Re: sendmail vunerability?
Timo Sirainen <[email protected]>
| Newsgroups | gmane.comp.security.full-disclosure,gmane.comp.security.vulnerabilities.watch.general |
|---|---|
| Message-ID | <1048924348.6856.12.camel@hurina> |
On Wed, 2003-11-19 at 14:55, nag wrote: > there is some rumor spreading about new sendmail vulnerability. i do not see > any news at www.sendmail.org but supposedly ths is remote buffer overflow. > i received the patch (see below), but o not have any exploit, so please > don't mail me about it. > > does this look legitimate? any comments? is this something old? is > this some scam backdoor? Assuming the address can contain high-ascii characters and it can be more than twice as long as pvpbsize, this is exploitable. I don't know if those conditions are prevented anywhere. Pretty interesting hole anyway, took a while to figure it out. Another problem may be that it calls isspace() and isascii() for negative values. Some systems don't like that. _______________________________________________ Full-Disclosure - We believe in it. Charter: http://lists.netsys.com/full-disclosure-charter.html