Orinoco OEM Residential Gateway SNMP Community String Remote Configuration 
Vulnerability
BugTraq ID: 5436
Remote: Yes
Date Published: Aug 09 2002 12:00AM
Relevant URL:
http://www.securityfocus.com/bid/5436
Summary:

Orinoco is the manufacturer of various wireless network components,
including access points and network cards.

A problem with some systems manufactured by Orinoco may allow remote users
to gain access to sensitive information, and potentially make AP
configuration changes.

The Orinoco series OEM products typically use a unique identification
string to provide access control to the management interface.  This
identification string is unique and static.  This identification string is
used as the authentication string for performing configuration of the
access point.

It is possible to remotely gain access to the identification string used
for configuration of OEM access points manufactured by Orinoco through
SNMP.  By sending a custom-crafted SNMP query to a vulnerable access
point, the access point will return system credentials, including the
identification string.  This identification string can be used as the
administrative community string.

Through the use of this identification string as a SNMP community string,
a remote user may make configuration changes to the access point.  These
changes may include the alteration of domain name servers, and the wired
equivalent privacy key.

[ hardware ]

ISDN4Linux IPPPD Device String SysLog Format String Vulnerability
BugTraq ID: 5437
Remote: No
Date Published: Aug 10 2002 12:00AM
Relevant URL:
http://www.securityfocus.com/bid/5437
Summary:

isdn4linux is a freely available, open source package of isdn
compatibility tools.  It is available for Linux operating systems.

A problem with isdn4linux may make local code execution and privilege
elevation possible.

isdn4linux contains a format string vulnerability in the ipppd utility.
In some installations, this utility is installed with setuid root
privileges.  Exploitation of this vulnerability could lead to a local
attacker executing code with administrative privileges.

The problem is in handling of device strings.  By executing ipppd with an
excessively long device string (256 or greater bytes), and embedding
format string specifiers in the device string, it is possible to execute
arbitrary attacker-supplied instructions.

SecurityFocus staff have determined that this vulnerability has apparently
been fixed in version 3.2p1 of the software.  This has not been confirmed
by the vendor.

OpenBSD select() Buffer Overflow Vulnerability
BugTraq ID: 5442
Remote: No
Date Published: Aug 12 2002 12:00AM
Relevant URL:
http://www.securityfocus.com/bid/5442
Summary:

OpenBSD is a freely available, open source operating system designed with
security in mind. It is maintained and distributed by the OpenBSD project.

A buffer overflow vulnerability has been reported for the select(2)
function. The vulnerability occurs when using the select(2) call.
select(2) provides programmers with the facility to examine I/O
descriptors.

The size parameter for the select() function is a signed integer.
Reportedly, select() evaluates the upper boundary checks in a signed
context. As a result, an attacker is able to cause the kernel to overwrite
arbitrary locations in memory when supplying select() with certain
negative values for the size parameter.

An attacker can exploit this vulnerability by causing the kernel to
overwrite memory locations with arbitrary attacker-supplied values. This
may result in the attacker causing the kernel to execute malicious,
attacker-supplied code.

PGP / GnuPG Chosen Ciphertext Message Disclosure Vulnerability
BugTraq ID: 5446
Remote: Yes
Date Published: Aug 12 2002 12:00AM
Relevant URL:
http://www.securityfocus.com/bid/5446
Summary:

PGP and GnuPG are two popular implementations of the OpenPGP encryption
specification. Both are available for a range of platforms, including
Microsoft Windows and Linux based systems.

A weakness in the OpenPGP specification, as implemented by both products,
may allow an attacker to learn the plaintext contents of encrypted
communications. While some degree of user interaction is required, the
attack is very plausible against non-technical end users.

In order to exploit this issue, an attacker E must first intercept an
encrypted message of interest between two users, B and A. The attacker may
modify this message and inject additional content into the encrypted
content.  This modified message must then be transmitted to A, the
recipient of the original message.

The attacker must then entice A into decrypting this message, and
revealing the results of the decrypted message. This may occur if A
responds to the malicious message with text that includes the decrypted
contents. As the results of decryption will appear garbled and
meaningless, it is conceivable that A would reply and include the original
"quoted" message in an attempt to determine what has gone wrong.

Given the decrypted version of the malicious message, and the original
encrypted message, the attacker may recover a portion of the original
plaintext. In general the attacker will be able to recover at best half of
the plaintext content per attack, as it is difficult to modify the
encrypted length of the message, and an equal amount of injected content
is required in order to implement the attack. Under many applications this
will be sufficient, however multiple attacks may result in full disclosure
of the plaintext message.

