SMB dissector in Ethereal 0.9.3 and earlier allows remote attackers to cause a denial of service (crash) or execute arbitrary code via malformed packets that cause Ethereal to dereference a NULL pointer.
Sudo before 1.6.6 contains an off-by-one error that can result in a heap-based buffer overflow that may allow local users to gain root privileges via special characters in the -p (prompt) argument, which are not properly expanded.
Macintosh clients, when using NT file system volumes on Windows 2000 SP1, create subdirectories and automatically modify the inherited NTFS permissions, which may cause the directories to have less restrictive permissions than intended.
Autogalaxy stores usernames and passwords in cleartext in cookies, which makes it easier for remote attackers to obtain authentication information and gain unauthorized access via sniffing or a cross-site scripting attack.
The default "basic" security setting' in config.php for TWIG webmail 2.7.4 and earlier stores cleartext usernames and passwords in cookies, which could allow attackers to obtain authentication information and gain privileges.
Pathways Homecare 6.5 uses weak encryption for user names and passwords, which allows local users to gain privileges by recovering the passwords from the pwhc.ini file.
6tunnel 0.08 and earlier does not properly close sockets that were initiated by a client, which allows remote attackers to cause a denial of service (resource exhaustion) by repeatedly connecting to and disconnecting from the server.
ValiCert Enterprise Validation Authority (EVA) Administration Server 3.3 through 4.2.1 uses insufficiently random data to (1) generate session tokens for HSMs using the C rand function, or (2) generate certificates or keys using /dev/urandom instead of another source which blocks when the entropy pool is low, which could make it easier for local or remote attackers to steal tokens or certificates via brute force guessing.
Internet Explorer 6 and earlier, when used with the Telnet client in Services for Unix (SFU) 2.0, allows remote attackers to execute commands by spawning Telnet with a log file option on the command line and writing arbitrary code into an executable file which is later executed, aka a new variant of the Telnet Invocation vulnerability as described in CVE-2001-0150.
Perception LiteServe 1.25 allows remote attackers to obtain source code of CGI scripts via URLs that contain MS-DOS conventions such as (1) upper case letters or (2) 8.3 file names.
By default, DNS servers on Windows NT 4.0 and Windows 2000 Server cache glue records received from non-delegated name servers, which allows remote attackers to poison the DNS cache via spoofed DNS responses.
prefs.php in phpBB 1.4.0 and earlier allows remote authenticated users to execute arbitrary PHP code via an invalid language value, which prevents the variables (1) $l_statsblock in prefs.php or (2) $l_privnotify in auth.php from being properly initialized, which can be modified by the user and later used in an eval statement.
dnskeygen in BIND 8.2.4 and earlier, and dnssec-keygen in BIND 9.1.2 and earlier, set insecure permissions for a HMAC-MD5 shared secret key file used for DNS Transactional Signatures (TSIG), which allows attackers to obtain the keys and perform dynamic DNS updates.
Task Manager in Windows 2000 does not allow local users to end processes with uppercase letters named (1) winlogon.exe, (2) csrss.exe, (3) smss.exe and (4) services.exe via the Process tab which could allow local users to install Trojan horses that cannot be stopped with the Task Manager.
Transsoft Broker 5.9.5.0 allows remote attackers to read arbitrary files and directories by uploading a .lnk (link) file that points to the target file.
ArGoSoft FTP Server 1.2.2.2 allows remote attackers to read arbitrary files and directories by uploading a .lnk (link) file that points to the target file.
WFTPD 3.00 allows remote attackers to read arbitrary files by uploading a (link) file that ends in a ".lnk." extension, which bypasses WFTPD's check for a ".lnk" extension.
FTP service in IIS 5.0 and earlier allows remote attackers to cause a denial of service via a wildcard sequence that generates a long string when it is expanded.
sash before 3.4-4 in Debian GNU/Linux does not properly clone /etc/shadow, which makes it world-readable and could allow local users to gain privileges via password cracking.
The Winsock2ProtocolCatalogMutex mutex in Windows NT 4.0 has inappropriate Everyone/Full Control permissions, which allows local users to modify the permissions to "No Access" and disable Winsock network connectivity to cause a denial of service, aka the "Winsock Mutex" vulnerability.
IBM WebSphere server 3.0.2 allows a remote attacker to view source code of a JSP program by requesting a URL which provides the JSP extension in upper case.
Unify eWave ServletExec allows a remote attacker to view source code of a JSP program by requesting a URL which provides the JSP extension in upper case.
The default configuration of BEA WebLogic 3.1.8 through 4.5.1 allows a remote attacker to view source code of a JSP program by requesting a URL which provides the JSP extension in upper case.
Eudora 4.x allows remote attackers to bypass the user warning for executable attachments such as .exe, .com, and .bat by using a .lnk file that refers to the attachment, aka "Stealth Attachment."
IIS 4.0 and 5.0 allows remote attackers to cause a denial of service by sending many URLs with a large number of escaped characters, aka the "Myriad Escaped Characters" Vulnerability.
Windows NT 4.0 does not properly shut down invalid named pipe RPC connections, which allows remote attackers to cause a denial of service (resource exhaustion) via a series of connections containing malformed data, aka the "Named Pipes Over RPC" vulnerability.
Lynx 2.x does not properly distinguish between internal and external HTML, which may allow a local attacker to read a "secure" hidden form value from a temporary file and craft a LYNXOPTIONS: URL that causes Lynx to modify the user's configuration file and execute commands.
Compaq/Microcom 6000 Access Integrator does not disconnect a client after a certain number of failed login attempts, which allows remote attackers to guess usernames or passwords via a brute force attack.