Joey Hess
2006-Mar-13 12:28 UTC
[Secure-testing-team] summary of what''s blocking security fixes from testing
Another pass over security holes that are fixed in unstable but not testing. Not sure if these are still useful to send to -release. RM summary: m68k is killing us with ICE after ICE and contributing to blocking hald of the fixes. The transitions arn''t hurting as much after the last heroic britney run, although kde/qt is of course a problem. Testing team summary: well, of these asterisk, inkscape, some kde stuff, lm-sensors, mysql-dfsg-4.1, and texmacs seem like the most likely candidates for upload to secure-testing, although some of the holes may not warrant a DTSA. apache m68k build needs requeued once deps are met asterisk 30 days old m68k build needs requeued once deps are met blocked indirectly by qt transition bzip 8/10 days old chmlib 3/10 days old clamav (fixed in secure-testing) 33 days old blocked by gmp courier too young fftw3 m68k FTBFS gxine too young inkscape 20 days old blocked by libsigc++-2.0 kdeedu FTBFS on arm (ICE) missing hppa and m68k builds kdegraphics kde transtion kdelibs kde transition kismet (fixed in secure-testing) 23 days old blocked by gmp lm-sensors 23 days old indirectly blocked by perl mozilla (partially fixed in secure-testing) 41 days old, AKA, is this package being maintained? rc bugs, FTBDS, etc mozilla-firefox (partially fixed in secure-testing) too young mozilla-thunderbird 41 days old FTBFS on alpha, arm, m68k mysql-dfsg-4.1 26 days old rc bug FTBFS on m68k net-snmp too young blocked by perl netpbm 8/10 days old FTBFS on m68k (ICE) ntp 177 days old 3 RC bugs, max 98 days old, none with responses from maintainers recommend removal from testing (and/or debian) openmotif 106 days old non-free package, still missing s390 build (I tried and failed to build this on raptor, machine is too unstable.) openssh frozen, rc bug security hole is minor (CAN-2005-2666) qt-x11-free too young rc bug pdns FTBFS on m68k rc bugs php4 (fixed in secure-testing) needs requeue on m68k once deps are satisfied blocked by qt-x11-free, perl, etc python2.1 alpha build succeeded 2 weeks ago but gone missing mips, mipsel, powerpc builds ditto blocked by gmp python2.2 FTBFS m68k (ICE) FTBFS hppa blocked by gmp python2.3 FTBFS m68k (ICE) FTBFS hppa blocked by gmp rpm FTBFS m68k (ICE) smb4k hppa build needs requeue once build deps are met m68k ditto arm ditto kde transition squid too young sqwebmail too young texmacs 59 days old FTBFS arm, hppa, m68k (ICE) turqstat too young blocked by qt-x11-free xorg-x11 too young build needs retried on arm -- see shy jo -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://lists.alioth.debian.org/pipermail/secure-testing-team/attachments/20050913/cbc01822/attachment.pgp
Steve Langasek
2006-Mar-13 12:28 UTC
[Secure-testing-team] Re: summary of what''s blocking security fixes from testing
On Tue, Sep 13, 2005 at 11:45:52PM -0400, Joey Hess wrote:> Another pass over security holes that are fixed in unstable but not > testing. Not sure if these are still useful to send to -release.Yes, I think they are.> Testing team summary: well, of these asterisk, inkscape, some kde stuff, > lm-sensors, mysql-dfsg-4.1, and texmacs seem like the most likely > candidates for upload to secure-testing, although some of the holes may > not warrant a DTSA.FWIW, inkscape is in the libsigc++-2.0/libgc transition that''s currently at the top of my hit list.> apache > m68k build needs requeued once deps are met > fftw3 > m68k FTBFS > netpbm > 8/10 days old > FTBFS on m68k (ICE) > rpm > FTBFS m68k (ICE)I''m forcing these in spite of the lack of m68k builds. Between ICEs and general sluggishness, m68k is not keeping up. I know the m68k porters are talking about putting new buildds on-line, but there are also a lot of KDE uploads coming that are going to bog it down further, and lots of m68k-specific toolchain problems that still need to be fixed. If we don''t see improvement soon, I think the necessary next step is to ignore it for all packages (i.e., exclude it from the list of release candidates for the time being).