sendmail 8.14 has a serious memory corruption bug in it
wbh at conducive.org
Tue Feb 19 15:59:41 PST 2008
Claus Assmann wrote:
On Tue, Feb 19, 2008, Constantine A. Murenin wrote:
On 19/02/2008, Claus Assmann <dragonfly-kernel at esmtp.org> wrote:
No, they don't. I asked twice. (I could explain to you why they
I'm quite curious what the reason is -- do you mind sharing it?
I have only a single static IP (it's an old contract). PacBell/AT&T
only delegates reverse DNS to you if you have at least 6 (8?) static
IP addresses. Changing to that doubles the amount of money I would
have to pay. I've considered changing ISPs instead (BTW: why are
static IP so expensive in the US? It's more than twice the amount of
One may have a dsl router that holds onto the same dynamic IP for the
best part of a year, but over time, and 'on average' a dynamic IP can be
used to serve a great deal more than twice as many customers as a
Sometimes several hundred times (think dynamic PPPoE) - with near-zero
admin cost vs the fixed-IP, and with carrier retention of control over
yet-another toolset to block 'revenue loss' to, for example, VoIP, which
they offer in competiton to public/free(ish) services.
On a side note, if I were you, I'd probably complain to the ISP for
not specifying in their rDNS that your IP-address is static.
How should they do that? I don't know of any "policy" at AT&T to do
so... (or any "official" standard).
More information about the Kernel