VCS key in spec files and some scripts
by Colin Walters
Dear Fedora Developers,
So recently I found myself desiring to update a .spec file for a
snapshot of a git tree in an automated fashion. As you may or may not
know, this actually has a surprising number of flaming hoops through
which you must jump.
The first one, when scripting such a thing, is pulling the source tree
and packing it into a source file. To this purpose, I propose we
begin adding the following key to .spec files:
#VCS: <url format entry for version control system>
Note the use of a comment. There's a ticket to support this key more
explicitly in RPM:
http://rpm.org/attachment/ticket/143
However I didn't want to block my scripts on that patch landing; in
any case we currently add human-consumable comments
(https://fedoraproject.org/wiki/Packaging:SourceURL) for VCS snapshots
now, so think of this key as just more of that. Having the version
control system in the spec file will help a lot for other things like
automating source tarball verification.
The second hoop is correctly handling the Release tag.
https://fedoraproject.org/wiki/Packaging:NamingGuidelines#Package_Version
has lots of gory details. But basically, this can be automated.
Another hoop is the autotools; you have a variety of choices here, but
it's most common to add BuildRequires on them, and figure out how to
bootstrap. This latter part requires a bit of build intelligence (my
scripts recognize the case of autotools, and the more specific case of
gnome-autogen.sh).
I'd actually like to break out the build recognition system into a
separate python library of some sort.
So without further ado, my current scripts:
http://fedorapeople.org/gitweb?p=walters/public_git/fedpkg-make-pull.git;...
I'd actually like to get these into Jesse's new fedpkg work, but for
now these operate on distcvs.
Executive summary: if you find yourself needing to automate git-spec
integration, you should use my scripts, because they're cool. I'm
sure I'm not the first person to write these scripts, but I'd like to
get these upstream into fedpkg. Patches for other version control
systems (and major build systems like distutils and cmake) are happily
accepted!
What am I using them for? Automating gnome-shell stack snapshots for F12:
$ fedpkg-pull-build-chain --arch=i386 --arch=x86_64 --release=F-12
--resultdir=_repo gobject-introspection gir-repository gjs clutter
mutter gnome-shell
14 years
Proposal: move comps to fedorahosted git
by Bill Nottingham
I'd like to propose moving comps to fedorahosted git.
Why? Because CVS is a pain.
I can work on fixing the automated releng tasks that use comps.
What I'd like to know is if doing this at some point over the
next few weeks (say, post-Alpha) would be a problem for people.
If it is, we can push it off until after F13 ships.
Bill
14 years
FESCo wants to ban direct stable pushes in Bodhi (urgent call for feedback)
by Kevin Kofler
Hi,
at the FESCo meeting on Tuesday, everyone except me seemed to be set on
wanting to disable the possibility to queue updates directly to stable in
Bodhi. The only reason this was not decided right there (with no outside
feedback) is that Matthew Garrett (mjg59) wants to write down a precise
policy (which may end up even more restrictive, like some arbitrary minimum
time period of testing).
He also noted that doing so "gives us an opportunity to discuss various
consequences with affected teams". But sadly, the people driving this
proposed change haven't used this opportunity to discuss this issue in a
transparent way as I would have expected (and I've been waiting for almost 3
days!), so I am doing it now. (We really need more transparency in decision
making!)
I would like to collect feedback on this issue. If you want to disable
direct stable pushes, why? Could there be a less radical solution to that
problem (e.g. a policy discouraging direct stable pushes for some specific
types of changes rather than a blanket ban)? On the other hand, if (like me)
you DON'T want that feature to go away, please provide valid use cases.
Some situations where I and others have used direct stable pushes in the
past and where I think they're really warranted and should be used:
* A new package which doesn't replace anything, and which I verified to work
fine for me. It's clearly not a completely broken package and there's no way
it can break anybody's existing setup as nobody has that package yet.
* A regression which causes big breakage at least for some people slipped
through testing for whatever reason. We urgently want the fix to get out
ASAP.
* A regression slipped through testing for whatever reason and the patch is
trivial. We want the fix to get out ASAP, and the risk of breakage is very
low.
* A trivial bugfix (like a one-line diff), tested and confirmed to fix the
bug by at least one person. The risk of breakage is extremely low.
If you can think of more, please post them! But even if you just agree with
me, please reply so the other FESCo members don't think it's just me!
Kevin Kofler
14 years
Why online recovery in pgpool is disabled?
by Michał Piotrowski
Hi,
Is there any particular reason why online recovery is disabled in F11 pgpool-II?
Online recovery is a very important feature (fundamental, must have)
and I have to build pgpool-II just to enable it. Can't it be enabled
in spec?
