Contactez-nous Suivez-nous sur Twitter En francais English Language

De la Théorie à la pratique

Freely subscribe to our NEWSLETTER

Newsletter FR

Newsletter EN



Vigil@nce: Lynx, two vulnerabilities

November 2008 by Vigil@nce

Two vulnerabilities of Lynx can be used by a local or remote attacker to execute code on victim’s computer.

- Gravity: 2/4
- Consequences: user access/rights
- Provenance: internet server
- Means of attack: no proof of concept, no attack
- Ability of attacker: expert (4/4)
- Confidence: confirmed by the editor (5/5)
- Diffusion of the vulnerable configuration: high (3/3)
- Number of vulnerabilities in this bulletin: 2
- Creation date: 28/10/2008


- Mandriva Corporate
- Mandriva Linux
- Mandriva Multi Network Firewall
- Red Hat Enterprise Linux
- Unix - plateform


The Lynx program is a text mode web browser. It has two vulnerabilities.

The "lynxcgi://" uris are used to execute local CGI scripts installed on user’s computer. In the interactive mode, Lynx warns the user before running these commands. However, when Lynx is run automatically as a url handler, the CGI command is directly executed. [grav:2/4; 468184, CVE-2008-4690]

The PERSONAL_MAILCAP (.mailcap) and PERSONAL_EXTENSION_MAP (.mime.types) files indicate how to open documents depending on their MIME type. For example, to open a document with the "image/gif" MIME type, run "gimp %s". However, Lynx searches these files in the current directory. An attacker can therefore create these files in a directory, and then invite the victim to start Lynx in this directory to see a GIF image, which will run the gimp program. [grav:1/4; 214205, CVE-2006-7234]


- Identifiers: 214205, 468184, CVE-2006-7234, CVE-2008-4690, MDVSA-2008:217, MDVSA-2008:218, RHSA-2008:0965-01, VIGILANCE-VUL-8206
- Url:

See previous articles


See next articles