Does this affect Suse 9.0 (mutt 1.4.1i)? Thanks Jörg ----- Forwarded message from Thomas Roessler <roessler@does-not-exist.org> ----- Mailing-List: contact bugtraq-help@securityfocus.com; run by ezmlm Delivered-To: mailing list bugtraq@securityfocus.com Delivered-To: moderator for bugtraq@securityfocus.com Date: Wed, 11 Feb 2004 15:21:14 +0100 From: Thomas Roessler <roessler@does-not-exist.org> To: BUGTRAQ@securityfocus.com Subject: Mutt-1.4.2 fixes buffer overflow. Mail-Followup-To: BUGTRAQ@securityfocus.com Mutt-1.4.2 has just been released; this version fixes a buffer overflow that can be triggered by incoming messages. There are reports about spam that has actually triggered this problem and crashed mutt. It is recommended that users of mutt versions prior to 1.4.2 upgrade to this version, or apply the patch included below. Users of "unstable" mutt versions after 1.3.28 (including 1.5.*) do not need to upgrade, as this problem had been fixed in the unstable branch in February 2002; unfortunately, the fix was not backported before 1.4 was released. mutt-1.4.2 can be found at ftp://ftp.mutt.org/mutt/. Distribution files: MD5 checksum file name size 44fc379c317109f516894a7c3fd43cc9 diff-1.4.1i-1.4.2i.gz (23k) 6045b47cbba8170d6a9fdccc1aa817b9 mutt-1.4.2i.tar.gz (2.4M) Linux distributors are expected to release updated mutt packages shortly. Credits: The problem in the stable mutt code base was originally reported to Red Hat, and was brought to my attention by Mark Cox. The Common Vulnerabilities and Exposures project (cve.mitre.org) has assigned the name CAN-2004-0078 to this issue. Regards, -- Thomas Roessler · Personal soap box at <http://log.does-not-exist.org/>. Index: menu.c =================================================================== RCS file: /cvs/mutt/mutt/menu.c,v retrieving revision 2.27.2.1 diff -u -r2.27.2.1 menu.c --- menu.c 28 Jan 2002 10:18:50 -0000 2.27.2.1 +++ menu.c 11 Feb 2004 10:05:52 -0000 @@ -148,30 +148,13 @@ menu->make_entry (s, l, menu, i); } -void menu_pad_string (char *s, size_t l) +void menu_pad_string (char *s, size_t n) { - size_t n = mutt_strlen (s); int shift = option (OPTARROWCURSOR) ? 3 : 0; - - l--; /* save room for the terminal \0 */ - if (l > COLS - shift) - l = COLS - shift; + int cols = COLS - shift; - /* Let's just pad the string anyway ... */ - mutt_format_string (s, INT_MAX, l, l, 0, ' ', s, n, 1); - return; - -#if !defined (HAVE_BKGDSET) && !defined (USE_SLANG_CURSES) - /* we have to pad the string with blanks to the end of line */ - if (n < l) - { - while (n < l) - s[n++] = ' '; - s[n] = 0; - } - else -#endif - s[l] = 0; + mutt_format_string (s, n, cols, cols, 0, ' ', s, strlen (s), 1); + s[n - 1] = 0; } void menu_redraw_full (MUTTMENU *menu) ----- End forwarded message ----- -- Joerg Mayer <jmayer@loplof.de> We are stuck with technology when what we really want is just stuff that works. Some say that should read Microsoft instead of technology.