> mysql-dfsg-4.1 > 26 days old > rc bug > FTBFS on m68kAnd no build log for the m68k failure to let people usefully debug it... The RC bug was apparently meant to be downgraded, and the maintainer missed. Downgrading now, and forcing in without m68k.> bzip > 8/10 days old > chmlib > 3/10 days old > courier > too young > gxine > too young > squid > too young > sqwebmail > too youngFeel free to add urgent hints for any of these.> clamav (fixed in secure-testing) > 33 days old > blocked by gmp > kismet (fixed in secure-testing) > 23 days old > blocked by gmpThis mainly means "blocked by kaffe", I think.> kdeedu > FTBFS on arm (ICE) > missing hppa and m68k buildsThose will almost certainly be all the same ICE, actually.> lm-sensors > 23 days old > indirectly blocked by perl > net-snmp > too young > blocked by perl<grumble>> mozilla (partially fixed in secure-testing) > 41 days old, AKA, is this package being maintained? > rc bugs, FTBDS, etcI''ll NMU this if no one else does, but it''ll probably take me a day or two to get to it.> mozilla-firefox (partially fixed in secure-testing) > too youngMore that arm hasn''t finished building it yet.> mozilla-thunderbird > 41 days old > FTBFS on alpha, arm, m68k... with a patch in the BTS, if someone wants to NMU...> ntp > 177 days old > 3 RC bugs, max 98 days old, none with responses from maintainers > recommend removal from testing (and/or debian)Are these different security bugs than the ones already fixed via proposed-updates?> openmotif > 106 days old > non-free package, still missing s390 build > (I tried and failed to build this on raptor, machine is too > unstable.)This package really doesn''t appear to have the necessary baseline support from porters and/or the maintainer to let us keep it around. There''s a total of one package in testing still depending on openmotif; I think we should give the arb maintainer a shot at fixing it, and then drop it from testing if he doesn''t get anywhere.> openssh > frozen, rc bug > security hole is minor (CAN-2005-2666)Pushed in. (The RC bug was reported against the version in stable, and should not be a blocker.)> php4 (fixed in secure-testing) > needs requeue on m68k once deps are satisfiedAlready in dep-wait. The version in unstable is stuck for a while, but the sarge security update is waiting in t-p-u for m68k to catch up.> python2.1 > alpha build succeeded 2 weeks ago but gone missing > mips, mipsel, powerpc builds ditto > blocked by gmp > python2.2 > FTBFS m68k (ICE) > FTBFS hppa > blocked by gmpNo hope that we can get rid of these yet...?> xorg-x11 > too young > build needs retried on armCurrently listed as building on tofee. Thanks, -- Steve Langasek Give me a lever long enough and a Free OS Debian Developer to set it on, and I can move the world. vorlon@debian.org http://www.debian.org/ -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://lists.alioth.debian.org/pipermail/secure-testing-team/attachments/20050913/dc2b7a49/attachment.pgp
Joey Hess
2006-Mar-13 12:28 UTC
[Secure-testing-team] Re: summary of what''s blocking security fixes from testing