Regards,
Michal
14 years
Incompatible upgrade - Is this workaround ok? (mysql-mmm)
by BJ Dierkes
Hello all,
I maintain Multi-Master Replication Manager for MySQL in both Fedora and EPEL. With changes from 2.0.11 -> 2.1.0 there was an incompatible change in that the daemon scripts were renamed:
mmmd_agent -> mmm_agentd
mmmd_mon -> mmm_mond
Upgrades obviously break because the INIT scripts and configuration files reference the path to the files. Would a sufficient work-around be a symlink to the old path, or would that not be kosher for any reason?
Thank you for your feedback.
---
derks
14 years, 1 month
creating file in koji allowed?
by Thomas Spura
Hi list,
I never saw this anywhere before, so I'd like to ask here first, before
doing so ;)
Is it allowed to create a file ~/.mpd.conf, when building in koji and
deleting afterwards?
I need to write down a password into that file, for running a testsuite.
If that file does not exist, I can't run mpich2 tests.
Here is the snipped, I intend to use:
"""
%{_mpich2_load}
# create ~/.mpd.conf, if it does not yet exist
if [ -e ~/.mpd.conf ]; then
# working locally, don't delete ~/.mpd.conf
DONT_DEL="TRUE"
else
DONT_DEL="FALSE"
echo MPD_SECRETWORD=$(pwgen -s 50 1) > ~/.mpd.conf
chmod 600 ~/.mpd.conf
fi
mpd --daemon
PYTHONPATH=%{buildroot}%{python3_sitearch}/mpich2 \
python3 test/runalltest.py
mpdallexit
# delte ~/.mpd.conf again
if [ $DONT_DEL = "FALSE" ]; then
rm ~/.mpd.conf
fi
%{_mpich2_unload}
"""
This prefectly works with my local rpmbuild and local mock and there
won't probably be an issue with that, but is this 'the safe way' to do?
(This also use my local ~/.mpd.conf and does not owerwrite it or deletes
it afterwards.)
Thanks in advance.
Thomas
14 years, 1 month
Read this if your package BuildRequires qt(4)-devel!!!
by Kevin Kofler
Hi,
for all maintainers of packages which BuildRequire qt4-devel (or qt-devel, but
the versioned virtual Provides is preferred): please, when you plan to push
updates for your packages, ALWAYS CHECK what version of Qt your package got
built against and DO NOT PUSH your update to stable before that version of Qt
goes stable! A package built against Qt 4.6 WILL NOT WORK AT ALL with Qt
4.5!!! (This is always the case, Qt is backwards- but not forwards-
compatible.)
Currently, buildroot overrides for Qt 4.6 are in effect (intermittently, as Qt
4.6 can and will be untagged from the buildroot on request to build updates
which need to go out soon, but we need it in the buildroot to build anything
related to KDE 4.4), so a package built now CANNOT go to stable before the big
Qt 4.6 / KDE 4.4 / SIP 4.10 update does. If you need to push an urgent update,
please ask Rex Dieter (rdieter on Freenode IRC) or another rel-eng member to
get the stuff out of the buildroot for a moment, and follow the instructions
given on IRC. If your update is not urgent, I recommend just not pushing it
out to stable before the big Qt/KDE/SIP update.
NOT FOLLOWING THOSE INSTRUCTIONS WILL LEAD TO YOUR PACKAGE BEING BROKEN IN THE
STABLE UPDATES!!! YOU HAVE BEEN WARNED!
I am sorry if I sound abrasive, but we already had at least 2 packages which
were broken due to this issue (just for 4.6, there were more such issues with
previous upgrades) and it looks like our previous devel-announce message was
not clear enough. Please double-check before you hit that "push to stable"
button! Thanks in advance.
We will look into using some less dangerous process (special build tags?) for
future Qt updates as this is just not working, but for now please be careful.
Kevin Kofler
_______________________________________________
devel-announce mailing list
devel-announce(a)lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel-announce
14 years, 1 month
Call for Participation - Fedora 13 Talking Points
by Robyn Bergeron
Talking points are key highlights of the new release. They should be
compelling, but they will not necessarily be comprehensive. There are
different types of talking points for different types of people:
general desktop users/everyone, developers, and sysadmins. For the
Fedora 13 cycle, we will also have talking points to address some of
the Spins. They are meant to provide a short, effective answer to the
question, "What cool stuff is in the latest release of Fedora?"
