FuzzManager配置与使用

用户1423082

发布于 2024-12-31 20:09:21

8600

代码可运行

文章被收录于专栏：giantbranch's blog

运行总次数：0

代码可运行

Server

下载

git clone https://github.com/MozillaSecurity/FuzzManager.git

安装依赖（注：EC2SpotManager是管理Amazon Cloud的实例的，一般用不到，用的话需要安装redis-server——apt install redis-server）

cd FuzzManager
pip install -r server/requirements.txt

这Server使用Django编写

Django manage.py migrate根据搜索到的文档，是创建数据库，数据表

cd server
python manage.py migrate

报错解决：只要复制FTB目录到server目录里面即可

root@bogon:~/FuzzManager/server# python manage.py migrate
Traceback (most recent call last):
  File "manage.py", line 10, in <module>
    execute_from_command_line(sys.argv)
  File "/usr/local/lib/python2.7/dist-packages/django/core/management/__init__.py", line 364, in execute_from_command_line
    utility.execute()
  File "/usr/local/lib/python2.7/dist-packages/django/core/management/__init__.py", line 338, in execute
    django.setup()
  File "/usr/local/lib/python2.7/dist-packages/django/__init__.py", line 27, in setup
    apps.populate(settings.INSTALLED_APPS)
  File "/usr/local/lib/python2.7/dist-packages/django/apps/registry.py", line 108, in populate
    app_config.import_models()
  File "/usr/local/lib/python2.7/dist-packages/django/apps/config.py", line 202, in import_models
    self.models_module = import_module(models_module_name)
  File "/usr/lib/python2.7/importlib/__init__.py", line 37, in import_module
    __import__(name)
  File "/root/FuzzManager/server/crashmanager/models.py", line 15, in <module>
    from FTB.ProgramConfiguration import ProgramConfiguration
  File "/root/FuzzManager/server/crashmanager/FTB/ProgramConfiguration.py", line 24, in <module>
    from FTB.ConfigurationFiles import ConfigurationFiles
ImportError: No module named FTB.ConfigurationFiles

复制完就可以了

root@bogon:~/FuzzManager/server# python manage.py migrate
Operations to perform:
  Apply all migrations: admin, auth, authtoken, contenttypes, covmanager, crashmanager, ec2spotmanager, sessions
Running migrations:
  Applying contenttypes.0001_initial... OK
  Applying auth.0001_initial... OK
  Applying admin.0001_initial... OK
  Applying admin.0002_logentry_remove_auto_add... OK
  Applying contenttypes.0002_remove_content_type_name... OK
  Applying auth.0002_alter_permission_name_max_length... OK
  Applying auth.0003_alter_user_email_max_length... OK
  Applying auth.0004_alter_user_username_opts... OK
  Applying auth.0005_alter_user_last_login_null... OK
  Applying auth.0006_require_contenttypes_0002... OK
  Applying auth.0007_alter_validators_add_error_messages... OK
  Applying auth.0008_alter_user_username_max_length... OK
  Applying authtoken.0001_initial... OK
  Applying authtoken.0002_auto_20160226_1747... OK
  Applying crashmanager.0001_squashed_0020_add_app_permissions... OK
  Applying covmanager.0001_initial... OK
  Applying covmanager.0002_increase_collection_filename_length... OK
  Applying covmanager.0003_collection_file_optional... OK
  Applying covmanager.0004_reportconfiguration_reportsummary... OK
  Applying covmanager.0005_report... OK
  Applying ec2spotmanager.0001_squashed_0013_add_gce_fields... OK
  Applying sessions.0001_initial... OK

创建fuzzmanager用户

root@bogon:~/FuzzManager/server# python ./manage.py createsuperuser
Username (leave blank to use 'root'):
Email address: fuzzmanager@test.com
Password:
Password (again):
Superuser created successfully.

获取fuzzmanager authorization token（下面的root是上面新建的用户名，这个看README的话是可以给Apache+WSGI设置虚拟主机用的，用token生成.htpasswd文件htpasswd -cb .htpasswd root 4a253efa90f514bd89ae9a86d1dc264aa3133945）

root@bogon:~/FuzzManager/server# python manage.py get_auth_token root
4a253efa90f514bd89ae9a86d1dc264aa3133945

本地测试

python manage.py runserver

访问http://127.0.0.1:8000/即可

这个只是监听127.0.0.1，假如是服务器，还得开个反向代理或者ssh代理才能访问，所以可以下面这样

python manage.py runserver 0.0.0.0:8000

访问了一下，发现需要在配置文件FuzzManager/server/server/settings.py中添加ALLOWED_HOSTS，就是HTTP请求的Host字段，添加本机的ip地址，假如有域名添加域名也行。

客户端

可以使用下面命令向服务器提交

python Collector.py --autosubmit mybadprogram --someopt yourtest

当然这之前得有配置文件~/.fuzzmanagerconf，下面是示例，那个sigdir是signatures存放目录，

[Main]
sigdir = /home/example/signatures
serverhost = 127.0.0.1
serverport = 8000
serverproto = http
serverauthtoken = 4a253efa90f514bd89ae9a86d1dc264aa3133945

尝试fuzz upx，提交试试，首先配置服务器信息~/.fuzzmanagerconf

[Main]
sigdir = /root/fuzz/upx/sigs
serverhost = 192.168.XX.XX
serverport = 8000
serverproto = http
serverauthtoken = 4a253efa90f514bd89ae9a86d1dc264aa3133945

配置程序信息upx.out.fuzzmanagerconf，放在二进制文件当前目录

[Main]
platform = x86-64
product = upx
product_version = UPX-git-d7ba31+
os = linux

[Metadata]
pathPrefix = /root/fuzz/upx
buildFlags =

之后运行命令即可（把/usr/local/lib/python2.7/dist-packages/Collector中的Collector.py出来即可使用）

python Collector.py --tool afl --autosubmit ./upx.out ./afl_out/crashes/id\:000000\,sig\:11\,src\:000120\,op\:arith8\,pos\:16168\,val\:+2

但是这个需要你的二进制程序是加了asan参数进行编译的，下面没有报错就是成功了

不然默认x86-64是不支持的，上传其实是上传了的，只不过没有获取到crash地址（后来发现是装了gdb插件的问题）

看了下源码，不加asan应该只支持x86和arm的自动提交https://github.com/MozillaSecurity/FuzzManager/blob/0ccde1820f6412da23f20da17eb3b5b9091a563e/FTB/Signatures/CrashInfo.py#L1245，需要向上翻翻看

但是后来发现是我的gdb装了peda，pwndgb插件，导致Collector.py脚本没识别出来。。。

后来禁用插件后就可以了

或者自己写代码提交，下面这个来源于https://www.fuzzingbook.org/html/FuzzingInTheLarge.html

from FTB.Signatures.CrashInfo import CrashInfo
from Collector.Collector import Collector


collector = Collector()

cmd = ["simply-buggy/simple-crash"]
result = subprocess.run(cmd, stderr=subprocess.PIPE, stdout=subprocess.PIPE)
stderr = result.stderr.decode().splitlines()
stdout = result.stdout.decode().splitlines()
crashInfo = CrashInfo.fromRawCrashData(stdout, stderr, configuration)
print(crashInfo)


collector.submit(crashInfo)

但是这个现在已经不能正常运行了，而且这个是基于python3的，我写了一个python2的版本

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Date    : 2020-03-23 11:37:56
# @Author  : giantbranch
# @Link    : http://www.giantbranch.cn/
# @tags : 

from FTB.Signatures.CrashInfo import CrashInfo
from FTB.ProgramConfiguration import ProgramConfiguration
from Collector.Collector import Collector
import subprocess

binary = "./src/upx.out_x86-64"
binaryArgs = "./afl_out/crashes/id:000000,sig:11,src:000120,op:arith8,pos:16168,val:+2"

FTB_GDB_SCRIPT_PATH = "/usr/local/lib/python2.7/dist-packages/FTB/Running/GDB.py"

configuration = ProgramConfiguration.fromBinary(binary)
print("configuration:")
print(configuration.product, configuration.platform)

gdbArgs = [
    "--batch",
    "-ex",
    "source %s" % FTB_GDB_SCRIPT_PATH,
    "-ex",
    "run %s" % binaryArgs
]

gdbArgs.extend([
    "-ex", "set pagination 0",
    "-ex", "set backtrace limit 128",
    "-ex", "bt",
    "-ex", "python printImportantRegisters()",
    "-ex", "x/2i $pc",
    "-ex", "quit",
])

cmdArgs = []
# cmd = ["gdb " + binary + " -ex \"r ./afl_out/crashes/id:000000,sig:11,src:000120,op:arith8,pos:16168,val:+2\""]

cmdArgs.append("gdb")
cmdArgs.extend(gdbArgs)
cmdArgs.append(binary)

print cmdArgs

process = subprocess.Popen(
    cmdArgs,
    stdin=subprocess.PIPE,
	stdout=subprocess.PIPE,
	stderr=subprocess.PIPE,
)


(stdout, stderr) = (process.stdout.read(), process.stderr.read())
# 假如不是传文件路径，是标准输入的，可能需要下面代码，input就是要传入的标准输入
# try:
#     stdout, stderr = process.communicate(input)
# except:
#     process.kill()
#     process.wait()

# print "===============stdoutstdout==============================="
# print stdout

# Detect where the GDB trace starts/ends
traceStart = stdout.rfind("Program received signal SIG")
traceStop = stdout.rfind("A debugging session is active")

# Alternative GDB start version when using core dumps
if traceStart < 0:
    traceStart = stdout.rfind("Program terminated with signal")

if traceStop < 0:
    traceStop = len(stdout)

# Move the trace from stdout to auxCrashData
auxCrashData = stdout[traceStart:traceStop]
stdout = stdout[:traceStart] + stdout[traceStop:]

# print "==============stdout============="
# print stdout.splitlines()


# print "==============stderr============="
# print auxCrashData.splitlines()

crashInfo = CrashInfo.fromRawCrashData(stdout.splitlines(), auxCrashData.splitlines(), configuration)
print(crashInfo)

testcase = binaryArgs
(testCaseData, isBinary) = Collector.read_testcase(testcase)
crashInfo.testcase = testCaseData

collector = Collector(tool="afl")

collector.submit(crashInfo, testcase)