Repository navigation
Expand file tree
/
Copy pathrender.yaml
More file actions
45 lines (45 loc) · 2.12 KB
/
Copy pathrender.yaml
File metadata and controls
45 lines (45 loc) · 2.12 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
# ---------------------------------------------------------------------------
# MINEGUARD AI — Render Blueprint (infrastructure as code)
#
# Apply with the CLI (no dashboard clicking):
# export PATH="$PATH:$HOME/.local/bin"
# render login # browser-authorised CLI token
# render blueprints validate ./render.yaml
# render services create ./render.yaml # or: New Blueprint in the dashboard
#
# One service does everything: the Dockerfile compiles the SPA and FastAPI
# serves both the UI and /api from the same origin, so there is nothing else to
# configure (no CORS, no second URL, no client-side API base).
#
# NOTE on the Free plan (deliberate, and worth knowing before a demo):
# * idle services spin down after 15 min; the next request waits ~1 min
# * the filesystem is EPHEMERAL — data/store.json and data/uploads/ are reset
# on every restart / redeploy / spin-down. api/store.py reseeds the same
# deterministic baseline each boot, which is exactly the documented
# `POST /api/admin/reset` behaviour the deck demonstrates.
# * persistent disks are paid-only; to keep state across restarts, switch
# `plan` to a paid tier and add a `disk` (see docs/DEPLOY.md §4)
# ---------------------------------------------------------------------------
services:
- type: web
name: mineguard-ai
runtime: docker
plan: free
# Oregon is where the Free plan is guaranteed; singapore/frankfurt are
# lower-latency for India but may require a paid plan in your workspace.
region: oregon
branch: main
repo: https://github.com/dev-dipeshkumar/-MINEGUARD-AI
dockerfilePath: ./Dockerfile
healthCheckPath: /api/health
autoDeploy: true
numInstances: 1
envVars:
# Rolling window the engine scores against; keeps the demo anchored to the
# current date (the seeded corpus is generated relative to today).
- key: MINEGUARD_HISTORY_DAYS
value: "90"
# rule-based today; `ml` is only accepted once an MLModelRiskScoringStrategy
# implementation is registered (api/services/risk_engine.py).
- key: MINEGUARD_ENGINE
value: rule-based