Steve Langasek wrote:> > bzip > > 8/10 days old > > chmlib > > 3/10 days old > > courier > > too young > > gxine > > too young > > squid > > too young > > sqwebmail > > too young > > Feel free to add urgent hints for any of these.None of them are that serious holes, IRC.> > ntp > > 177 days old > > 3 RC bugs, max 98 days old, none with responses from maintainers > > recommend removal from testing (and/or debian) > > Are these different security bugs than the ones already fixed via > proposed-updates?Thanks, you''re right it''s not vulnerable.> > openmotif > > 106 days old > > non-free package, still missing s390 build > > (I tried and failed to build this on raptor, machine is too > > unstable.) > > This package really doesn''t appear to have the necessary baseline > support from porters and/or the maintainer to let us keep it around. > There''s a total of one package in testing still depending on openmotif; > I think we should give the arb maintainer a shot at fixing it, and then > drop it from testing if he doesn''t get anywhere.Agreed, although I think you might as well let the maintainers of ida and motv have a crack at it too.> > python2.1 > > alpha build succeeded 2 weeks ago but gone missing > > mips, mipsel, powerpc builds ditto > > blocked by gmp > > python2.2 > > FTBFS m68k (ICE) > > FTBFS hppa > > blocked by gmp > > No hope that we can get rid of these yet...?Very little depends on python2.1. python2.2 has a bit more stuff but certianly not too much that filing some RC bugs and dropping it might not be the easiest way to fix this issue. -- see shy jo -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://lists.alioth.debian.org/pipermail/secure-testing-team/attachments/20050914/43baeaf3/attachment.pgp
Moritz Muehlenhoff
2006-Mar-13 12:28 UTC
[Secure-testing-team] Re: summary of what''s blocking security fixes from testing
Steve Langasek wrote:> > mozilla-thunderbird > > 41 days old > > FTBFS on alpha, arm, m68k > > ... with a patch in the BTS, if someone wants to NMU...The maintainer posted a followup to the firefox side of the latest Mozilla IDN buffer overflow, so I very much suspect he''s currently preparing fixed thunderbird packages.> > ntp > > 177 days old > > 3 RC bugs, max 98 days old, none with responses from maintainers > > recommend removal from testing (and/or debian) > > Are these different security bugs than the ones already fixed via > proposed-updates?No, it the same vulnerability as already fixed in t-p-u. Cheers, Moritz
Moritz Muehlenhoff
2006-Mar-13 12:28 UTC
[Secure-testing-team] summary of what''s blocking security fixes from testing
Joey Hess wrote:> kdeedu > FTBFS on arm (ICE) > missing hppa and m68k builds > kdegraphics > kde transtionI had a look at these, but it''s worth a fix, xpdf has been more critical, because someone could indirectly mess with the print spooler, but kpdf as a simple app doesn''t warrant a fix IMO. And the kdeedu issue is really minor.> kdelibs > kde transitionI have fixed packages ready, I''ll send them to you this evening, after some more testing. kdebase will be fixed next.> mozilla (partially fixed in secure-testing) > 41 days old, AKA, is this package being maintained? > rc bugs, FTBDS, etc633 open bugs, 10 of them RC. Should probably be team-maintained.> mysql-dfsg-4.1 > 26 days old > rc bug > FTBFS on m68kThis (and mysql-dfsg) are perfect candidates for DTSAs/NMUs, the fix is a simple one-liner.> ntp > 177 days old > 3 RC bugs, max 98 days old, none with responses from maintainers > recommend removal from testing (and/or debian)Fix is a one-liner as well.> squid > too youngCould very well be bumped, it only contains the (IMO non-invasive) security fixes and typo fixes. Cheers, Moritz
Mark Brown
2006-Mar-13 12:28 UTC
[Secure-testing-team] Re: summary of what''s blocking security fixes from testing
On Tue, Sep 13, 2005 at 11:45:52PM -0400, Joey Hess wrote:> Another pass over security holes that are fixed in unstable but not > testing. Not sure if these are still useful to send to -release.There''s also zlib, which is held up in both testing-proposed-updates and unstable by missing builds. -- "You grabbed my hand and we fell into it, like a daydream - or a fever."
