Re: freeimage and CVE-2019-12214
Hi Santiago and Salvatore
Now I realize what I did wrong. I should have put openjpeg on a line later.
Do you want me to fix that or do you want to do that yourself?
This is my proposal if I change only for buster.
ola@tigereye:~/git/security-tracker$ git diff
diff --git a/data/CVE/list b/data/CVE/list
index eeae88cade..7d8f058247 100644
--- a/data/CVE/list
+++ b/data/CVE/list
@@ -347440,13 +347440,17 @@ CVE-2019-12214 (In FreeImage 3.18.0, an
out-of-bounds access occurs because of m
- freeimage <unfixed> (bug #947478)
[bookworm] - freeimage <postponed> (Revisit when upstream
fixes are available)
[bullseye] - freeimage <postponed> (Revisit when upstream
fixes are available)
- [buster] - freeimage <postponed> (Revisit when upstream fixes
are available)
+ [buster] - freeimage <not-affected> (Do not include openjpeg
copy since 3.10.0-3)
[stretch] - freeimage <postponed> (Revisit when upstream fixes
are available)
[jessie] - freeimage <postponed> (Revisit when upstream fixes
are available)
+ - openjpeg2 2.1.0-1
NOTE: https://sourceforge.net/p/freeimage/discussion/36111/thread/e06734bed5/
NOTE: very few information regarding this vulnerability, which
is seemingly located
NOTE: in libopenjpeg, not freeimage. Without reproducer or
stacktrace, this is
NOTE: nearly unfixable.
+ NOTE: Turned out that the issue is not in freeimage at all,
but rather in openjpeg.
+ NOTE: For more information see
https://lists.debian.org/debian-lts/2024/04/msg00058.html
+ NOTE: and more specifically
https://lists.debian.org/debian-lts/2024/04/msg00081.html
CVE-2019-12213 (When FreeImage 3.18.0 reads a special TIFF file, the
TIFFReadDirectory ...)
{DSA-4593-1 DLA-2031-1}
- freeimage 3.18.0+ds2-3 (bug #929597)
Or do you want me to adjust also the no-affected to apply to all relases?
If so it would look like this:
ola@tigereye:~/git/security-tracker$ git diff
diff --git a/data/CVE/list b/data/CVE/list
index eeae88cade..0ce35fbff4 100644
--- a/data/CVE/list
+++ b/data/CVE/list
@@ -347437,16 +347437,15 @@ CVE-2019-12216 (An issue was discovered in
libSDL2.a in Simple DirectMedia Layer
CVE-2019-12215 (A full path disclosure vulnerability was discovered
in Matomo v3.9.1 w ...)
- matomo <itp> (bug #448532)
CVE-2019-12214 (In FreeImage 3.18.0, an out-of-bounds access occurs
because of mishand ...)
- - freeimage <unfixed> (bug #947478)
- [bookworm] - freeimage <postponed> (Revisit when upstream
fixes are available)
- [bullseye] - freeimage <postponed> (Revisit when upstream
fixes are available)
- [buster] - freeimage <postponed> (Revisit when upstream fixes
are available)
- [stretch] - freeimage <postponed> (Revisit when upstream fixes
are available)
- [jessie] - freeimage <postponed> (Revisit when upstream fixes
are available)
+ - freeimage <not-affected> (bug #947478)
+ - openjpeg2 2.1.0-1
NOTE: https://sourceforge.net/p/freeimage/discussion/36111/thread/e06734bed5/
NOTE: very few information regarding this vulnerability, which
is seemingly located
NOTE: in libopenjpeg, not freeimage. Without reproducer or
stacktrace, this is
NOTE: nearly unfixable.
+ NOTE: Turned out that the issue is not in freeimage at all,
but rather in openjpeg.
+ NOTE: For more information see
https://lists.debian.org/debian-lts/2024/04/msg00058.html
+ NOTE: and more specifically
https://lists.debian.org/debian-lts/2024/04/msg00081.html
CVE-2019-12213 (When FreeImage 3.18.0 reads a special TIFF file, the
TIFFReadDirectory ...)
{DSA-4593-1 DLA-2031-1}
- freeimage 3.18.0+ds2-3 (bug #929597)
// Ola
On Mon, 15 Apr 2024 at 15:30, Santiago Ruano Rincón
<santiagorr@riseup.net> wrote:
>
> Hi,
>
> Cyrille, thank you for checking this. However, I don't think the contact
> address you had sent the email is correct.
> CVE is maintained by MITRE (not NIST). And there exist several CNAs that
> could issue CVE IDs for specific products/domains.
> According to https://www.cve.org/CVERecord?id=CVE-2019-12214, that CVE
> was assigned by MITRE Corporation.
>
> According to https://cve.mitre.org/ (or https://www.cve.org/), the
> correct way to request an update in a CVE entry assigned by MITRE is
> filling out the form that you can find at: https://cveform.mitre.org/,
> choosing the appropriate request type.
>
> Cyrille, would you mind submitting your update to MITRE instead?
>
>
> Ola, the changes you have made to the security-tracker have been
> reverted by Salvatore. See dd2656be1f868274d60b1f38aa7a884e3c8123f2.
> Please, let us know if would you like to propose a proper update. I
> think it is worth to mention the finding in #947478.
>
> Thank you,
>
> El 14/04/24 a las 13:39, Ola Lundqvist escribió:
> > Hi Cyrille
> >
> > Thank you very much.
> >
> > I'll update the security tracker accordingly.
