Files
HealthLog/docs/self-hosting/reverse-proxy.md
T
Marc-André Bombeck f92e644154 chore(release): v1.4.49 — audit-backlog closure + server reminder suppression + diagnostic endpoint + analytics cold-path fix
Bundles the deferred v1.4.47 Mediums + Lows with the v1.4.48-deferred items
into a single release. Headline: server-side suppression of MEDICATION_REMINDER
APNs for iOS clients that own their local reminders (closes GH #206), a
push_attempts table backing /api/admin/notifications/diagnostic, and a sweep
of v1.4.48 QA forward-findings.

Performance (the user-visible win):
- /api/analytics cold-fallback path splits the heavy single-statement
  aggregate into two parallel queries. The linearly composable columns
  (count / min / max / mean) keep the full-partition scan; the windowed
  7 / 30 / 90-day avg / slope / r² columns take a 90-day outer cap so
  the planner does an index range scan on (user_id, type, measured_at)
  instead of a full-partition sequential scan. Output is bit-identical;
  cold slim slice drops from multi-second to sub-second on tenants with
  large measurement partitions.

Server-side notifications:
- GET + PATCH /api/auth/me/notification-prefs with deep-merge semantics.
  Initial category medication.clientManaged: boolean (default false). Cron
  medicationReminderJob skips dose-due APNs when the flag is true,
  emitting a medication_reminder.suppressed_client_managed wide-event
  annotation per skip.
- push_attempts table backs the diagnostic endpoint. Every APNS /
  Web-Push / Telegram / NTFY sender writes a fire-and-forget row.
  Daily 03:35 Europe/Berlin cleanup prunes rows older than 90 days.
- /api/admin/notifications/diagnostic — admin-only endpoint surfacing
  device APNs token presence (masked), channel config presence, and
  recent push attempts. Registered in OpenAPI.
- Per-channel APNs "Send test" button now fires with
  eventType: "MEDICATION_REMINDER" so the dispatcher's time-sensitive +
  priority-10 branch runs. A successful self-test now exercises the
  exact path production reminders take; an unsuccessful one points the
  user at iOS Settings → Time Sensitive Notifications rather than
  appearing to be a server-side bug.

Withings reliability:
- OAuth callback consumes the nonce atomically (delete-first with
  P2025 catch for the replay branch). The previous read-then-delete
  pattern allowed a race between two concurrent callbacks.
- ?reason=state redirect differentiated into csrf1 / replay / expired /
  cross_user so operators can tell mismatch types apart without DB
  shell access.
- connect endpoint rate-limited to 10/60s/user; row-create failures
  redirect to a consistent error surface.

Observability hardening:
- redactSensitiveFields(body) denylist helper covers password / token /
  secret / apiKey / authorization / csrfState / nonce, recursive,
  case-insensitive. Caught: /api/devices invalid_format issues were
  echoing apnsToken values verbatim into the audit log.
- sanitiseZodIssues(issues, { stripValuesFromMessage: true }) overload
  applied to 14 audit-log sites.

Migrations: 0079 (notification_prefs), 0080 (push_attempts).

Tests: 4815 → 5272 unit + integration. Workout-batch integration suite
that had been red on main since v1.4.45 (three release commits shipped
through it) now green again — updated to match v1.4.42 cross-source
dedup semantics.

