Files
website/server/src/utils/trustProxy.js
Claude d38c98ad9e Harden admin login: RBAC-safe controls, 2FA, bot-scoring, rate limits (#9)
Adds a layered set of protections around the admin login and the app edge.

Trust proxy (server/src/utils/trustProxy.js)
- Configurable via TRUST_PROXY; pin to the newt agent ("ptero") LAN IP so
  X-Forwarded-For is trusted ONLY from that peer. A blanket "true" is
  rejected (coerced to 1) to prevent XFF spoofing that would dodge every
  IP-based control. DEBUG_TRUST_PROXY logs peer/XFF/req.ip to re-verify the
  proxy IP without a redeploy. Documents the Omada static-reservation
  assumption.

Login throttling (server/src/middleware/loginProtection.js, rateLimit.js)
- express-slow-down progressive delay + the existing hard rate cap + a
  separate per-IP exponential backoff that persists across the rate window.
  All failures return one generic message (no user/pass disclosure).

Honeypot (login form + auth.controller)
- Hidden, plausibly-named field ("company"); a filled value fails
  generically and is scored as an unambiguous bot.

Optional per-user TOTP 2FA (speakeasy/qrcode)
- totp_secret/totp_enabled columns (+ idempotent migration). Self-service
  Account page: enroll via QR, confirm a code to enable, code-gated disable.
- Login is two-step for enrolled users: after the password, a short-lived
  signed challenge (stage:'totp', not a session) is required before the
  real session is issued.

Bot / scanner scoring + IP ban (server/src/middleware/botScore.js)
- Weighted CMS-scanner paths (this app uses none). Junk paths 404 FIRST,
  unconditionally — independent of score/ban state, so a scanner rotating
  through fresh Cloudflare IPs gets no free pass. /wp-admin/install.php is
  the top-weighted near-1-hit ban (worst offender in prod logs). Per-IP
  score with quiet-period decay temp-bans an IP from ALL routes once past a
  (deliberately low) threshold, to protect /admin from credential stuffing.
  Failed logins and honeypot hits feed the same score.
- Periodic sweep evicts stale, unbanned, quiet entries so the in-memory
  store can't grow unbounded; the interval is unref'd and cleared on
  graceful shutdown.

Tests: node --test suite (40) covering trust-proxy parsing + live req.ip
(incl. pinned-IP), rate limiter + exponential backoff, honeypot rejection,
TOTP verify (enabled/disabled) + challenge-isn't-a-session, bot-score
threshold/decay/ban + junk-404-independence + install.php + store sweep.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-02 23:22:35 -05:00

103 lines
4.5 KiB
JavaScript

// ── Trust proxy configuration ──────────────────────────────────────────────
//
// Real request path for this deployment:
//
// client → Pangolin → newt tunnel agent ("ptero", separate VM)
// → this app (its own VM), over the LAN
//
// The hop that actually opens the TCP connection to this app is ptero, so from
// Express's point of view ptero is THE trusted proxy and its LAN IP is the
// right-most/most-recently-added entry to reconcile against. The real client IP
// arrives in X-Forwarded-For. req.ip / req.secure must reflect the real client
// because the rate limiter, exponential backoff, bot-scoring ban, and activity
// log all key on req.ip — so this is a prerequisite for every other control.
//
// Recommended production value: pin TRUST_PROXY to ptero's LAN IP exactly. That
// is stricter than a hop count: Express will only honour XFF on connections that
// actually come from ptero, so nothing else on the LAN can inject a forwarded
// header. See TRUST_PROXY in .env.example for how to set it.
//
// IMPORTANT ASSUMPTION: pinning ptero's IP assumes ptero holds a STATIC IP
// (a DHCP reservation in Omada). If that reservation does not exist, a lease
// change would silently move ptero to a new IP and every XFF would stop being
// trusted — req.ip would collapse to ptero's (new) address for all clients,
// breaking rate limiting/bans. Verify the reservation before relying on this,
// and use DEBUG_TRUST_PROXY (below) to re-check the observed proxy IP without a
// code redeploy if it ever needs to change.
//
// We deliberately DO NOT use a blanket `true`. `true` trusts every hop and takes
// the left-most (client-supplied, spoofable) XFF entry, letting an attacker forge
// their apparent IP to dodge rate limits / bans.
//
// TRUST_PROXY accepts:
// - unset / '' -> 1 (single proxy hop fallback)
// - an integer -> that many trusted hops (e.g. "2")
// - "false" -> false (no proxy; direct connections only)
// - "loopback" etc. -> the express preset string, passed through
// - a CSV of IPs / CIDRs -> that list (e.g. ptero's LAN IP: "10.0.0.42")
//
// `true` is intentionally rejected (coerced to 1) with a warning, so it can't be
// set by accident.
const log = require('./logger')('trustproxy')
function truthyEnv(v) {
return ['1', 'true', 'yes', 'on'].includes(String(v || '').trim().toLowerCase())
}
const PRESETS = new Set(['loopback', 'linklocal', 'uniquelocal'])
function parseTrustProxy(raw = process.env.TRUST_PROXY) {
const val = (raw == null ? '' : String(raw)).trim()
if (val === '') return 1 // default: one hop (Pangolin)
if (val.toLowerCase() === 'false') return false
if (val.toLowerCase() === 'true') {
log.warn('TRUST_PROXY=true is unsafe (trusts spoofable client XFF); using 1 hop instead')
return 1
}
// Pure integer → hop count.
if (/^\d+$/.test(val)) return Number(val)
// Single express preset keyword.
if (PRESETS.has(val.toLowerCase())) return val.toLowerCase()
// Otherwise treat as a comma-separated list of trusted IPs / CIDRs (and/or
// preset keywords), which Express accepts as an array.
const list = val
.split(',')
.map((s) => s.trim())
.filter(Boolean)
return list.length === 1 ? list[0] : list
}
// Apply the setting to an Express app and log what was chosen.
function applyTrustProxy(app, raw = process.env.TRUST_PROXY) {
const setting = parseTrustProxy(raw)
app.set('trust proxy', setting)
log.info('trust proxy configured', { setting: Array.isArray(setting) ? setting.join(',') : setting })
return setting
}
// Temporary diagnostic middleware, OFF by default. Set DEBUG_TRUST_PROXY=1 to
// log, per request, the raw peer address and forwarded header alongside the IP
// Express resolved — so the real proxy IP (ptero) can be re-verified in-place
// without a code change if it ever moves. Mounted before the bot guard so it
// still fires for scanner/junk requests (whose source IPs are what we want to
// see). Turn it back off once verified; it is noisy.
function trustProxyDebug(req, res, next) {
if (truthyEnv(process.env.DEBUG_TRUST_PROXY)) {
log.info('trust-proxy debug', {
remoteAddress: req.socket && req.socket.remoteAddress,
xForwardedFor: req.headers['x-forwarded-for'] || null,
resolvedIp: req.ip,
method: req.method,
path: req.originalUrl,
})
}
return next()
}
module.exports = { parseTrustProxy, applyTrustProxy, trustProxyDebug }