Three load-bearing operational pieces before any new features: * .env.example committed, .env gitignored — per-developer API keys (THREATFOX_AUTH_KEY, OTX_API_KEY, NVD_API_KEY) ready for the registrations ahead; python-dotenv loads it in the venv CLI; compose picks it up via env_file: .env on the cockpit service. * Cockpit /api/inference-status endpoint + a topbar status chip that polls it on page load — "model · live" green when up, "model · offline" amber when the inference server is unreachable. No more manual checking. Compose also gains a healthcheck on the inference service (applies on next recreate). * New `psyc backup` command — tars the audit trail (db + sealed packages + recipient keys + ledger + datasets) to data/backups/psyc-data-<ts>.tar.gz. Excludes the HF model cache, mock-cert receipts, and the re-trainable adapters — the goal is the irrecoverable evidence, not bulk artifacts. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
79 lines
2.6 KiB
YAML
79 lines
2.6 KiB
YAML
# psyc — neuronetz.ai deployment stack.
|
|
#
|
|
# docker compose up -d --build # cockpit + mock-cert (no GPU)
|
|
# docker compose --profile gpu up -d --build # + the live model (needs an NVIDIA GPU)
|
|
#
|
|
# The cockpit is fronted by the external `backend` network's nginx-proxy as
|
|
# psyc.neuronetz.ai (point DNS for that name at the proxy host). mock-cert and
|
|
# the inference server stay internal — no VIRTUAL_HOST, reachable only inside
|
|
# `backend` by service name.
|
|
#
|
|
# WARNING: psyc has no built-in authentication. The reverse proxy / network
|
|
# perimeter is the security boundary. See docs/deploy.md.
|
|
|
|
services:
|
|
cockpit:
|
|
build: .
|
|
image: psyc:latest
|
|
command: ["psyc", "serve", "--host", "0.0.0.0", "--port", "8767"]
|
|
env_file: .env # per-dev API keys (gitignored). cp .env.example .env first.
|
|
environment:
|
|
VIRTUAL_HOST: psyc.neuronetz.ai
|
|
VIRTUAL_PORT: "8767"
|
|
PSYC_MOCK_CERT_URL: http://mock-cert:8770
|
|
PSYC_INFERENCE_URL: http://inference:8771
|
|
ports:
|
|
- "8767:8767" # direct/debug access; the proxy serves psyc.neuronetz.ai on :80
|
|
volumes:
|
|
- ./data:/data
|
|
networks: [backend]
|
|
restart: unless-stopped
|
|
healthcheck:
|
|
test: ["CMD", "python", "-c", "import urllib.request; urllib.request.urlopen('http://localhost:8767/healthz')"]
|
|
interval: 30s
|
|
timeout: 5s
|
|
retries: 3
|
|
|
|
mock-cert:
|
|
image: psyc:latest
|
|
command: ["psyc", "mock-cert", "--host", "0.0.0.0", "--port", "8770"]
|
|
volumes:
|
|
- ./data:/data
|
|
networks: [backend]
|
|
restart: unless-stopped
|
|
healthcheck:
|
|
test: ["CMD", "python", "-c", "import urllib.request; urllib.request.urlopen('http://localhost:8770/healthz')"]
|
|
interval: 30s
|
|
timeout: 5s
|
|
retries: 3
|
|
|
|
# The live fine-tuned model behind the Classifier bot. GPU-only — opt in with
|
|
# `--profile gpu`. Uses the psyc-trainer image (built from Dockerfile.train).
|
|
inference:
|
|
image: psyc-trainer
|
|
command: ["/scripts/serve_model.py", "--adapter", "/data/adapters/psyc-v4/final", "--host", "0.0.0.0", "--port", "8771"]
|
|
volumes:
|
|
- ./data:/data
|
|
- ./scripts:/scripts
|
|
networks: [backend]
|
|
restart: unless-stopped
|
|
profiles: ["gpu"]
|
|
healthcheck:
|
|
test: ["CMD", "python", "-c", "import urllib.request; urllib.request.urlopen('http://localhost:8771/healthz')"]
|
|
interval: 30s
|
|
timeout: 5s
|
|
retries: 3
|
|
start_period: 90s
|
|
deploy:
|
|
resources:
|
|
reservations:
|
|
devices:
|
|
- driver: nvidia
|
|
count: all
|
|
capabilities: [gpu]
|
|
|
|
networks:
|
|
backend:
|
|
name: backend
|
|
external: true
|