Resolves PR #207.
2026-05-23 12:51:41 +02:00

221 lines
8.1 KiB
Markdown

# Reverse proxy configuration
The bundled `docker-compose.yml` exposes HealthLog on plain HTTP at
`localhost:3000`. Production deployments front the container with a
reverse proxy that terminates TLS, sets the forwarded headers the
rate-limiter trusts, and preserves the WebSocket upgrade path the
streaming Coach surface depends on.
This guide gives copy-paste blocks for five common fronts. Pick one;
the env-var pairing is identical across them.
## Pre-flight — env vars the proxy interacts with
| Variable | Required for | Notes |
| -------- | ------------ | ----- |
| `NEXT_PUBLIC_APP_URL` | Every public-facing absolute URL the UI emits | Must match the hostname the proxy serves |
| `APP_URL` | Server-side absolute URLs from background jobs | Mirror of the above; both must match |
| `TRUST_PROXY_HOPS` | `X-Forwarded-For` chain trust | Defaults to `1` — one trusted hop |
Set the URL pair to the public origin (`https://your-instance.example.com`)
and restart the `app` container. The image reads both at startup; no
rebuild needed.
### `TRUST_PROXY_HOPS` — what to set it to
`src/lib/api-response.ts` (`getClientIp`) reads the `X-Forwarded-For`
chain from the **right**, counting back `TRUST_PROXY_HOPS` entries.
This guards against a client rotating `X-Forwarded-For` per request
to defeat IP-based rate limits.
| Topology | `TRUST_PROXY_HOPS` |
| -------- | ------------------ |
| App is internet-facing with no proxy | `0` (XFF ignored entirely) |
| Single proxy in front (Caddy / Traefik / Nginx / Coolify) | `1` (default) |
| Cloudflare → your proxy → app | `2` |
| Cloudflare → Coolify-Tunnel → Coolify → app | `3` |
A misconfigured count logs a one-shot warning to stderr and collapses
every anonymous caller into one shared rate-limit bucket. Match the
value to the actual hop count or set it to `0` and let `x-real-ip`
drive the IP resolution.
## Caddy
`Caddyfile`:
```
your-instance.example.com {
encode zstd gzip
reverse_proxy localhost:3000 {
header_up X-Forwarded-For {remote_host}
header_up X-Forwarded-Proto {scheme}
header_up Host {host}
flush_interval -1
}
}
```
Caddy handles the WebSocket upgrade automatically. `flush_interval -1`
disables response buffering so the streaming insight events surface
without batching delays. With Caddy as the only hop, leave
`TRUST_PROXY_HOPS=1`.
## Traefik
`docker-compose.override.yml` snippet:
```yaml
services:
app:
labels:
- "traefik.enable=true"
- "traefik.http.routers.healthlog.rule=Host(`your-instance.example.com`)"
- "traefik.http.routers.healthlog.entrypoints=websecure"
- "traefik.http.routers.healthlog.tls.certresolver=letsencrypt"
- "traefik.http.services.healthlog.loadbalancer.server.port=3000"
- "traefik.http.middlewares.healthlog-headers.headers.customrequestheaders.X-Forwarded-Proto=https"
- "traefik.http.routers.healthlog.middlewares=healthlog-headers"
```
Traefik forwards `X-Forwarded-For` by default and respects the WebSocket
upgrade. Pair with `TRUST_PROXY_HOPS=1`. If Traefik sits behind
Cloudflare, set it to `2` so the rate-limiter reads the real client IP
that Cloudflare placed in the chain.
## Nginx Proxy Manager
In the NPM UI:
1. **Proxy Hosts → Add Proxy Host.**
2. **Domain Names:** `your-instance.example.com`.
3. **Scheme:** `http`; **Forward Hostname / IP:** the Docker host or
internal hostname; **Forward Port:** `3000`.
4. Enable **Websockets Support** (required for streaming Coach
responses).
5. **SSL tab:** request a Let's Encrypt certificate, force SSL.
6. **Advanced tab — Custom Nginx Configuration:**
```nginx
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
proxy_set_header Host $host;
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
proxy_read_timeout 300s;
```
The 300-second read timeout matters for long-running insight
generations — anything shorter will cut streamed responses mid-flight.
Pair with `TRUST_PROXY_HOPS=1`.
## Coolify
Coolify's bundled Traefik handles most of this. The pieces that need
operator attention:
1. **Application → Domains:** set `https://your-instance.example.com`.
Coolify provisions the certificate and adds the Traefik labels.
2. **Application → Environment Variables:** set
`NEXT_PUBLIC_APP_URL`, `APP_URL`, and any optional secrets
(`WITHINGS_*`, `APNS_*`, `LOKI_*`).
3. **Application → General → Pull policy:** `always`. The bundled
`docker-compose.yml` declares this explicitly (v1.4.34.2 fix) so
Coolify re-checks the GHCR digest on every redeploy instead of
reusing the cached `:latest` layer.
4. **Notifications → Webhook:** point at `/api/internal/deploy-webhook`
with the `X-Deploy-Webhook-Secret` header set to
`DEPLOY_WEBHOOK_SECRET`. The app gates the endpoint with a
timing-safe compare against that env var.
### DO NOT set as runtime env vars
The following variables are **build-time only** — they are injected
via `--build-arg` by the `docker-publish` workflow and baked into the
shipped image. Setting them under Coolify's *Environment Variables*
panel has zero effect on the running container, and the stale value
becomes a misleading artefact next time someone audits the deploy
config:
- `NEXT_PUBLIC_APP_VERSION` — sourced from the GHCR tag at build
time. Next.js inlines the value into both the client bundle (so
the `<VersionPoller>` self-healing reload compares against the
shipped shell) and the `/api/version` server route (so the public
endpoint returns the same string). The `env: {…}` block in
`next.config.ts` short-circuits the runtime `process.env` read.
- `NEXT_PUBLIC_APP_BUILD_SHA`, `NEXT_PUBLIC_APP_BUILT_AT` — same
pattern. The CI workflow stamps the short SHA and build timestamp
into the image during `pnpm build`.
If you spot any of these in the Coolify panel, delete the entry. The
image already carries the correct value.
Coolify-Tunnel adds one hop to the chain; pair with
`TRUST_PROXY_HOPS=2`. If you front Coolify itself with Cloudflare,
bump to `3`.
## Bare Nginx
`/etc/nginx/sites-available/healthlog`:
```nginx
upstream healthlog {
server 127.0.0.1:3000;
keepalive 16;
}
server {
listen 443 ssl http2;
listen [::]:443 ssl http2;
server_name your-instance.example.com;
ssl_certificate /etc/letsencrypt/live/your-instance.example.com/fullchain.pem;
ssl_certificate_key /etc/letsencrypt/live/your-instance.example.com/privkey.pem;
client_max_body_size 1600M; # Apple Health export.zip cap is 1.5 GB
location / {
proxy_pass http://healthlog;
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
proxy_read_timeout 300s;
proxy_buffering off;
}
}
server {
listen 80;
listen [::]:80;
server_name your-instance.example.com;
return 301 https://$host$request_uri;
}
```
`client_max_body_size 1600M` is the load-bearing setting if you plan
to support Apple Health `export.zip` uploads — the app caps the
upload at 1.5 GB and Nginx will reject the body before the app sees
it otherwise. `proxy_buffering off` keeps streamed insight events
flowing without buffer pauses. Pair with `TRUST_PROXY_HOPS=1`.
## Verifying the proxy chain
Once the proxy is in front, hit `https://your-instance.example.com/api/version`
from outside the host and check `docker compose logs app` for the
request. The wide-event log line should carry a non-null client IP
that matches your real source address. A `null` IP or a single
`unknown` rate-limit bucket means `TRUST_PROXY_HOPS` is set higher
than the actual proxy hop count, or the proxy is dropping the
`X-Forwarded-For` header entirely.
The one-shot stderr warning `[getClientIp] TRUST_PROXY_HOPS=N but
X-Forwarded-For carried M entries` confirms a mismatch. Adjust the
env var down to the actual chain length and restart.