Bug 109705 – media-libs/netpbm buffer overflow (CAN-2005-2978)

• Status: RESOLVED FIXED
• Product: Gentoo Security

[Description Thierry Carrez (RETIRED) 2005-10-18 07:04:56 UTC]

CAN-2005-2978 :
RedHat discovered a buffer overflow in the netpbm utility pnmtopng.

Prepared ebuild should be committed direct to stable on the following archs :
alpha amd64 hppa ppc ppc64 sparc x86

Also media-libs/urt-3.1b-r1 should be pushed to ppc64 stable at the same time.

[Comment 1 Thierry Carrez (RETIRED) 2005-10-18 07:05:38 UTC]

*** Bug 107609 has been marked as a duplicate of this bug. ***

[Comment 2 Thierry Carrez (RETIRED) 2005-10-18 07:06:34 UTC]

vapier: please commit your ebuild from bug 107609.

[Comment 3 Thierry Carrez (RETIRED) 2005-10-18 08:30:28 UTC]

Hm. In fact 10.29 is fixed, so we should move to that.

Calling arch testers again (sorry bout that):
Target KEYWORDS="alpha amd64 arm hppa ia64 mips ppc ppc64 sparc x86"

Stable any >=10.29 of your liking.

[Comment 4 Lares Moreau 2005-10-18 10:07:15 UTC]

x86:
emerged 10.29-r1, without issue. 
this ebuild depends on media-libs/urt which is also unstable in this arch.

perhaps a bug for media-libs/urt stabilization is in order, to handle the
stablization of this dependency.

[Comment 5 SpanKY 2005-10-18 10:32:13 UTC]

no one said 10.29-r1 needs to be the one stabilized

[Comment 6 Lares Moreau 2005-10-18 10:44:31 UTC]

Thierry Carrez wrote: "Hm. In fact 10.29 is fixed, so we should move to that."

is not 10.29-r1 the logical package to stabilize?

[Comment 7 Michael Hanselmann (hansmi) (RETIRED) 2005-10-18 11:10:05 UTC]

ppc and hppa done.

[Comment 8 Andrej Kacian (RETIRED) 2005-10-18 12:12:53 UTC]

If you don't want 10.29-r1 stabilized, don't say things like "Stable any >=10.29
of your liking." (comment #3). My liking was to stabilize 10.29-r1, because
changelog entry for that revision says it contains multiple fixes.

[Comment 9 SpanKY 2005-10-18 13:05:32 UTC]

no

what is logical is that you move to whatever package is the easiest or whichever
version an arch team decides on

[Comment 10 Andrej Kacian (RETIRED) 2005-10-18 14:04:22 UTC]

Oh, so we don't care about quality now, but about having to do least possible
amount of work now? Sorry I asked then.

[Comment 11 Andrej Kacian (RETIRED) 2005-10-18 14:25:30 UTC]

10.29 stable on x86

[Comment 12 Bryan Østergaard (RETIRED) 2005-10-18 16:04:41 UTC]

Alpha stable.

[Comment 13 Luis Medinas (RETIRED) 2005-10-18 16:21:53 UTC]

amd64 done

[Comment 14 Brent Baude (RETIRED) 2005-10-18 19:37:26 UTC]

Marked 10.29 ppc64 stable

[Comment 15 Thierry Carrez (RETIRED) 2005-10-19 00:25:15 UTC]

(In reply to comment #10)
> Oh, so we don't care about quality now, but about having to do least possible
> amount of work now? Sorry I asked then.

When multiple security-fixed versions are available, we (security) don't dictate
which fixed version the arch teams must choose. This is their choice to decide
which version is best fit for their arch stable tree. As long as the
vulnerability is fixed, we are ok with it. That's what vapier was trying to say
in his own words.

[Comment 16 Gustavo Zacarias (RETIRED) 2005-10-19 08:34:06 UTC]

10.29 sparc stable.

[Comment 17 Thierry Carrez (RETIRED) 2005-10-20 04:43:09 UTC]

GLSA 200510-18
mips should mark stable to benefit from GLSA

[Comment 18 Aaron Walker (RETIRED) 2005-10-21 05:14:42 UTC]

10.29 stable on mips.