Matthias Klose
2006-Mar-13 12:28 UTC
[Secure-testing-team] Re: summary of what''s blocking security fixes from testing
Joey Hess writes:> RM summary: m68k is killing us with ICE after ICE and contributing to > blocking hald of the fixes. The transitions arn''t hurting as much after > the last heroic britney run, although kde/qt is of course a problem.updated gcc-4.0 packages, which should fix some of the ICEs are uploaded today. Currently bootstrapping, it may take a while. A bigger problem seems to be the new binutils on m68k.> python2.2 > FTBFS m68k (ICE) > FTBFS hppa > blocked by gmpneeds a new upload.> python2.3 > FTBFS m68k (ICE) > FTBFS hppa > blocked by gmphppa is a glibc problem.
Adeodato Simó
2006-Mar-13 12:28 UTC
[Secure-testing-team] Re: summary of what''s blocking security fixes from testing
* Joey Hess [Tue, 13 Sep 2005 23:45:52 -0400]:> kdegraphics > kde transtion4:3.3.2-2sarge1 was propagated to t-p-u, but not accepted because of #325254 (uninstallable on powerpc). Anyway, 4:3.3.2-3 is now in t-p-u and waiting for buildds (haven''t seen any, though).> kdelibs > kde transition4:3.3.2-6.2 seems to be in NEW? And is blocking kdegraphics below.> qt-x11-free > too young > rc bugIs this listed here for some security issue, or because it''s blocking some stuff? It''ll be a long while before it can enter testing, though. -- Adeodato Sim? EM: asp16 [ykwim] alu.ua.es | PK: DA6AE621 Loan-department manager: "There isn''t any fine print. At these interest rates, we don''t need it."
Steve Langasek
2006-Mar-13 12:28 UTC
[Secure-testing-team] Re: summary of what''s blocking security fixes from testing
On Wed, Sep 14, 2005 at 11:05:08AM +0200, Moritz Muehlenhoff wrote:> > > ntp > > > 177 days old > > > 3 RC bugs, max 98 days old, none with responses from maintainers > > > recommend removal from testing (and/or debian)> > Are these different security bugs than the ones already fixed via > > proposed-updates?> No, it the same vulnerability as already fixed in t-p-u.Right, that version''s already present in testing then. -- Steve Langasek Give me a lever long enough and a Free OS Debian Developer to set it on, and I can move the world. vorlon@debian.org http://www.debian.org/ -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://lists.alioth.debian.org/pipermail/secure-testing-team/attachments/20050914/02c64d96/attachment.pgp
Frank Lichtenheld
2006-Mar-13 12:28 UTC
[Secure-testing-team] Re: summary of what''s blocking security fixes from testing
On Mon, Sep 19, 2005 at 03:16:41AM -0700, Steve Langasek wrote:> On Wed, Sep 14, 2005 at 10:00:05AM -0400, Joey Hess wrote: > > Agreed, although I think you might as well let the maintainers of ida > > and motv have a crack at it too. > > So Frank has already NMUed this with a proposed fix for the s390 failure. > Binaries only uploaded for i386 and m68k so far, though.I''ve added it to the list of non-free buildds. But they will only pick it up after dinstall since they don''t build out of incoming... Gruesse, -- Frank Lichtenheld <djpig@debian.org> www: http://www.djpig.de/
Matthias Klose
2006-Mar-13 12:28 UTC
[Secure-testing-team] Re: summary of what''s blocking security fixes from testing
Joey Hess writes:> > > python2.1 > > > alpha build succeeded 2 weeks ago but gone missing > > > mips, mipsel, powerpc builds ditto > > > blocked by gmp > > > python2.2 > > > FTBFS m68k (ICE) > > > FTBFS hppa > > > blocked by gmp > > > > No hope that we can get rid of these yet...? > > Very little depends on python2.1. python2.2 has a bit more stuff but > certianly not too much that filing some RC bugs and dropping it might > not be the easiest way to fix this issue.these should slowly fade away. Just waiting to clear up the transition business to start this one. Matthias
Andreas Barth
2006-Mar-13 12:28 UTC
[Secure-testing-team] Re: summary of what''s blocking security fixes from testing