Each cycle, the Marketing team compiles a short list of approximately
three talking points for each of these audiences for the upcoming
release. For Fedora 13, they're found here:
https://fedoraproject.org/wiki/Fedora_13_Talking_Points
If you have a talking point that you feel meets the criteria found on
the talking points SOP page at
https://fedoraproject.org/wiki/Talking_points_SOP, add it to the the
table on the F13 page with supporting information. Please make your
contributions and changes on the wiki page, so that the Marketing team
can efficiently capture and consider your input.
The Marketing team will make final adjustments to the list of talking
points at their meeting on February 23, which will be announced on the
marketing list and is open to everyone. If you are interested in
attending the meeting, the agenda, location, and time details can be
found at http://fedoraproject.org/wiki/Marketing_meetings. Following
the meeting, the finalized list of talking points will be announced,
and posted to https://fedoraproject.org/wiki/Fedora_13_Talking_Points.
We welcome you to participate in the process!
Cheers,
-Robyn
_______________________________________________
devel-announce mailing list
devel-announce(a)lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel-announce
14 years, 1 month
gconf settings for gdm login
by Jon Masters
Folks,
What's up with the login screen on rawhide these days? It's defaulting
to no longer listing available users. I can see security sensitive
reasons for this, but it's a behavior change - is it intentional?
Jon.
14 years, 1 month
F-13 Branched report: 20100228 changes
by Fedora compose checker
Compose started at Sun Feb 28 09:15:10 UTC 2010
Broken deps for i386
----------------------------------------------------------
anaconda-13.32-1.fc13.i686 requires python-urlgrabber >= 0:3.9.1-5
balsa-2.4.6-3.fc13.i686 requires libgmime-2.4.so.2
blahtexml-0.6-5.fc12.i686 requires libxerces-c.so.28
doodle-0.6.7-5.fc12.i686 requires libextractor.so.1
easystroke-0.5.2-1.fc13.i686 requires libboost_serialization-mt.so.5
edje-0.9.9.050-6.fc12.i686 requires libembryo.so.0
fusecompress-2.6-3.fc12.i686 requires libboost_serialization-mt.so.5
fusecompress-2.6-3.fc12.i686 requires libboost_system-mt.so.5
fusecompress-2.6-3.fc12.i686 requires libboost_program_options-mt.so.5
fusecompress-2.6-3.fc12.i686 requires libboost_filesystem-mt.so.5
fusecompress-2.6-3.fc12.i686 requires libboost_iostreams-mt.so.5
gnome-python2-totem-2.29.1-4.fc13.i686 requires libtotem-plparser.so.12
koan-2.0.3.1-1.fc13.noarch requires mkinitrd
libvirt-qpid-0.2.17-3.fc12.i686 requires qpidc >= 0:0.5.790661
linphone-2.1.1-4.fc12.i686 requires libortp.so.7
matahari-0.0.4-7.fc13.i686 requires qpidc >= 0:0.5.819819
ovaldi-5.5.25-2.fc13.i686 requires libxerces-c.so.28
ovirt-server-0.100-4.fc12.noarch requires qpidd
ovirt-server-0.100-4.fc12.noarch requires qpidc
qmf-devel-0.5.819819-4.fc13.i686 requires qpidc-client-devel = 0:0.5.819819-4.fc13
qpidc-server-devel-0.5.819819-4.fc13.i686 requires qpidc-client-devel = 0:0.5.819819-4.fc13
qpidc-server-rdma-0.5.819819-4.fc13.i686 requires qpidc-client-rdma = 0:0.5.819819-4.fc13
qpidc-server-ssl-0.5.819819-4.fc13.i686 requires qpidc-client-ssl = 0:0.5.819819-4.fc13
zikula-module-menutree-2.2-1.fc13.noarch requires zikula >= 0:1.2
Broken deps for x86_64
----------------------------------------------------------
anaconda-13.32-1.fc13.x86_64 requires python-urlgrabber >= 0:3.9.1-5
balsa-2.4.6-3.fc13.x86_64 requires libgmime-2.4.so.2()(64bit)
blahtexml-0.6-5.fc12.x86_64 requires libxerces-c.so.28()(64bit)
doodle-0.6.7-5.fc12.i686 requires libextractor.so.1
doodle-0.6.7-5.fc12.x86_64 requires libextractor.so.1()(64bit)
