2 Commits

Author SHA1 Message Date
ztaylor e78795bf4f docs: update deploy design for public-ingress pivot + publicOnlyTransport finding
Records that the in-cluster Service DNS could not be used for a dashboard-registered
worker (engine publicOnlyTransport SSRF guard blocks internal addresses), the pivot to
st-app chart + public ingress at arcade-eval-ref.st.dev (CNAME -> k8s-backstage.st.dev),
and the verified end-to-end whoami result.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-22 12:44:55 -04:00
Zeb Taylor 9009237a14 deploy: containerize arcade-eval-ref MCP server + ACR build/push action (#4)
Replace the cloudflared quick-tunnel dev pattern with a permanent in-cluster
deployment so the self-hosted Arcade engine reaches the echo/add/whoami reference
server over stable cluster DNS.

- lib/mcp_server/Dockerfile: python:3.12-slim, pip install ., HTTP transport via
  ARCADE_SERVER_{TRANSPORT,HOST,PORT} env overrides (no server.py change needed),
  non-root user, port 8000.
- .github/workflows/build-push-acr.yml: build + push
  servicetitandev.azurecr.io/arcade-eval-ref:1.0.<run_number>. Adapted from
  servicetitan/mem0; needs repo secrets ACR_DEV_USERNAME / ACR_DEV_PASSWORD.
- docs/superpowers/specs design record.

K8s manifests live in k8s-backstage-v2 apps/mcp/arcade-eval-ref/ (separate branch).

