Cloudanix VulnDB

CVE-2021-27250

Medium · 6.5

CVSS

6.5

EPSS

67.4%

Published

Apr 14, 2021

CISA due date

—

Cloudanix analysis

Next patch cycle SSVC: Attend

A nearby attacker could trick an affected network device or service into revealing saved passwords or configuration secrets.

Chainable
Exploitation
Adjacent Network
Attacker needs
None
Complexity
Low
Impact
Data Exposure
Component
Firmware
Class
Path Traversal
Remediation
Upgrade
Effort
Medium
Patch risk
Medium

What it actually does

This is an unauthenticated path traversal issue in CGI request handling. The vulnerable code uses the user-controlled errorpage request parameter in file operations without sufficiently normalizing and constraining the path, allowing a network-adjacent attacker to request files outside the intended web/CGI directory. Successful exploitation can disclose sensitive local files, including stored credentials or configuration data. The CVSS vector indicates adjacent-network reachability, low attack complexity, no authentication, no user interaction, and high confidentiality impact, with no direct integrity or availability impact.

Why it matters to the business

An attacker on the same local network or Wi-Fi segment as an affected device or service could steal saved credentials or configuration secrets and use them to expand access into management interfaces, wireless infrastructure, source-control systems, or other connected services. Although it is not listed in CISA KEV and is not known to be used in ransomware campaigns, the EPSS score is high at approximately 0.67 and has remained stable, indicating meaningful likelihood of exploitation in the wild or by opportunistic scanners. Because the issue has been public for several years, any still-exposed vulnerable deployment should be treated as a priority remediation item.

How an attack unfolds

An attacker first gains network adjacency, such as access to the same LAN, Wi-Fi network, guest network, or routed management segment that can reach the affected CGI endpoint. The attacker then sends crafted HTTP requests to the vulnerable CGI script using the errorpage parameter with traversal sequences such as ../ or encoded equivalents. Because the application does not properly validate the supplied path before using it in file operations, the request can cause the device or application to return files outside the expected directory. The attacker focuses on files likely to contain credentials, hashes, configuration secrets, wireless keys, or administrative settings. Any recovered secrets can then be used to authenticate to the device, pivot to related management interfaces, or compromise other systems that reuse the exposed credentials.

How to fix it

Upgrade affected software or firmware to a fixed release. For source-based deployments identified as usvn/usvn (GIT), update to or beyond commit 0efafc31671dbb9a128cf310ce1baf253a094bef. For affected D-Link DAP-2020 v1.01rc001 installations, replace the vulnerable firmware with a vendor-fixed version if available, retire unsupported hardware, or isolate it from untrusted and user-accessible networks. As compensating controls, restrict access to device management and CGI interfaces to a dedicated management VLAN or trusted administrator IP ranges, disable remote management from wireless or guest networks, block direct access to vulnerable CGI paths where operationally possible, and rotate any credentials that may have been stored on or accessible through the affected system. Long-term fixes should ensure file paths are canonicalized, traversal sequences are rejected, and file access is restricted to an allowlisted directory.

How to confirm the fix worked

Confirm the installed product and version or source commit against the fixed version information, specifically verifying that source deployments include commit 0efafc31671dbb9a128cf310ce1baf253a094bef or later where applicable. For appliance deployments, verify firmware version and vendor support status. From a trusted test host on the same network segment, safely confirm that requests using traversal payloads in the errorpage parameter are rejected with an error and do not return local file contents. Review web server, embedded HTTP daemon, reverse proxy, firewall, and network IDS logs for historical requests containing traversal sequences in the errorpage parameter. After remediation, retest access controls to ensure management/CGI endpoints are not reachable from guest, user, or untrusted adjacent networks.

Typical attack path

  1. Identify reachable affected devices or services from an adjacent network segment.
  2. Probe CGI endpoints that accept an errorpage parameter.
  3. Send path traversal payloads using ../ sequences or URL-encoded variants.
  4. Read sensitive local files or configuration files returned by the vulnerable handler.
  5. Extract stored credentials, password hashes, keys, or management configuration data.
  6. Use the disclosed credentials to log in, change settings, pivot laterally, or support further compromise.

How to detect it

Look for HTTP requests to CGI endpoints containing the errorpage parameter with path traversal patterns. Relevant log sources include embedded web server logs, reverse proxy logs, WAF logs, firewall logs, wireless controller logs, DHCP/NAC records for adjacent clients, NetFlow, packet captures, and IDS alerts. Focus on requests using ../, ..%2f, %2e%2e%2f, double-encoded traversal strings, or attempts to read common sensitive files such as passwd, shadow, configuration, or password storage files. Correlate suspicious requests with successful HTTP 200 responses, unusually large response bodies, administrative logins shortly afterward, configuration changes, new device sessions, or authentication attempts using local administrator accounts. Because exploitation is adjacent-network scoped, identify the source MAC/IP, wireless association history, switch port, VLAN, SSID, and any guest-network presence for the requesting client. If credentials may have been exposed, review downstream authentication logs for reuse of device, service, or administrative credentials after the suspicious request time.

  • errorpage=../
  • errorpage=..%2f
  • errorpage=%2e%2e%2f
  • errorpage=..%252f
  • errorpage=%252e%252e%252f
  • errorpage=....//
  • errorpage=..\
  • errorpage=..%5c
  • errorpage=%2e%2e%5c
  • errorpage=../../etc/passwd
  • errorpage=../../etc/shadow
  • errorpage=../../etc/config
  • errorpage=../../var/passwd
  • errorpage=../../var/etc/passwd
  • GET /cgi-bin/
  • POST /cgi-bin/
  • /cgi-bin/
  • ../etc/passwd
  • %2e%2e%2fetc%2fpasswd
  • ..%2f..%2f..%2f

What your security team should do

  • Inventory affected D-Link DAP-2020 v1.01rc001 devices and any source-based deployments mapped to usvn/usvn that may require commit 0efafc31671dbb9a128cf310ce1baf253a094bef or later.
  • Prioritize remediation for devices reachable from Wi-Fi, guest, user, or shared network segments because exploitation requires only network adjacency and no authentication.
  • Patch, upgrade firmware, or retire unsupported affected systems; do not rely on obscurity of CGI paths as a control.
  • Restrict management and CGI interfaces to trusted administrator networks using firewall, ACL, VLAN, or VPN controls.
  • Rotate administrative, wireless, service, and reused credentials that may have been stored on affected systems before the fix.
  • Search historical logs for traversal attempts against the errorpage parameter and investigate any successful responses or unusual follow-on logins.
  • Add SIEM and network IDS detections for URL traversal payloads targeting CGI endpoints, especially requests containing errorpage= with ../ or encoded traversal sequences.

The record

This vulnerability allows network-adjacent attackers to disclose sensitive information on affected installations of D-Link DAP-2020 v1.01rc001 Wi-Fi access points. Authentication is not required to exploit this vulnerability. The specific flaw exists within the processing of CGI scripts. When parsing the errorpage request parameter, the process does not properly validate a user-supplied path prior to using it in file operations. An attacker can leverage this vulnerability to disclose stored credentials, leading to further compromise. Was ZDI-CAN-11856.

Affected packages

Ecosystem Package Introduced Fixed

Related vulnerabilities