stheg olloydson
2008-Apr-06 20:13 UTC
CVE-2008-1391 - Multiple BSD Platforms "strfmon()" Function Integer Overflow
Hello, According to the information at mitre.org, both 6.x and 7.0 are vulnerable. I see in NetBSD's CVS log for src/lib/libc/stdlib/strfmon.c, they have patched this on March 27. Looking at FreeBSD's CVS log at http://www.freebsd.org/cgi/cvsweb.cgi/src/lib/libc/stdlib/strfmon.c, shows that no changes have been made since Mon Sep 12, 2005. Is our strfmon() not vulnerable as reported? stheg ____________________________________________________________________________________ You rock. That's why Blockbuster's offering you one month of Blockbuster Total Access, No Cost. http://tc.deals.yahoo.com/tc/blockbuster/text5.com
Simon L. Nielsen
2008-Apr-06 21:15 UTC
CVE-2008-1391 - Multiple BSD Platforms "strfmon()" Function Integer Overflow
On 2008.04.06 12:47:11 -0700, stheg olloydson wrote:> According to the information at mitre.org, both 6.x and 7.0 are > vulnerable. I see in NetBSD's CVS log for > src/lib/libc/stdlib/strfmon.c, they have patched this on March > 27.Note that the change in NetBSD is possibly incomplete to fix the issue. I'm not sure what the final conclusion was on that.> Looking at FreeBSD's CVS log at > http://www.freebsd.org/cgi/cvsweb.cgi/src/lib/libc/stdlib/strfmon.c, > shows that no changes have been made since Mon Sep 12, 2005. > Is our strfmon() not vulnerable as reported?The FreeBSD version is affected and will be fixed in -CURRENT / HEAD shortly. The FreeBSD Security Team has yet to be able to come up with any real cases where this is an actual security issue, so unless we find any place where this is actually a problem, the issue will be handled as a normal bug and merged to -STABLE branches acordingly. Note that allowing untrusted format strings to be used is normally a bad idea, so any application where the strfmon issue is a problem are likely already broken. -- Simon L. Nielsen FreeBSD Security Team