CVE-2024-39358
A buffer overflow vulnerability exists in the adm.cgi set_wzap() functionality of Wavlink AC3000 M33A8.V5030.210505. A specially crafted HTTP request can lead to stack-based buffer overflow. An attacker can make an authenticated HTTP request to trigger this vulnerability.
The versions below were either tested or verified to be vulnerable by Talos or confirmed to be vulnerable by the vendor.
Wavlink AC3000 M33A8.V5030.210505
Wavlink AC3000 - https://www.wavlink.com/en_us/product/WL-WN533A8.html
9.1 - CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:C/C:H/I:H/A:H
CWE-120 - Buffer Copy without Checking Size of Input (‘Classic Buffer Overflow’)
The Wavlink AC3000 wireless router is predominately one of the most popular gigabit routers in the US, in part due to both its potential wireless and wired speed capabilities and extremely low price point (costing at the time of this writing ~$60 USD). Among the configuration options, it’s also able to act as a standalone wireless gateway, a basic network router, or a wireless repeater.
When interacting with and configuring the Wavlink AC3000 wifi router, as is typical of most wifi routers, an administrator logs in via some web portal and configures appropriate options via the HTTP interface. In the case of this particular router, and in another somewhat common execution pattern, these HTML pages can invoke .cgi binaries due to how the lighttpd server is configured. Since all of these .shtml and .cgi files are located in the web root, anyone with network access to the device doesn’t actually need to log in to the device to interact with these .cgi files, and it usually is the responsibility of the .cgi binary to check if the authentication is completed successfully. On this device, one will see a check_valid_user() function in each individual .cgi binary which will check the session cookie of the HTTP request to see if it’s coming from a validly logged in user.
Assuming that we’ve passed this check in the adm.cgi binary, we then run into a set of functions that we can call based off of what we pass for the page= parameter in our HTTP POST request. Of the available commands, we focus on the following:
00401884 else if (strcmp(webget_page, "wzdap") == 0)
00401cb0 set_wzdap(contlen_buf)
If we provide page=wzdap, we enter the set_wzdap function and our provided POST data is further parsed therein:
0040c680 int32_t $v0 = nvram_bufget(0, "TouchLinkEn")
0040c69c int32_t $v0_1 = nvram_bufget(0, "Model")
0040c6b8 int32_t $v0_2 = nvram_bufget(0, "CountryCode")
// [...]
0040c8c4 nvram_bufset(0, "UserInit", &data_415db0)
0040c8fc int32_t $v0_10 = strdup(web_get("wl_Method", arg1, 0))
0040c934 int32_t wlan_ssid2g_dup = strdup(web_get("wlan_ssid2g", arg1, 0)) // [1]
0040c96c int32_t $v0_13 = strdup(web_get("wlan_ssid5g", arg1, 0))
0040c9a4 int32_t $v0_15 = strdup(web_get("wlan_ssid5g_2", arg1, 0))
0040c9dc int32_t $v0_17 = strdup(web_get("web_pskValue", arg1, 0))
0040ca14 int32_t $v0_19 = strdup(web_get("static_en", arg1, 0))
0040ca4c int32_t $v0_21 = strdup(web_get("EncrypType", arg1, 0))
0040ca84 int32_t $v0_23 = strdup(web_get("skiplist1", arg1, 0))
0040cabc int32_t $v0_25 = strdup(web_get("skiplist2", arg1, 0))
0040caf4 int32_t $v0_27 = strdup(web_get("skiplist3", arg1, 0))
0040cb14 nvram_bufset(2, "AutoChannelSkipList", $v0_23)
0040cb30 nvram_bufset(0, "AutoChannelSkipList", $v0_25)
0040cb4c nvram_bufset(1, "AutoChannelSkipList", $v0_27)
0040cb68 nvram_bufset(0, "UserInit", &data_415db0)
0040cb84 nvram_bufset(0, "OperationMode", "0")
0040cba0 nvram_bufset(0, "ENWISP", "0")
0040cbc0 nvram_bufset(0, "wanConnectionMode", "DHCP")
0040cbdc nvram_bufset(0, "IEEE80211H", "0")
0040cbf8 nvram_bufset(0, "natEnabled", "0")
0040cc14 nvram_bufset(2, "ApCliEnable", "0")
0040cc30 nvram_bufset(2, "SSID1", wlan_ssid2g_dup)
0040cc54 if (strcmp($v0, "1") == 0)
0040d090 void var_90
0040d090 sprintf(&var_90, "%s_Touch", wlan_ssid2g_dup) // [2]
At [1], the binary copies our wlan_ssid2g POST parameter to the heap and then copies our input buffer to the stack at [2]. Since there’s no length checking on our input buffer, we can easily provide a buffer longer than 0x90 bytes, which will overwrite the return address of this function and quickly result in arbitrary code execution.
Thread 2.1 "adm.cgi" hit Breakpoint 3, 0x0040d090 in set_wzdap ()
(gdb) x/1s $a2
0x5730c8: "\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177\bp\377\177AAAAAAAA"
(gdb) c
Continuing.
[Detaching after fork from child process 29513]
Thread 2.1 "adm.cgi" received signal SIGSEGV, Segmentation fault.
0x41414141 in ?? ()
(gdb) info reg
zero at v0 v1 a0 a1 a2 a3
R0 00000000 00000000 00000000 ffffffff 00000003 7fd1ebd4 00000000 00000000
t0 t1 t2 t3 t4 t5 t6 t7
R8 00000000 00000000 00000000 00000000 00000001 00000000 00000000 00000001
s0 s1 s2 s3 s4 s5 s6 s7
R16 7fff7008 7fff7008 7fff7008 7fff7008 7fff7008 7fff7008 7fff7008 7fff7008
t8 t9 k0 k1 gp sp s8 ra
R24 00000000 76fbf450 00000010 00000000 7705c490 7fd1ed28 41414141 41414141
status lo hi badvaddr cause pc
0100fc13 00000000 00000000 41414140 50800008 41414141
fcsr fir hi1 lo1 hi2 lo2 hi3 lo3
00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
dspctl restart
00000000 00000000
2024-07-25 - Initial Vendor Contact
2024-07-29 - Requesting reply from vendor
2024-07-30 - Vendor confirms receipt
2024-07-30 - Vendor Disclosure
2024-07-30 - Vendor confirms receipt
2024-09-02 - Status update request sent
2024-10-15 - Status update request. Upcoming expiration date announced.
2024-10-22 - Vendor replies product has been discontinued, but patches are being worked on
2024-11-04 - Status update request for patch release dates
2024-11-12 TALOS advisory release date announced
2025-01-14 - Public Release
Discovered by Lilith >_> of Cisco Talos.