dbus system bus switching to non-permissive soon for F11
by Colin Walters
Hi,
I plan to switch the DBus system bus away from the "permissive"
release to the one which closes
https://bugs.freedesktop.org/show_bug.cgi?id=18229
relatively soon in the Fedora 11 tree.
I've been going through bugzillas and getting policies upstream for a
while, and I think the vast majority of things are fixed now. The
permissive mode has been in rawhide for some time, and will have been
generating messages of the form:
Feb 2 12:17:38 space-ghost dbus: Rejected send message, 20
matched rules; type="method_return", sender=":1.19" (uid=0 pid=2863
comm="/usr/sbin/wpa_supplicant -c /etc/wpa_supplicant/wp")
interface="(unset)" member="(unset)" error name="(unset)"
requested_reply=0 destination=":1.18" (uid=0 pid=2852
comm="NetworkManager --pid-file=/var/run/NetworkManager/"))
in syslog. If you're seeing these and I didn't find your bugzilla,
and don't know how to fix, please ask for help at
dbus(a)lists.freedesktop.org or #dbus on Freenode.
15 years, 2 months
Re: rpms/dbus/F-9 dbus-alpha-unaligned.patch, NONE, 1.1 dbus.spec, 1.150, 1.151
by Colin Walters
On Thu, Feb 26, 2009 at 4:25 AM, Oliver Falk <oliver(a)fedoraproject.org> wrote:
> Author: oliver
>
> Update of /cvs/pkgs/rpms/dbus/F-9
> In directory cvs1.fedora.phx.redhat.com:/tmp/cvs-serv24148
[...]
> URL: http://www.freedesktop.org/software/dbus/
> Source0: http://dbus.freedesktop.org/releases/dbus/%{name}-%{version}.tar.gz
> Source1: doxygen_to_devhelp.xsl
> @@ -41,6 +41,7 @@
> Patch0: start-early.patch
> Patch1: dbus-1.0.1-generate-xml-docs.patch
> Patch6: dbus-1.2.1-increase-timeout.patch
> +Patch7: dbus-alpha-unaligned.patch
Public reminder - include a comment link to the upstream bug tracker
entries above any patches you add.
In this case, that would be:
# http://bugs.freedesktop.org/show_bug.cgi?id=20137
https://fedoraproject.org/wiki/Packaging:Guidelines#All_patches_should_ha...
15 years, 2 months
Disable OSS sound by default in Fedora 11, compat-broken-oss-sound
by Warren Togami
https://bugzilla.redhat.com/show_bug.cgi?id=472741#c12
Here is an improved plan to disable OSS by default in Fedora 11. This
plan makes it possible to restore the old behavior with a single compat
package. This will prevent the pulseaudio/OSS conflict that currently
causes weird behavior and mistaken bug reports filed against numerous
other packages.
This should also encourage us to fix the few remaining applications that
output OSS by default.
Proposal
========
1) Keep the OSS kernel modules.
2) Remove all snd-oss* lines from /etc/modprobe.d/modprobe.conf.dist.
3) (We might want to put blacklist snd-oss* lines into
alsa-plugins-pulseaudio? Will this be needed or helpful at all?)
4) Put the lines removed in #2 into a new package called
compat-broken-oss-sound. This package conflicts with
alsa-plugins-pulseaudio.
5) Release notes say OSS is DISABLED BY DEFAULT. All applications
should switch to ALSA or pulseaudio output. As a backup measure,
OSS-only applications can use padsp to emulate an OSS device, which
works for most applications.
If we are going to do this, we should do it soon before F11 Beta.
FAQ
===
1) Why is it called compat-broken-oss-sound?
OSS causes broken behavior in combination with pulseaudio.
OSS completely blocks sound from other applications.
We want the package name to suggest there is something wrong if you need
to use it.
2) Why can't we simply blacklist?
It doesn't work due to the install lines. The install lines need to be
removed.
3) How does OSS conflict with pulseaudio?
OSS grabs the entire sound device, causing pulseaudio to get stuck.
Applications attempting to output via pulseaudio themselves get stuck
and sometimes crash. This behavior is very non-obvious and often
pulseaudio or the application is blamed instead of OSS.
Warren Togami
wtogami(a)redhat.com
15 years, 2 months
RE: Ready for new RPM version?
by Shawn Starr
> -----Original Message-----
> From: fedora-devel-list-bounces(a)redhat.com
> [mailto:fedora-devel-list-bounces@redhat.com]On Behalf Of Panu
> Matilainen
> Sent: Thursday, February 26, 2009 12:36 PM
> To: fedora-devel-list(a)redhat.com
> Subject: Ready for new RPM version?
>
>
>
> Are we ready to consider a brand new RPM version for F11,
> amidst all this
> mass-rebuild-for-strong-hash chaos and just days to go to development
> freeze?
>
> We just put out first rpm 4.7.0 beta:
> http://lists.rpm.org/pipermail/rpm-announce/2009-February/0000
16.html.
