oss-sec mailing list archives

Re: GHOST gethostbyname() heap overflow in glibc (CVE-2015-0235)


From: Alexander Cherepanov <ch3root () openwall com>
Date: Fri, 30 Jan 2015 07:44:49 +0300

On 2015-01-30 03:28, Kees Cook wrote:
On Thu, Jan 29, 2015 at 4:02 PM, Solar Designer <solar () openwall com> wrote:
Paul, Kees -

On Thu, Jan 29, 2015 at 08:00:48AM -0800, Paul Pluzhnikov wrote:
On Thu, Jan 29, 2015 at 4:09 AM, Hanno B??ck <hanno () hboeck de> wrote:
And yes: I'd like people to cry alarm every time they see a buffer
overflow in glibc or any other core lib.

What is the appropriate forum to cry alarm on?

As a moderator for oss-security, I'd appreciate it if you cry alarm in
here.  And if this ever becomes too noisy, that would be an interesting
problem to have and we'll find a way to deal with it then. :-)

We are not a distro, and (AFAICT) are not on any of the closed lists.
But maybe we should be.

Actually, Chrome OS is listed as a member of linux-distros here:

http://oss-security.openwall.org/wiki/mailing-lists/distros

and the person subscribed on behalf of Chrome OS is Kees Cook
(previously representing Ubuntu).  Given your comment above, we have to
double-check whether this is currently correct.  Is Kees Cook currently
representing Chrome OS on linux-distros?  If so, why were you not aware
of that?  (I think this is unrelated to the handling of GHOST, but since
this was brought up we just have to deal with it as well.)

I'm representing Chrome OS on linux-distros, yes. As for GHOST, I
wasn't aware of the issue when it was fixed back in April in the
Chrome OS bug tracker -- it was handled by the package maintainers, it
seems, and never got escalated, unfortunately.

-Kees

I cannot help but ask: is https://code.google.com/p/chromium/issues/detail?id=364511#c9 from you? (Sorry, I see only a partial email address there, and I cannot expand it for some reason.) Was it automated then?

https://code.google.com/p/chromium/issues/detail?id=364511#c10 is from Tim Willis and he is from the Chrome security team, right? I'm not sure what you mean by "escalated" but comments #10 and #14 show that the Chrome security team can catch relevant issues itself (which is nice).

--
Alexander Cherepanov


Current thread: