Remote Desktop Exposed to the Internet

Remote Desktop can be reached directly from the internet, where it is a constant target for password guessing and a common way in for ransomware groups.

SeverityMedium
CVSS 3.15.6 CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:L Open in the calculator
CVSS 4.06.3 CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N Open in the calculator
CWECWE-1327: Binding to an Unrestricted IP Address
MITRE ATT&CKT1133 External Remote Services, T1110.003 Password Spraying, T1021.001 Remote Desktop Protocol
CategorySecurity Misconfiguration

Description

The Remote Desktop service (TCP port 3389 by default, and UDP 3389 where enabled) can be reached directly from the internet. Remote Desktop gives a full interactive session on the host, which makes its sign-in a high-value target: internet-facing Remote Desktop services are found by continual automated scanning and receive password guessing and credential stuffing attempts, and national cyber security agencies, including the NCSC, have named exposed Remote Desktop as one of the most common ways ransomware groups gain initial access.

An attacker who guesses or obtains the password of any account allowed to use Remote Desktop gains an interactive session inside the network, from which they can steal data, move to other systems and deploy ransomware. Accounts with weak or reused passwords are the usual entry point. The likelihood rises with the number of accounts allowed to sign in and falls where an account lockout policy, strong passwords and Network Level Authentication are in place, but none of these gives the protection of a second factor. Any future pre-authentication vulnerability in Remote Desktop Services would also be exploitable directly from the internet.

Exposure is confirmed by a port scan from the internet and by connecting with a Remote Desktop client or the Nmap rdp-ntlm-info script, which, through the NTLM exchange used by Network Level Authentication, also discloses the host name, domain name and operating system build without credentials. Moving the service to a non-standard port does not prevent it being found.

Remediation

Remove direct internet access to Remote Desktop and provide remote access through a controlled gateway:

  • Block TCP and UDP port 3389, and any other port the service listens on, at the internet boundary.
  • Provide remote access through a VPN or a Remote Desktop Gateway that requires multi-factor authentication, or through a zero trust access service, so that only authenticated users can reach Remote Desktop hosts.
  • Where direct exposure must remain for a short time, restrict it to known source addresses with firewall rules.
  • Enforce Network Level Authentication, strong passwords and an account lockout policy, and limit the Remote Desktop Users group to the accounts that need it.
  • Monitor sign-in events, including failed sign-ins (event ID 4625) and successful Remote Desktop sign-ins (event ID 4624 with logon type 10), and alert on unusual patterns.

Because exposed Remote Desktop services are attacked routinely, review the host for signs of earlier compromise. Confirm the fix with a port scan from the internet, which should no longer reach the service.

References


Add This Finding to Your Library

In Pental this write-up is in Settings, Library. Add to my library copies it into your own, rated in your firm's CVSS version, ready to put in your own words.

Free trial  ·  No card required

Only need the reporting? Our reporting tool is free and open source, with the same library.


More Network Services Findings

Every Network Services finding