>This is nothing like the 4.4.2.x -> 4.6.0 leap-of-faith upgrade last year
>but it's still non-trivial amount of changes to get the kind of memory use
>and performance improvements that 4.7.0 has, which is why it's not just
>4.6.1.
Do you have RPM/.spec available? I'd like to use it regardless if Fedora 11 has it or not.
Thanks,
Shawn.
15 years, 2 months
Mass Rebuild Status
by Jesse Keating
We're now roughly 4 days into the mass rebuild. Things are going very
smoothly. Some interesting numbers:
Roughly 5650 submissions have been made to the build system thus far.
Roughly 4640 builds have completed in dist-f11-rebuild.
Of those, 4585 would be tagged into dist-f11 if we started tagging
today.
Checking the builds in both dist-f11 and dist-f11-rebuild for things
that have been built after the run started shows 2201 builds still
needing to be done.
There have been 505 build failures (these will be posted later broken
down by maintainer), although some of these were due to buildsystem
errors and have been resubmitted.
There are so far 68 packages that failed in some way or another prior to
the build submission that have to be manually looked at.
All in all it has been quite smooth, and I'm very impressed at how well
the buildsystem has handled the onslaught of builds. At first, ppc
builders were the bottle neck, but the Infrastructure rock stars were
able to bring a couple more PPC builders online. After that the CVS
server quickly became the bottleneck, slowing down my ability to submit
packages faster than the buildsystem could handle them. At that point I
broke my script apart and had multiple threads submitting builds at the
same time which enabled me to fill the build queue once again. We're
down to a final thread, working its way through the p packages (into
perl-H* now). If the buildsystem catches up, I'll break that further
down into threads.
I expect the builds to finish by Friday or Saturday of this week, and
then I'll start the tagging. The rawhide report for the next day will
be late, and probably too big to post to the mailing list. Rest
assured, every package changed (:
--
Jesse Keating
Fedora -- Freedom² is a feature!
identi.ca: http://identi.ca/jkeating
_______________________________________________
Fedora-devel-announce mailing list
Fedora-devel-announce(a)redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-announce
15 years, 2 months
Re: comps comps-f11.xml.in,1.125,1.126
by Nils Philippsen
This time with the fedoraproject.org address, sorry for the spam (I'll
give up if this one doesn't get through).
On Thu, 2009-02-26 at 10:40 -0500, Bill Nottingham wrote:
> Nils Philippsen (nphilipp(a)fedoraproject.org) said:
> > Author: nphilipp
> >
> > Update of /cvs/pkgs/comps
> > In directory cvs1.fedora.phx.redhat.com:/tmp/cvs-serv5807
> >
> > Modified Files:
> > comps-f11.xml.in
> > Log Message:
> > add "online-docs" group with system-config-*-docs packages that get installed
> > dependent on the respective tool
>
> Why wouldn't these just be Required: by the tools themselves?
This is so you can install the tools without online documentation e.g.
to satisfy space constraints on live media.
Nils
--
Nils Philippsen "Those who would give up Essential Liberty to purchase
Red Hat a little Temporary Safety, deserve neither Liberty
nils(a)redhat.com nor Safety." -- Benjamin Franklin, 1759
PGP fingerprint: C4A8 9474 5C4C ADE3 2B8F 656D 47D8 9B65 6951 3011
15 years, 2 months
rawhide mock problem
by Neal Becker
I'm trying to see why igraph-0.5.1 fails to build on rawhide. On my F10
box, I did:
mock -r fedora-rawhide-x86_64 --init
mock -r fedora-rawhide-x86_64 igraph-0.5.1-4.fc11.src.rpm
...
ERROR: Command failed:
# /usr/bin/yum --installroot /var/lib/mock/fedora-rawhide-x86_64/root/
groupinstall buildsys-build
Error unpacking rpm package kernel-headers-2.6.29-0.157.rc6.git2.fc11.x86_64
error: unpacking of archive failed on file /usr/include/asm-generic/errno-
base.h;49a6b30f: cpio: MD5 sum mismatch
Error unpacking rpm package bash-4.0-2.fc11.x86_64
error: unpacking of archive failed on file /bin/bash;49a6b30f: cpio: MD5 sum
mismatch
warning: %post(info-4.13a-1.fc11.x86_64) scriptlet failed, exit status 255
warning: %post(libsepol-2.0.35-1.fc11.x86_64) scriptlet failed, exit status
255
warning: %post(libselinux-2.0.77-5.fc11.x86_64) scriptlet failed, exit
status 255
Error unpacking rpm package popt-1.13-5.fc11.x86_64
error: unpacking of archive failed on file /lib64/libpopt.so.0.0.0;49a6b30f:
cpio: MD5 sum mismatch
Error unpacking rpm package audit-libs-1.7.12-1.fc11.x86_64
error: unpacking of archive failed on file /etc/libaudit.conf;49a6b30f: cpio:
MD5 sum mismatch
warning: %post(sed-4.1.5-11.fc11.x86_64) scriptlet failed, exit status 255
warning: %post(readline-5.2-13.fc9.x86_64) scriptlet failed, exit status 255
warning: %post(grep-2.5.3-3.fc11.x86_64) scriptlet failed, exit status 255
warning: %post(libidn-0.6.14-9.x86_64) scriptlet failed, exit status 255
warning: %post(nspr-4.7.3-3.fc11.x86_64) scriptlet failed, exit status 255
warning: %post(nss-3.12.2.0-4.fc10.x86_64) scriptlet failed, exit status 255
warning: %post(findutils-1:4.4.0-1.fc10.x86_64) scriptlet failed, exit status
255
warning: %post(binutils-2.19.51.0.2-12.fc11.x86_64) scriptlet failed, exit
status 255
warning: %post(cpio-2.9.90-3.fc11.x86_64) scriptlet failed, exit status 255
warning: %post(diffutils-2.8.1-22.fc11.x86_64) scriptlet failed, exit status
255
warning: %post(cpp-4.4.0-0.21.x86_64) scriptlet failed, exit status 255
...