easystroke-0.5.2-1.fc13.x86_64 requires libboost_serialization-mt.so.5()(64bit)
edje-0.9.9.050-6.fc12.i686 requires libembryo.so.0
edje-0.9.9.050-6.fc12.x86_64 requires libembryo.so.0()(64bit)
fusecompress-2.6-3.fc12.x86_64 requires libboost_system-mt.so.5()(64bit)
fusecompress-2.6-3.fc12.x86_64 requires libboost_program_options-mt.so.5()(64bit)
fusecompress-2.6-3.fc12.x86_64 requires libboost_iostreams-mt.so.5()(64bit)
fusecompress-2.6-3.fc12.x86_64 requires libboost_filesystem-mt.so.5()(64bit)
fusecompress-2.6-3.fc12.x86_64 requires libboost_serialization-mt.so.5()(64bit)
gnome-python2-totem-2.29.1-4.fc13.x86_64 requires libtotem-plparser.so.12()(64bit)
koan-2.0.3.1-1.fc13.noarch requires mkinitrd
libvirt-qpid-0.2.17-3.fc12.x86_64 requires qpidc >= 0:0.5.790661
linphone-2.1.1-4.fc12.i686 requires libortp.so.7
linphone-2.1.1-4.fc12.x86_64 requires libortp.so.7()(64bit)
matahari-0.0.4-7.fc13.x86_64 requires qpidc >= 0:0.5.819819
ovaldi-5.5.25-2.fc13.x86_64 requires libxerces-c.so.28()(64bit)
ovirt-server-0.100-4.fc12.noarch requires qpidd
ovirt-server-0.100-4.fc12.noarch requires qpidc
qmf-devel-0.5.819819-4.fc13.i686 requires qpidc-client-devel = 0:0.5.819819-4.fc13
qmf-devel-0.5.819819-4.fc13.x86_64 requires qpidc-client-devel = 0:0.5.819819-4.fc13
qpidc-server-devel-0.5.819819-4.fc13.i686 requires qpidc-client-devel = 0:0.5.819819-4.fc13
qpidc-server-devel-0.5.819819-4.fc13.x86_64 requires qpidc-client-devel = 0:0.5.819819-4.fc13
qpidc-server-rdma-0.5.819819-4.fc13.x86_64 requires qpidc-client-rdma = 0:0.5.819819-4.fc13
qpidc-server-ssl-0.5.819819-4.fc13.x86_64 requires qpidc-client-ssl = 0:0.5.819819-4.fc13
zikula-module-menutree-2.2-1.fc13.noarch requires zikula >= 0:1.2
New package perl-Config-MVP
Multivalue-property package-oriented configuration
Updated Packages:
anaconda-13.32-1.fc13
---------------------
* Thu Feb 25 2010 David Lehman <dlehman(a)redhat.com> - 13.32-1
- Check for the real device-mapper nodes in /proc/swaps. (#567840) (dlehman)
- It's necessary to give each vfprintf invocation a fresh va_list (#568235)
(akozumpl)
- Don't unconditionally unskip the partition step on failure (#567889).
(clumens)
- rpm doesn't always give the callback a tuple (#567878). (clumens)
* Wed Feb 24 2010 David Cantrell <dcantrell(a)redhat.com> - 13.31-1
- Revert "There is no kernel-PAE package anymore, use kernel for xen
(#559347)." (dcantrell)
- logging: make loader say 'loader' (#563009). (akozumpl)
- Make loader log into syslog (so remote logging works for it as well)
(#524980) (akozumpl)
* Tue Feb 23 2010 Chris Lumens <clumens(a)redhat.com> - 13.30-1
- Revert "Add back hald for Xorg input device queries (#553780)" (clumens)
- No longer remove persistent udev rules files (#566948). (clumens)
- When BUILDARCH==ppc64, set BASEARCH to ppc (#524235). (dcantrell)
- There is no kernel-PAE package anymore, use kernel for xen (#559347).
(dcantrell)
- Fix a typo, leaving one less string needing translation (#567427).
(clumens)
- Don't show BIOS RAID and multipath members in the cleardisks UI (#567281).
(clumens)
* Mon Feb 22 2010 David Cantrell <dcantrell(a)redhat.com> - 13.29-1
- DiskLabel.status can't be determined so return False. (#563526,#561074)
(dlehman)
- Remove getDasdDevPort() and getDasdState() from isys.py. (dcantrell)
- Replace calls to isys.getDasdPorts() with calls to new getDasdPorts()
(dcantrell)
- Add getDasdPorts() to storage/dasd.py. (dcantrell)
- Remove isys/dasd.c, functions no longer needed in isys. (dcantrell)
- Fix creation of encrypted md members and pvs in kickstart. (#567396)
(dlehman)
- Don't align free space geometries in getFreeRegions. (#565692) (dlehman)
- Align extended partitions like we do other partitions. (dlehman)
- Don't allow the host's LD_LIBRARY_PATH affect get_dso_deps (#565887).
(clumens)
- Remove a couple redundant network bring up calls. (clumens)
- Reset the resolver cache after bringing up the network (#562209). (clumens)
- Let's have /etc/xorg.conf.d in stage2 (#566396) (akozumpl)
- Add the filter UI screens to the list of translatable files (#567216).
(clumens)
- Don't traceback when a user tries to put /boot on an LV (#566569)
(hdegoede)
- RescueInterface should inherit from InstallInterfaceBase too (hdegoede)
* Fri Feb 19 2010 Chris Lumens <clumens(a)redhat.com> - 13.28-1
- Allow --ignoremissing to work for @base and @core (#566752).
(clumens)
- Add device node names to the filter UI, hidden by default (#566375).
(clumens)
- logging: initialize tty3 logging in anaconda_log, along with all other
basic loggers. (akozumpl)
- logging: introduce stderr logger and use it for critical situations in
kickstart.py. (akozumpl)
- logging: Loggers live a cosmopolitan life, forget about them after
created. (akozumpl)
- logging: remove AnacondaLog's unused default parameter. (akozumpl)
- logging, fix: setting remote logging from kicstart (akozumpl)
- logging: addFileHandler does not set autoLevel by default (akozumpl)
- Allow deleting the interface property, too (#566186). (clumens)
coriander-2.0.0-1.fc13
----------------------
* Thu Feb 25 2010 Tim Niemueller <tim(a)niemueller.de> - 2.0.0-1
- Update to latest stable release
- Added patch for implicit DSO linking
frepple-0.7.1-2.fc13
--------------------
* Fri Feb 19 2010 Johan De Taeye <jdetaeye(a)users.sourceforge.net> 0.7.1-2
- patch to work with xerces-c 3.x
geeqie-1.0-1.fc13
-----------------
* Fri Feb 19 2010 Michael Schwendt <mschwendt(a)fedoraproject.org> - 1.0-1
- update to 1.0 final release
gnome-packagekit-2.29.4-0.2.20100211git.fc13
--------------------------------------------
* Wed Feb 24 2010 Matthias Clasen <mclasen(a)redhat.com> - 2.29.4-0.2.20100211git
- Fix an infinite loop in the update viewer
kdebase-4.4.0-5.fc13
--------------------
* Fri Feb 26 2010 Kevin Kofler <Kevin(a)tigcc.ticalc.org> - 6:4.4.0-5
- fix disabling automatic spell checking in the Konqueror UI (kde#228593)
* Thu Feb 25 2010 Kevin Kofler <Kevin(a)tigcc.ticalc.org> - 6:4.4.0-4
- backport fix for folderview getting resorted on file creation (kde#227157)
kdebase-workspace-4.4.0-8.fc13
------------------------------
* Fri Feb 26 2010 Kevin Kofler <Kevin(a)tigcc.ticalc.org> - 4.4.0-8
- fix the Games menu in the classic menu mixing up Name and Description
* Fri Feb 19 2010 Rex Dieter <rdieter(a)fedoraproject.org> - 4.4.0-7
- version solid-bluetooth(-devel) better
* Fri Feb 19 2010 Rex Dieter <rdieter(a)fedoraproject.org> - 4.4.0-6
- solid-bluetooth and Requires: bluez ... pulls unwanted baggage (#566306)
* Tue Feb 16 2010 Rex Dieter <rdieter(a)fedoraproject.org> - 4.4.0-5.1
- Requires: kbluetooth (<f13)
kdesdk-4.4.0-4.fc13
-------------------
* Fri Feb 26 2010 Kevin Kofler <Kevin(a)tigcc.ticalc.org> - 4.4.0-4
- also remove hardcoded setCheckSpellingEnabled(true) calls (kde#228587)
* Fri Feb 26 2010 Kevin Kofler <Kevin(a)tigcc.ticalc.org> - 4.4.0-3
- fix Cervisia not to unconditionally enable auto spell checking (kde#228587)
- BR binutils-static (F13+)
* Wed Feb 24 2010 Rex Dieter <rdieter(a)fedoraproject.org> - 4.4.0-2
- -utils: Requires: qt4 dep
- -libs: Requires: %name ...
mysql-connector-java-5.1.12-1.fc13
----------------------------------
* Fri Feb 19 2010 Miloš Jakubíček <xjakub(a)fi.muni.cz> - 1:5.1.12-1
- Update to 5.1.12
perl-Cache-FastMmap-1.35-1.fc13
-------------------------------
* Sat Feb 20 2010 Iain Arnell <iarnell(a)gmail.com> 1.35-1
- update to latest upstream version
Summary:
Added Packages: 1
Removed Packages: 0
Modified Packages: 10
14 years, 1 month