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-22 11:29:23 -04:00
3 changed files with 188 additions and 0 deletions
+58
View File
@@ -0,0 +1,58 @@
name: Build and Push to ACR
# Builds the arcade-eval reference MCP server image and pushes it to the
# ServiceTitan dev Azure Container Registry. The image is consumed by
# apps/mcp/arcade-eval-ref/ in k8s-backstage-v2 (backstage-wus2-v4).
#
# Adapted from servicetitan/mem0 .github/workflows/build-push-acr.yml.
# Requires repo secrets ACR_DEV_USERNAME and ACR_DEV_PASSWORD.
on:
workflow_dispatch:
push:
branches:
- main
paths:
- 'lib/mcp_server/**'
- '.github/workflows/build-push-acr.yml'
env:
REGISTRY: servicetitandev.azurecr.io
IMAGE: arcade-eval-ref
VERSION_PREFIX: "1.0"
jobs:
build-and-push:
runs-on: ubuntu-latest
permissions:
contents: read
steps:
- name: Checkout
uses: actions/checkout@v4
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v3
- name: Login to ACR
uses: docker/login-action@v3
with:
registry: ${{ env.REGISTRY }}
username: ${{ secrets.ACR_DEV_USERNAME }}
password: ${{ secrets.ACR_DEV_PASSWORD }}
- name: Generate image tag
id: meta
run: |
echo "tag=${{ env.VERSION_PREFIX }}.${{ github.run_number }}" >> "$GITHUB_OUTPUT"
- name: Build and push
uses: docker/build-push-action@v6
with:
context: lib/mcp_server
file: lib/mcp_server/Dockerfile
push: true
tags: |
${{ env.REGISTRY }}/${{ env.IMAGE }}:${{ steps.meta.outputs.tag }}
cache-from: type=gha,scope=${{ env.IMAGE }}
cache-to: type=gha,mode=max,scope=${{ env.IMAGE }}
@@ -0,0 +1,95 @@
# Deploy arcade-eval reference MCP server to backstage k8s
**Date:** 2026-06-22
**Status:** DONE — deployed and verified end-to-end.
## Goal
Replace the ephemeral cloudflared **quick tunnel** (used to register the
`arcade-eval-ref` server with the self-hosted Arcade engine) with a permanent
deployment on `backstage-wus2-v4`, so the engine reaches the server over a stable
URL instead of a `trycloudflare.com` URL that dies on restart.
Relevant eval categories: cat-4 (custom server dev), cat-8 (deployment), cat-9 (DX).
## Key finding that shaped the final design
The first attempt registered the in-cluster **Service DNS**
(`http://arcade-eval-ref.arcade-eval-ref.svc.cluster.local:8000`) as a dashboard
worker. Health went green but **0 tools loaded**. Engine logs showed:
```
Failed to get worker tools: Get ".../worker/tools":
dial tcp 10.0.192.27:8000: publicOnlyTransport: blocked connection to internal address
```
**The Arcade engine has an SSRF guard (`publicOnlyTransport`) that blocks
dashboard-registered worker URIs resolving to internal/private (RFC1918) addresses.**
Only workers declared in the **engine config file** (e.g. the bundled `arcade-worker-main`
at `http://arcade-worker-main:8001`) may use internal URIs. Health checks aren't guarded
(hence green), but the authenticated `/worker/tools` discovery is. The cloudflared tunnel
worked only because it was a *public* URL.
⇒ A dashboard-registered in-cluster worker **must be exposed on a public URL**. (The
worker secret was a red herring — the connection is refused before auth.)
## Architecture / data flow (final)
```
Claude Code ──▶ gateway zeb-gateway-test ──▶ Arcade engine ──HTTPS /worker/*──▶
https://arcade-eval-ref.st.dev (Cloudflare CNAME → k8s-backstage.st.dev → nginx ingress)
└─▶ Service → Deployment: python:3.12 running mcp_server.server over HTTP :8000
(echo / add / whoami). /mcp also served; /worker/* auth = ARCADE_WORKER_SECRET.
```
### Runtime facts (verified by introspecting `arcade-mcp-server` 1.17)
- `app.run()` honors env overrides via `_get_configuration_overrides()`:
`ARCADE_SERVER_TRANSPORT=http`, `ARCADE_SERVER_HOST=0.0.0.0`, `ARCADE_SERVER_PORT=8000`
— so the hardcoded `127.0.0.1` in `server.py` is overridden at runtime (no code change).
- `ARCADE_WORKER_SECRET` enables worker routes at `/worker/*`; the engine authenticates with
an HS256 JWT (`aud=worker`, `ver=1`) signed with that secret. MCP is served at `/mcp`.
## Components (three repos)
### 1. `arcade-eval` — image
- `lib/mcp_server/Dockerfile``python:3.12-slim`, `pip install .`, HTTP transport via env,
non-root, port 8000.
- `.github/workflows/build-push-acr.yml` — pushes
`servicetitandev.azurecr.io/arcade-eval-ref:1.0.<run_number>` (secrets
`ACR_DEV_USERNAME`/`ACR_DEV_PASSWORD`). Adapted from `servicetitan/mem0`.
### 2. `k8s-backstage-v2` — `apps/mcp/arcade-eval-ref/`
- `namespace.yaml` — ns `arcade-eval-ref`.
- `server.yaml`**st-app HelmRelease** (chart 2.0.72): `image` pinned to `1.0.1`,
`service.internalPort: 8000`, **`ingress.enabled` host `arcade-eval-ref.st.dev`
class `nginx`, `oAuth.enabled: false`** (no SSO wall over `/worker/*` or `/mcp`),
worker secret via `envFrom` from the SealedSecret, probes off. TLS = ingress default
`*.st.dev` wildcard cert.
- `sealedsecret.yaml``arcade-eval-ref-worker-secret` (key `ARCADE_WORKER_SECRET`),
strict scope, sealed with the backstage-wus2-v4 sealed-secrets cert.
### 3. `iac-terraform-workspaces` — DNS
- CNAME `arcade-eval-ref.st.dev``k8s-backstage.st.dev` (st.dev zone), mirroring the
`anvil`/`alerts` pattern.
## Registration (dashboard)
Add/repoint the worker: URI `https://arcade-eval-ref.st.dev`, Secret = the worker-secret
plaintext (git-ignored at `results/arcade-eval-ref-worker-secret.txt`). The engine then
fetches `/worker/tools` over the public URL → tools load → add to `zeb-gateway-test`.
## Verified
- `https://arcade-eval-ref.st.dev/worker/health` → 200 (valid `*.st.dev` LE cert);
`/worker/tools` with a correct worker JWT → 200, tools `Echo/Add/Whoami`.
- Through the gateway: `ArcadeEvalRef_Whoami()` → the caller's Entra `sub`
(`GvgRofe5…`), proving per-user execution across the full
client → gateway → engine → public URL → in-cluster pod chain.
## Alternative considered (not taken)
Declare the server as a static worker in the **engine config** (`tools.directors[].workers`,
like `arcade-worker-main`) — that path allows internal URIs and avoids public exposure, but
edits the vendor Helm release (`apps/arcade`) and loses the dashboard per-project workflow.
Public ingress was chosen as the lower-touch option.
+35
View File
@@ -0,0 +1,35 @@
# syntax=docker/dockerfile:1
#
# arcade-eval reference MCP server (echo / add / whoami).
#
# Runs over HTTP so the self-hosted Arcade engine can reach it in-cluster via a
# stable Service URL — replacing the ephemeral cloudflared tunnel used in dev.
# Deployed to backstage-wus2-v4 under apps/mcp/arcade-eval-ref/ (k8s-backstage-v2).
FROM python:3.12-slim
WORKDIR /app
# Install the package + runtime deps (arcade-mcp-server, httpx) declared in pyproject.toml.
COPY pyproject.toml ./
COPY src ./src
RUN pip install --no-cache-dir .
# arcade_mcp_server's app.run() reads these env vars via _get_configuration_overrides():
# - ARCADE_SERVER_TRANSPORT=http -> serve MCP at /mcp and worker routes at /worker/*
# - ARCADE_SERVER_HOST=0.0.0.0 -> bind all interfaces (server.py hardcodes 127.0.0.1;
# this env override is what makes it reachable in-cluster)
# - ARCADE_SERVER_PORT=8000
# ARCADE_WORKER_SECRET is injected by Kubernetes at runtime (from a SealedSecret); it
# authenticates the engine->worker connection and enables the /worker/* routes.
ENV ARCADE_SERVER_TRANSPORT=http \
ARCADE_SERVER_HOST=0.0.0.0 \
ARCADE_SERVER_PORT=8000
# Run as an unprivileged user.
RUN useradd --create-home --uid 10001 appuser
USER appuser
EXPOSE 8000
# server.py's __main__ calls app.run(); the env vars above override transport/host/port.
CMD ["python", "-m", "mcp_server.server"]