lots more
15 years, 2 months
Features/ArchitectureSupport - changing what we build for
by Bill Nottingham
Kevin Fenzi (kevin(a)scrye.com) said, in the FESCo meeting summary:
> * Architecture Support
> https://fedoraproject.org/wiki/Features/ArchitectureSupport
>
> FESCo tabled this to revisit this next week. There were questions
> about OLPC support and LSTP client machines. Discussion to continue
> on fedora-devel list.
As a followup to the discussion on compiler flags, this feature was
written up to describe a plan for what to do overall about what
architectures we support. The main points are:
- install x86_64 kernel on 32-bit OS where appropriate
- install PAE kernel on other 32-bit OS installs where appropriate
- build only i686 and above for Fedora
The last one is the issue that caused discussion during the FESCo
meeting. The main points raised:
- XO support
There is concern that building for i686 might break binaries on the XO.
The Geode on the XO self-reports as i586; this may need slight frobbing.
Aside from that, it "supports the i686 (Pentium Pro) instruction set, MMX,
the parts of SSE that do not involve SSE registers, 3DNow! Enhanced, a
couple Geode-specific instructions and a few SSE2 instructions."
(http://wiki.laptop.org/go/Geode_instruction_set)
This implies it should work; however, there have been bugs in the past
(https://bugzilla.redhat.com/show_bug.cgi?id=200330). Chris Ball stated
"my input as OLPC software dude is that we're probably okay with dropping
586, but it would be great if someone heled us test that everything's
looking okay first."
- K12Linux/thin client
Some older thin client hardware is not i686 compatible. Warren has more
info on this. One question I have is that some of this is mentioned
as being Geode - wouldn't this fall into the same category as the XO?
Also, just how much of this hardware is out there using Fedora now
(as oppposed to RHEL, CentOS, or some other distribution?)
More info is on the feature page. Note that in the smolt stats, i586
accounts for less than one twentieth of one percent of active systems,
an order of magnitude less than PowerPC.
Bill
15 years, 2 months
seg fault building glibc on x86_64
by Chris Eveleigh
we appear to be getting a crash when attempting to rpmbuild glibc-2.9-3
on an up to date F10 x86_64 system.
i can build it fine on an up to date F10 i386 system.
this has come up in https://bugzilla.redhat.com/show_bug.cgi?id=473445
can anyone reproduce/fix/suggest?
it seems to crash at different places - for example:
rpmbuild/BUILDROOT/glibc-2.9-3.x86_64/usr/share/zoneinfo/right/Asia/Riyadh89] Segmentation fault (at least twice)
rpmbuild/BUILDROOT/glibc-2.9-3.x86_64/usr/share/zoneinfo/right/Australia/Darwin] Segmentation fault
rpmbuild/BUILD/glibc-20081113T2206/build-x86_64-linuxnptl/sunrpc/xmount.stmp] Segmentation fault
rpmbuild/BUILD/glibc-20081113T2206/build-x86_64-linuxnptl/sunrpc/rpcsvc/sm_inter.stmp] Segmentation fault
i can't really see /what/ is crashing, you just get a line like:
make[2]: ***
[/home/cje/rpmbuild/BUILDROOT/glibc-2.9-3.x86_64/usr/share/zoneinfo/right/Australia/Darwin] Segmentation fault
15 years, 2 months
Intel 4 series chipset not working with f10
by Rodd Clarkson
Right,
I've recently purchased an intel 4 series chipset as part of a G41
motherboard and neither the i810 or the intel driver work, which leaves
me with vesa (at 1280x1024 on a 1680x1050 monitor).
I bought the card because it does hdmi, but it turns out it doesn't even
do intel drivers, so that might be a pipe dream.
Attached at the log files and xorg.conf files after trying both the i810
and the intel drivers.
Any suggestions?
R.
--
MOOSE technology
po box 6061, north croydon, vic 3136 mobile: 0403 338 731
http://www.moosetech.com.au phone: 03 9726 9457
mailto:rodd@moosetech.com.au fax: 03 9726 9456
"Share your knowledge. It is a way to achieve immortality."
-- The Dalai Lama
--
"It's a fine line between denial and faith.
It's much better on my side"
15 years, 2 months