kiwix
|
|
host: "kiwix.{{ caddy_domain }}"
|
|
backend: "https://kiwix.tail8d86e.ts.net"
|
|
- name: torrent
|
|
host: "torrent.{{ caddy_domain }}"
|
|
backend: "https://torrent.tail8d86e.ts.net"
|
|
- name: teslamate
|
|
host: "tesla.{{ caddy_domain }}"
|
|
backend: "https://tesla.tail8d86e.ts.net"
|
|
- name: immich
|
|
host: "photos.{{ caddy_domain }}"
|
|
backend: "https://photos.tail8d86e.ts.net"
|
|
- name: navidrome
|
|
host: "dj.{{ caddy_domain }}"
|
|
backend: "https://dj.tail8d86e.ts.net"
|
|
- name: homepage
|
|
host: "go.{{ caddy_domain }}"
|
|
backend: "https://go.tail8d86e.ts.net"
|
|
- name: docs
|
|
host: "docs.{{ caddy_domain }}"
|
|
backend: "https://docs.tail8d86e.ts.net"
|
|
- name: cv
|
|
host: "cv.{{ caddy_domain }}"
|
|
backend: "https://cv.tail8d86e.ts.net"
|
|
- name: nvr
|
|
host: "nvr.{{ caddy_domain }}"
|
|
backend: "https://nvr.tail8d86e.ts.net"
|
|
- name: authentik
|
|
host: "authentik.{{ caddy_domain }}"
|
|
backend: "https://authentik.tail8d86e.ts.net"
|
|
- name: ntfy
|
|
host: "ntfy.{{ caddy_domain }}"
|
|
backend: "https://ntfy.tail8d86e.ts.net"
|
|
- name: ollama
|
|
host: "ollama.{{ caddy_domain }}"
|
|
backend: "https://ollama.tail8d86e.ts.net"
|
|
- name: sifaka
|
|
host: "nas.{{ caddy_domain }}"
|
|
backend: "http://sifaka:5000"
|
|
|
|
# Layer 4 (TCP) services
|
|
# Format: { port: external_port, backend: "host:port" }
|
|
caddy_tcp_services:
|
|
- port: 2222
|
|
backend: "localhost:2200" # Forgejo SSH
|
|
- port: 5432
|
|
backend: "pg.tail8d86e.ts.net:5432" # PostgreSQL
|
|
- port: "{{ sifaka_node_exporter_port }}"
|
|
backend: "sifaka:{{ sifaka_node_exporter_port }}" # Sifaka node_exporter
|
|
- port: "{{ sifaka_smartctl_exporter_port }}"
|
|
backend: "sifaka:{{ sifaka_smartctl_exporter_port }}" # Sifaka smartctl_exporter
|