* Joey Hess (joeyh@debian.org) [050914 23:12]:> Adeodato Sim? wrote: > > > kdelibs > > > kde transition > > > > 4:3.3.2-6.2 seems to be in NEW? And is blocking kdegraphics below. > > Moritz Muehlenhoff wrote (privately): > > But according to http://ftp-master.debian.org/new.html it is in > > t-p-u? And it''s not shown in the PTS either? > > And why would it need to go through NEW, there were not binary > > packages added?> I share this confusion..Uh, that''s strongly technical. Basically it''s a corner case of the version propogation patches that doesn''t work right. Please see http://lists.debian.org/debian-dak/2005/09/msg00000.html for a possible fix. Cheers, Andi
Moritz Muehlenhoff
2006-Mar-13 12:28 UTC
[Secure-testing-team] Re: summary of what''s blocking security fixes from testing
Andreas Barth wrote:> > > 4:3.3.2-6.2 seems to be in NEW? And is blocking kdegraphics below. > > > > Moritz Muehlenhoff wrote (privately): > > > But according to http://ftp-master.debian.org/new.html it is in > > > t-p-u? And it''s not shown in the PTS either? > > > And why would it need to go through NEW, there were not binary > > > packages added? > > > I share this confusion.. > > Uh, that''s strongly technical. Basically it''s a corner case of the > version propogation patches that doesn''t work right. Please see > http://lists.debian.org/debian-dak/2005/09/msg00000.html for a possible > fix.Has the dak patch mentioned in above link been applied or is this more a mid-term solution (i.e. preparing a fix through a DTSA would be better for now?) Cheers, Moritz
Joey Hess
2006-Mar-13 12:28 UTC
[Secure-testing-team] Re: summary of what''s blocking security fixes from testing
Adeodato Sim? wrote:> > kdelibs > > kde transition > > 4:3.3.2-6.2 seems to be in NEW? And is blocking kdegraphics below.Moritz Muehlenhoff wrote (privately):> But according to http://ftp-master.debian.org/new.html it is in > t-p-u? And it''s not shown in the PTS either? > And why would it need to go through NEW, there were not binary > packages added?I share this confusion..> > qt-x11-free > > too young > > rc bug > > Is this listed here for some security issue, or because it''s blocking > some stuff? It''ll be a long while before it can enter testing, though.Not sure how I contrived to list that as a security issue, it''s only blocking other security fixes. -- see shy jo -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://lists.alioth.debian.org/pipermail/secure-testing-team/attachments/20050914/83916cff/attachment.pgp
Steve Langasek
2006-Mar-13 12:28 UTC
[Secure-testing-team] Re: summary of what''s blocking security fixes from testing
On Wed, Sep 14, 2005 at 10:00:05AM -0400, Joey Hess wrote:> > Feel free to add urgent hints for any of these.> None of them are that serious holes, IRC.Hmm, if they''re not serious enough to bump package urgency for, maybe it''s also not useful to have that part of the list posted to d-release. :)> > > openmotif > > > 106 days old > > > non-free package, still missing s390 build > > > (I tried and failed to build this on raptor, machine is too > > > unstable.)> > This package really doesn''t appear to have the necessary baseline > > support from porters and/or the maintainer to let us keep it around. > > There''s a total of one package in testing still depending on openmotif; > > I think we should give the arb maintainer a shot at fixing it, and then > > drop it from testing if he doesn''t get anywhere.> Agreed, although I think you might as well let the maintainers of ida > and motv have a crack at it too.So Frank has already NMUed this with a proposed fix for the s390 failure. Binaries only uploaded for i386 and m68k so far, though. -- Steve Langasek Give me a lever long enough and a Free OS Debian Developer to set it on, and I can move the world. vorlon@debian.org http://www.debian.org/ -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://lists.alioth.debian.org/pipermail/secure-testing-team/attachments/20050919/14d26c5a/attachment.pgp