Authelia now runs in front of Traefik, so the app was asking for a second
password at the same door. This removes its own authentication entirely
rather than layering the two.
## Breaking — the server needs both files in this deploy
`compose.yaml` and `.env` are not pulled from this repository. The new image
ignores `PASSWORD`, and the old image refuses to start without it, so the
image and the compose file have to move together or the container dies at
startup.
| Variable | Change |
|---|---|
| `PASSWORD` | **removed** — the app no longer reads it |
| `AUTH` | **new, required** — the Traefik middleware that authenticates the app, e.g. `authelia@docker` |
| `CERTRESOLVER` | **new, required** — the resolver issuing the certificate for `HOST` |
The `tls=true` label is replaced by `tls.certresolver=${CERTRESOLVER}`.
Naming a resolver implies TLS, so it stays one label rather than two — and
the resolver had been carried by hand on the server since the first deploy.
## What was removed
- `auth()` and `challenge()` — HTTP Basic over a single shared password
- `throttle.go` and its tests — the per-IP guess limiter and the
`X-Forwarded-For` handling that fed it
- `golang.org/x/time`, which existed only for that limiter
- Sixty `-u` flags from the smoke script
`routes()` returns the bare mux and `/healthz` is an ordinary route on it.
159 insertions against 446 deletions; nothing was written to replace what
went.
## What holds the app up now
Both invariants live in `compose.yaml`, next to comments saying why:
- **The router names the Authelia middleware through `AUTH`.** Traefik takes
a router out of service when its middleware does not resolve, so an unset
or misspelt value fails shut rather than serving the app open.
- **The container publishes no ports.** It is reachable only over the shared
proxy network. Publishing `8080` would now bypass authentication outright,
not merely TLS.
`/healthz` returns the version and nothing else, so it is safe to exempt in
Authelia if a monitor needs to reach it.
## Why this is stronger, not weaker
The layer being deleted was one shared secret with no sessions, no second
factor and no way to revoke access for one person. Authelia does all three,
configured once for every service on the host instead of reimplemented per
app. The weaker of the two prompts was the one being kept.
## Tests
`TestAuth` and `TestHealthzSkipsAuth` are replaced by a single test asserting
every route answers without credentials — a 401 from the app would now mean
authentication had crept back in. `make check` green; CI green on `dev`.
## Note on the commit list
Nine of the ten commits below are already in `main` via #4, squash-merged
under a different SHA. They contribute nothing to the diff, which is the
auth removal alone.
---------
Co-authored-by: Esa Kataja <[email protected]>
Reviewed-on: #5
The deployment is a public hostname behind Traefik rather than a LAN-only
box, which changes two things.
TLS is now terminated by the proxy, so Basic credentials are no longer in
cleartext. The container publishes no ports: doing so would leave an
unencrypted copy of the app on the host, bypassing the proxy. The hostname
lives in .env rather than compose.yaml, so no infrastructure detail is
committed and the MIT publication option stays open.
The password is now the only thing between the internet and the app, and a
500 ms sleep is not a defence at that exposure. Wrong guesses are rate
limited per client address: five in a burst, then one per ten seconds,
answered with 429.
Two details that decide whether this works at all:
- a request with no Authorization header is not charged. That is the
handshake every browser session opens with, and counting it would lock the
household out for simply opening the app a few times.
- X-Forwarded-For is believed only when the connection arrived from a private
address, i.e. through the proxy, and then only its last entry, which is the
one the proxy observed. A direct client could otherwise forge a fresh
address per attempt and walk past the limiter entirely.
None of this substitutes for a strong password. It removes brute force as a
practical route, nothing more. PRD §3, §9 and §10 are updated: "no external
internet exposure" is no longer true.
Bring up the stage 1 skeleton described in the PRD, enough that the app
builds, serves, and can be populated with dishes.
- net/http server with shared-password Basic auth, /healthz outside it,
graceful shutdown, and TZ-aware calendar days
- SQLite via modernc (pure Go, static binary), opened with WAL and a single
connection
- migration runner: numbered SQL files embedded and applied once each inside
a transaction, recorded in schema_migrations
- bundle import (PRD §7.3) as a live feature on the Ruoat tab: paste JSON or
upload a file, get a per-row Finnish report. The same importer is reachable
as `foodster -import` for repopulating a scratch database
- templ views and hand-written CSS with light-dark() theming; the Datastar
v1.0.3 client is vendored, since the Go SDK ships no browser asset and a
CDN would break an offline LAN
Names are normalized to sentence case rather than title case: Finnish
capitalizes only the first word of a phrase, so "Keitetyt perunat" is right
and "Keitetyt Perunat" is not. PRD §6 and §7.3 are amended to match.
Testing is behind make targets rather than ad-hoc commands: `make check` runs
lint, unit tests and scripts/smoke.sh, which exercises auth, static assets
and every import path against a scratch database on a spare port.
KessinenandEsa Kataja