It is not believed to be possible to exploit this weakness against message
content which is compressed during the OpenPGP encryption process.
Attacker supplied content will cause an error in the decompression process
with a high degree of probability, which may alert the end user or prevent
the display of the decrypted content. Compression is reported to be
enabled in both products by default. Files which are already compressed,
however, may not be compressed again, allowing exploitation.

It is important to note that exploitation of this issue will result in the
plaintext contents of a specific, intercepted message being disclosed to a
third party. The integrity of the private keys involved in the original
communication is not compromised, and widespread exploitation of this
weakness is extremely likely to be noticed by the end user.

W3C CERN httpd Proxy Cross-Site Scripting Vulnerability
BugTraq ID: 5447
Remote: Yes
Date Published: Aug 12 2002 12:00AM
Relevant URL:
http://www.securityfocus.com/bid/5447
Summary:

CERN httpd is a freely available HTTP server and HTTP proxy server
available from the W3C.

The httpd Proxy does not protect against cross-site scripting attacks.

When it cannot retrieve a web document, Proxomitron outputs an error
webpage.  The URL that it attempted to use is displayed without being
sanitized.  It is possible for attackers to construct urls that will cause
arbitrary HTML or script code to be embedded in the error page.  When the
client interprets the error page, the attacker-supplied code may execute
within the context of the proxy server.

This type of vulnerability may be used to steal cookies or perform other
web-based attacks.

[ bon, qui tourne encore ce truc ? :-> ]

L2TPD Weak Random Number Generator Seeding Vulnerability
BugTraq ID: 5451
Remote: Yes
Date Published: Aug 13 2002 12:00AM
Relevant URL:
http://www.securityfocus.com/bid/5451
Summary:

l2tpd is a Layer 2 Tunneling Protocol daemon, implementing the protocol
defined in RFC 2661.

Some versions of l2tpd fail to seed the random number generator before
calling the function rand(). This may result in predictable random numbers
being generated. Random numbers are used for a number of purposes within
l2tpd, including tunnel and session ids, and within the challenge /
response mechanism.

An attacker may be able to exploit this vulnerability to predict the
numbers which will be generated by l2tpd. This may allow a number of
attacks. The ability to guess session and tunnel ids may allow an attacker
to inject data into a valid conversation. The ability to predict the
behavior of the challenge / response mechanism may allow man in the middle
attacks, or some replay attacks. In both cases, the integrity of
connections made with l2tpd may be compromised.

The consequences of exploitation will be highly dependant on the
environment l2tpd is deployed in.

Red Hat Interchange Arbitrary File Read Vulnerability
BugTraq ID: 5453
Remote: Yes
Date Published: Aug 13 2002 12:00AM
Relevant URL:
http://www.securityfocus.com/bid/5453
Summary:

Interchange is a Web application development environment with a focus on
ecommerce and dynamic content management. It is available for Linux and
Unix variant operating systems.

A vulnerability has been reported for Interchange 4.8.5 and earlier.
Interchange may disclose contents of files to attackers.

The vulnerability occurs due to the placement of the 'doc' folder.
Reportedly, the folder will be installed as follows:
<INTERCHANGE_ROOT>/doc. This folder, by default, contains Interchange man
pages. This vulnerability is only exploitable when the Interchange service
runs in INET (Internet service) mode.

An attacker may exploit this vulnerability to the contents of restricted
files accessible to the Interchange process. The potentially sensitive
information obtained may be used to mount further attacks against a
vulnerable system.

It has been reported that this issue may be exploited through a '../'
directory traversal sequence in a HTTP request to the vulnerable server.
URLs may escape the document root in this manner, and request arbitrary
files on the system, subject to the permissions of the server process.

[ licence pas claire ]

Xinetd Open File Descriptor Denial Of Service Vulnerability
BugTraq ID: 5458
Remote: No
Date Published: Aug 13 2002 12:00AM
Relevant URL:
http://www.securityfocus.com/bid/5458
Summary:

Xinetd is intended as a secure replacement for inetd. It is designed for
use with Linux and Unix variant operating environments.

Reportedly, xinetd is vulnerable to a denial of service condition. The
vulnerability is the result of file descriptors for the signal pipe being
inherited by child processes launched by xinetd. This may result in a
malicious attacker access to pipes associated with xinetd thus having the
ability to communicate with xinetd.

Local attackers may misuse the open file descriptors by sending extraneous
or malformed data to xinetd which may cause the service to crash. This
results xinetd failing to respond to legitimate requests for service.

The signal pipe was introduced in version 2.3.4 of Xinetd.  Earlier
versions are not prone to this issue.


-
Pour poster une annonce: [EMAIL PROTECTED]

Répondre à