> >
> > // Ola
> >
> > On Sun, 14 Apr 2024 at 12:24, Cyrille Bollu <cyrille@bollu.be> wrote:
> >
> > > Hi,
> > >
> > > I've performed a more thoroughful investigation and have informed NIST
> > > that the offending line is actually to be found in openjpeg between
> > > version 2.0.0 up to (excluding) 2.1.0.
> > >
> > > Debian Buster isn't affected as it uses version 2.3.0-2+deb10u2.
> > >
> > > Hereunder copy of the email I've sent ot NIST.
> > >
> > > Best regards,
> > >
> > > Cyrille
> > >
> > > >Message-ID: <981f8fc77d9e0fee8399a19e6e4c9c64ceeea9a7.camel@bollu.be>
> > > >Subject: CVE-2019-12214: missing vulnerable configuration
> > > >From: Cyrille Bollu <cyrille@bollu.be>
> > > >To: cpe_dictionary@nist.gov
> > > >Date: Sun, 14 Apr 2024 12:01:43 +0200
> > > >Content-Type: text/plain; charset="UTF-8"
> > > >Content-Transfer-Encoding: quoted-printable
> > > >User-Agent: Evolution 3.46.4-2
> > > >MIME-Version: 1.0
> > > >X-Evolution-Identity: 953def08ae37ee7006cd76b472f065ecb205f7e1
> > > >X-Evolution-Fcc:
> > > >folder://d19e895bfc6f11c136a14747fb40c471b2a393e7/Sent
> > > >X-Evolution-Transport: 80f305883d50f910e4b81fcb40b6c46360542068
> > > >X-Evolution-Source:
> > > >
> > > >Dear NIST,
> > > >
> > > >As part of an investigation performed on-behalf of Debian-LTS team,
> > > >I've found out that CVE-2019-12214 is actualy located in code from the
> > > >openjpeg project (https://github.com/uclouvain/openjpeg) which
> > > >freeimage copied in its source tree.
> > > >
> > > >The offending line, "memcpy(l_cp->ppm_data_current, p_header_data,
> > > >l_N_ppm);", has been introduced in version 2.0.0 (see
> > > >
> > > https://github.com/uclouvain/openjpeg/archive/refs/tags/version.2.0.tar.gz
> > > )
> > > >and removed in version 2.1.1 (see
> > > >https://github.com/uclouvain/openjpeg/archive/refs/tags/v2.1.1.tar.gz)
> > > .
> > > >
> > > >So, all intermediatory versions (version 2.0.0 included) might be
> > > >vulnerables (I haven't investigated more than just the presence of
> > > >absence of this line though).
> > > >
> > > >I think it's worth updating CVE-2019-12214 with this information.
> > > >
> > > >Best regards,
> > > >
> > > >Cyrille Bollu
> > >
> > > Le samedi 13 avril 2024 à 09:56 +0200, Cyrille a écrit :
> > > > I don’t know anything about your procedures, but I don’t see why we
> > > > wouldn’t…
> > > >
> > > > I would also contact NIST (or whoever is in charge of the CVE
> > > > database; I can’t remember by heart who it is) to let them know this,
> > > > so they update the CVE’s vulnerable configurations. I’ll try to do
> > > > that next week, but I will probably first have to find out which
> > > > exact versions of openjpeg2 have been affected (which will probably
> > > > be quite difficult for me)
> > > >
> > > > Nice week-end
> > > >
> > > > Cyrille
> > > >
> > > > > Le 13 avr. 2024 à 00:22, Ola Lundqvist <ola@inguza.com> a écrit :
> > > > >
> > > > > Hi Cyrille
> > > > >
> > > > > > On Fri, 12 Apr 2024 at 16:32, Cyrille Bollu <cyrille@bollu.be>
> > > > > > wrote:
> > > > > >
> > > > > > Hi Ola,
> > > > > >
> > > > > > Thank you for your help.
> > > > > >
> > > > > > So, IIUC:
> > > > > >
> > > > > > 1. CVE-2019-12214 shouldn't be assigned to freeimage in Debian
> > > > > > Buster;
> > > > > > 2. CVE-2019-12214 might be assigned to source package openjpeg2
> > > > > > or
> > > > > > openjpeg (the later doesn't seem to be available in Buster
> > > > > > though)
> > > > >
> > > > > Yes, potentially so. At least if I understand the email from
> > > > > Santiago correctly.
> > > > >
> > > > > freeimage build depends on libopenjp2-7-dev which is built from
> > > > > openjpeg2 so in buster it is openjpeg2 where it should belong.
> > > > >
> > > > > But I do not know whether we typically re-assign things like this
> > > > > or
> > > > > not so I do not want to give advice for this. Better if someone
> > > > > else
> > > > > who knows the practice answers this.
> > > > >
> > > > > // Ola
> > > > >
> > > > > --
> > > > > --- Inguza Technology AB --- MSc in Information Technology ----
> > > > > > ola@inguza.com opal@debian.org |
> > > > > > http://inguza.com/ Mobile: +46 (0)70-332 1551 |
> > > > > ---------------------------------------------------------------
> > > >
> > >
> >
> >
> > --
> > --- Inguza Technology AB --- MSc in Information Technology ----
> > | ola@inguza.com opal@debian.org |
> > | http://inguza.com/ Mobile: +46 (0)70-332 1551 |
> > ---------------------------------------------------------------
--
--- Inguza Technology AB --- MSc in Information Technology ----
| ola@inguza.com opal@debian.org |
| http://inguza.com/ Mobile: +46 (0)70-332 1551 |
---------------------------------------------------------------
Reply to: