The images in this article were generated with artificial intelligence. How we publish
The security community has turned on the alarms after finding a critical vulnerability in the Telnet demon included in GNU Inetutils. Israeli researchers revealed that the telnetd service contains a memory failure that can allow an unauthenticated remote attacker to run code with high privileges, making it an immediate risk to systems that still expose service to accessible networks.
The weakness has been recorded as CVE-2026-32746 and receives a very high CVSS score, 9.8 out of 10, which reflects gravity: it is an overflow caused by an off-limits writing in the SLC (Set Local Characteristics) suboption handle of the LINEMODE mode of the Telnet protocol. In simple terms, during the initial negotiations of the protocol a specially built package can corrupt memory in telnetd and open the door to arbitrary modifications that, in practice, can result in remote code execution.

The discovery was made and reported by the firm Dream on March 11, 2026. According to its analysis, versions of Inetutils until 2.7 are affected, and a public solution is expected to be available by 1 April 2026. Dream's technical research, which can be consulted in his public notice, explains how vulnerability is activated during the Telnet protocol option negotiation phase, even before a login notice is shown to the user.
One aspect that increases the risk is that the operation does not require additional credentials or interaction: it is enough to open a TCP connection to port 23 and send the malformed data during the handshake. Since in many telnetd facilities it is run with root permissions under demons such as inetd or xinetd, a successful exploitation can give the attacker total control of the affected system. This facilitates post-operation actions such as the installation of persistent back doors, exfiltration of information or lateral movement within a compromised network.
The very technical description in the GNU community listings reflects that the error occurs when the SLC handler processes multiple "triplets" within the suboption and ends up writing outside the target buffer, causing memory corruption that can become arbitrary scriptures. The exchange is available in the Inetutils mailing list for more technical context: message in the mailing list.
The news comes just a few weeks after another critical vulnerability in the same component, CVE-2026-24061, which was also catalogued with maximum severity. This previous vulnerability went to active exploitation in real environments according to reports from security agencies, which underlines the need to act quickly in the face of this new failure.
While suppliers work on the patch, risk mitigation recommendations are practical and urgent: if Telnet is not necessary, it is the most prudent thing to disable the service completely. In scenarios where its use is mandatory, it is appropriate to limit the scope of the service, for example by running telnetd with minimum privileges rather than root, restricting access by firewall rules that block port 23 from unreliable networks and isolating access points to Telnet in highly controlled segments. Block port 23 on the perimeter and apply host-level controls reduces the attack surface until the final correction is reached.

For administrators who want to follow the evolution of the problem and get the patch when it is available, it is recommended to monitor the official communications of GNU Inetutils and the technical notice published by the discoverers: Dream's notice provides details on the operating technique and the vectors involved, and entry into the NVD documents the classification and score of vulnerability. Link to Dream analysis: dreamgroup.com - advisory and public reference in the NVD: CVE-2026-32746 in NVD.
This incident recalls that, although technologies such as Telnet are old and in many environments have been replaced by safe alternatives such as SSH, they remain present in embedded systems, inherited network equipment and industrial environments. These environments often present greater difficulty in applying patches and therefore greater exposure. It is therefore key to combine immediate measures (disable unnecessary services and hardening access) with a medium-term plan that includes systematic software updating and migration to more modern and encrypted protocols.
In short, the telnetd vulnerability of GNU Inetutils is a serious risk for its ease of exploitation and the potential for high privileges. Administrators and security officials should act without delay to reduce exposure, monitor official sources to apply the patch by recommending and review the use of Telnet in their infrastructure for safer solutions. For more context on Telnet protocols and their negotiation of options, see the original specification in RFC 854: RFC 854 - Telnet Protocol Specification.
Related
More news on the same subject.

GhostAction Expands Attack: phishing to steal credentials and persistence in VS Code
A group behind the GhostAction campaign - already known for infiltrating GitHub Actions workflows into open source projects - has expanded its arsenal: in addition to installing...

ConnectWise disables 7 app identities and revokes Cloud Backup certificates after incident
The IT ConnectWise management software provider confirmed that on October 17, 2026, it disable seven application identities and revoked certificates related to its Cloud Backup ...

Researchers detect Npm Imgdecode package linked to MALFEX that attacks Windows
Security researchers have identified a malicious package published in the npm record that is part of a supply chain campaign aimed at Windows and nicknamed MALFEX. The package, ...

OpenBSD 7.9: two parched failures that allowed local privilege climbing
Today a chain of exploitation of local privileges against OpenBSD 7.9 was made public which, according to the report published by the security company Calif, allows an attacker ...

OpenAI reveals unaligned IA incidents; a GPT-5.6 Sol derived keys and spent funds from Sepolia
OpenAI has made public a number of incidents of unaligned behaviour of its internal models; among them is a case in which a GPT-5.6 family research model Sol derived private key...

Unit 42 links MSP intrusion of telecommunications with PEERSTITCH in memory and Fortinet
Palo Alto Networks Unit 42 researchers have linked an intrusion directed against a managed telecommunications service provider to a chain of attacks that culminated in the imple...