header-logo
Suggest Exploit
explore-vulnerabilities

Explore Vulnerabilities

Version
Year

Explore all Exploits:

ProFTPD 1.2pre0-3 Remote Exploit

The vulnerability in 1.2pre1, 1.2pre3 and 1.2pre3 is a remotely exploitable buffer overflow, the result of a sprintf() in the log_xfer() routine in src/log.c. The vulnerability in 1.2pre4 is a mkdir overflow. The name of the created path can not exceed 255 chars. 1.2pre6 limits the command buffer size to 512 characters in src/main.c and modifies the fix from 1.2pre4.

Caldera OpenLinux 2.2 ,Debian Linux 2.1/2.2,RedHat Linux <= 6.0 Vixie Cron MAILTO Sendmail Vulnerability

Failure by the vixie cron daemon from validating the contents of a user supplied environment variable allow a malicious users to pass arbitrary command line arguments to sendmail while running as the root user. The cron daemon uses the cron_popen function to send email to the user that queue commands for execution by cron. The user can set a MAILTO environment variable before calling cron to have it send the email to a different address. The cron daemon passes the contents of this environment variable to sendmail via the command line while executing as the root user. This allow a local malicious user to obtain root access.

Microsoft Internet Explorer 5.0 for Windows 95/Windows 98/Windows NT 4 HTML Form Control DoS

Certain Microsoft applications (IE5, Outlook Express 5) are unable to display large HTML form fields within HTML tables. Attempts to view the offending table will cause the application to 'hang' and consume 100% CPU resources until the application is closed.

Vixie cron Buffer Overflow Vulnerability

The version of Vixie cron that ships with RedHat versions 4.2, 5.2 and 6.0 is vulnerable to a local buffer overflow attack. By utilizing the MAILTO environment variable, a buffer can be overflown in the cron_popen() function, allowing an attacker to execute arbitrary code. Vixie cron daemon is installed setuid root by default, allowing for a local root compromise. Recent versions of Debian GNU/Linux have been confirmed to not be vulnerable to this attack.

Microsoft Internet Explorer 5.0 for Windows 95/Windows 98/Windows NT 4 ActiveX ‘Object for constructing type libraries for scriptlets’ Vulnerability

The 'scriptlet.typlib' ActiveX control can create, edit, and overwrite files on the local disk. This means that an executable text file (e.g. a '.hta' file) can be written to the startup folder of a remote machine and will be executed the next time that machine reboots. Attackers can exploit this vulnerability via a malicious web page or an email message.

pt_chown Vulnerability

pt_chown is a program included with glibc 2.1.x that exists to aid the proper allocation of terminals for non-suid programs that don't have devpts support. It is installed setuid root, and is shipped with RedHat Linux 6.0. As it stands, pt_chown is vulnerable to an attack that allows malicious users to write aribtrary data to tty input/output streams (open file desciptors -> tty) that don't belong to them (you could theoretically get full control of the terminal). This is done by fooling the program into giving you access (it lacks security checks). Whether you can be compromised or not depends on the software you are using and whether it has support for devpts (screen, midnight commander, etc). The consequences are hijacking of terminals, possibly leading to a root compromise.

W3-mSQL CGI Script Directory Traversal

Under certain versions of Mini SQL, the w3-msql CGI script allows users to view directories which are set for private access via .htaccess files. W3-mSQL converts any form data passed to a script into global Lite variables and these variables can then be accessed by the script code. An attacker can use this vulnerability to gain access to protected directories and files by using two approaches. The first approach requires the attacker to know the location/directory structure of the site they are attacking. The second approach will gain the intruder a DES encrypted password which they can then attempt to crack it via any number of popular cracking utilities.

Buffer Overflow in libtermcap’s tgetent() Function

A buffer overflow existed in libtermcap's tgetent() function, which could cause the user to execute arbitrary code if they were able to supply their own termcap file. Versions of libtermcap 2.0.8 and earliear are vulnerable. Under Red Hat Linux 5.2 and 4.2, this could lead to local users gaining root privileges, as xterm (as well as other possibly setuid programs) are linked against libtermcap. Under Red Hat Linux 6.0, xterm is not setuid root. Debian and Caldera OpenLinux use the ncurses library instead of termcap and thus are not vulnerable.

libtermcap xterm exploit

A buffer overflow existed in libtermcap's tgetent() function, which could cause the user to execute arbitrary code if they were able to supply their own termcap file. Versions of libtermcap 2.0.8 and earliear are vulnerable. Under Red Hat Linux 5.2 and 4.2, this could lead to local users gaining root privileges, as xterm (as well as other possibly setuid programs) are linked against libtermcap. Under Red Hat Linux 6.0, xterm is not setuid root. Debian and Caldera OpenLinux use the ncurses library instead of termcap and thus are not vulnerable.

Oracle Intelligent Agent Vulnerability

A vulnerability in the Oracle Intelligent Agent allows local malicious users to execute arbitrary commands and to create world writable files as the root user. The problem lies in the dbsnmp program located in $ORACLE_HOME/bin. This setuid root and setgid dba program trusts the environment variable ORACLE_HOME without verifying its contents. This vulnerability can be exploited in a number of ways. The dbsnmp program calls a tcl script (nmiconf.tcl) located by default in $ORACLE_HOME/network/agent/config. A malicious user can craft his own nmiconf.tcl script and fool the dbsnmp program to execute as root. When run without ORACLE_HOME being set, dbsnmp will dump two log files out into the current working directory: dbsnmpc and dbsnmpt. If these files do not exist, dbsnmp will attempt to create them mode 666 and dump around 400 bytes of uncontrollable output into them. If the files do exist, dbsnmp will append these 400 bytes but not change the permissions. Thus a malicious user can create world writable files in the system that do not exist (e.g. /.rhosts).

Recent Exploits: