DevHeads.net

Postings by Michael Cronenworth

Rawhide armv7hl buildroot busted?

I've tried two builds over the past couple hours that result in failures. Other
architectures are building.

Is there something wrong in glibc/gcc/makefile for armv7hl?

build.log
<a href="https://kojipkgs.fedoraproject.org//work/tasks/408/24900408/build.log" title="https://kojipkgs.fedoraproject.org//work/tasks/408/24900408/build.log">https://kojipkgs.fedoraproject.org//work/tasks/408/24900408/build.log</a>

root.log
<a href="https://kojipkgs.fedoraproject.org//work/tasks/408/24900408/root.log" title="https://kojipkgs.fedoraproject.org//work/tasks/408/24900408/root.log">https://kojipkgs.fedoraproject.org//work/tasks/408/24900408/root.log</a>

devel mail spam

I'm seeing many duplicate messages coming from <a href="mailto: ... at citadel dot stu.neva.ru"> ... at citadel dot stu.neva.ru</a> for messages
from the past few days.

Anyone on it?

Cannot find cause for ARM build failing

Hello,

My attempt at building a wine update today failed for only the ARMv7 (32-bit) builds
across F26, F27, and Rawhide. I'm seeing the build log[1] show the following message:

error: variadic functions must use the base AAPCS variant

I'm unfamiliar with this type of failure. The code that is failing has not been
changed in years.

Urgent attention required; ImageMagick update breakage

Hi all,

An ImageMagick update (6.9 => 7.0) with an SONAME bump and other breakage has been
pushed to F25 and higher.

First, the update introduces regressions on s390x and ppc64 arches.
- <a href="https://bugzilla.redhat.com/show_bug.cgi?id=1484578" title="https://bugzilla.redhat.com/show_bug.cgi?id=1484578">https://bugzilla.redhat.com/show_bug.cgi?id=1484578</a>
- <a href="https://bugzilla.redhat.com/show_bug.cgi?id=1484579" title="https://bugzilla.redhat.com/show_bug.cgi?id=1484579">https://bugzilla.redhat.com/show_bug.cgi?id=1484579</a>

Secondly, rebuilds are required for:

autotrace-0.31.1-44.fc26.src.rpm
converseen-0.9.6.2-1.fc27.src.rpm
dmtx-utils-0.7.4-2.fc27.src.rpm
drawtiming-0.7.1-20.fc26.src.rpm
F gtatool-2.2.0-4.fc27.src.rpm
imageinfo-0.05-25.fc26.src.rpm
inkscape-0.92.1-5.20170510bzr15686.fc27.src.rpm
kx

F26 Mesa update

Hello,

Is there a reason for delaying the update for mesa-17.1.2 to F26?

We're getting buildroot failures for some F26 packages that depend on mesa (llvm was
updated).

Thanks,
Michael

Call to retire gstreamer-0.10

Hi,

As I'm sure everyone is well aware gstreamer-0.10 is obsolete and has been replaced
by gstreamer-1.0. An upstream release for gst-0.10 has not been made in over 4
years. Last month there were a handful of security vulnerabilities disclosed[1] and
upstream no longer maintains the 0.10 series. While we could patch and update our
Fedora packages this is a perfect opportunity to retire it.

Here is an approximate list of packages depending on gst-0.10. They themselves
should be considered obsolete or retirement material as their upstreams may not have
adopted gst-1.0.

Intel Vulkan driver status

Why is the Vulkan driver being left out right now?

Here's the RFE[1] from July asking for it to be enabled.

[1] <a href="https://bugzilla.redhat.com/show_bug.cgi?id=1356229" title="https://bugzilla.redhat.com/show_bug.cgi?id=1356229">https://bugzilla.redhat.com/show_bug.cgi?id=1356229</a>

ftplib license change

Hello,

The ftplib (and mingw-ftplib) package changed license types.

LGPLv2+ => Artistic 2.0

CC'ing Spot as he missed changing the License field in the package update.

Thanks,
Michael

Review swap, 7 php libraries

Hi,

Please look over the following reviews and let me know if you are able to swap.

They have been pre-reviewed so they should be easy to do.

<a href="https://bugzilla.redhat.com/show_bug.cgi?id=1231353" title="https://bugzilla.redhat.com/show_bug.cgi?id=1231353">https://bugzilla.redhat.com/show_bug.cgi?id=1231353</a> php-cssjanus
<a href="https://bugzilla.redhat.com/show_bug.cgi?id=1264179" title="https://bugzilla.redhat.com/show_bug.cgi?id=1264179">https://bugzilla.redhat.com/show_bug.cgi?id=1264179</a> php-mediawiki-at-ease
<a href="https://bugzilla.redhat.com/show_bug.cgi?id=1264191" title="https://bugzilla.redhat.com/show_bug.cgi?id=1264191">https://bugzilla.redhat.com/show_bug.cgi?id=1264191</a> php-wikimedia-assert
<a href="https://bugzilla.redhat.com/show_bug.cgi?id=1264199" title="https://bugzilla.redhat.com/show_bug.cgi?id=1264199">https://bugzilla.redhat.com/show_bug.cgi?id=1264199</a> php-wikimedia-avro
<a href="https://bugzilla.redhat.com/show_bug.cgi?id=1264209" title="https://bugzilla.redhat.com/show_bug.cgi?id=1264209">https://bugzilla.redhat.com/show_bug.cgi?id=1264209</a> php-wikimedia-ip-set
<a href="https://bugzilla.redhat.com/show_bug.cgi?id=1264213" title="https://bugzilla.redhat.com/show_bug.cgi?id=1264213">https://bugzilla.redhat.com/show_bug.cgi?id=1264213</a> php-wikimedia-wrappedstring
<a href="https://bugzilla.re" title="https://bugzilla.re">https://bugzilla.re</a>

PHP package reviews for libraries in mediawiki

Hello,

Mediawiki was bundling PHP libraries. The latest stable release is attempting to
un-bundle them by clearly splitting them out into a "vendor" sub-directory.

I have created package reviews for these libraries so they do not have to be
bundled.

Firefox addon signing

I'm sure those that need to know, know, but for those that haven't heard[1]
Mozilla's official Firefox build will enforce addons to contain a Mozilla signature
without any runtime option to disable the check.

Initially this prevents Fedora packaged addons since they are unsigned. The Mozilla
signing process takes time and can't be part of a package building process.

Is Fedora going to get authorization to build Firefox with a runtime disable option?

[1] <a href="https://blog.mozilla.org/addons/2015/02/10/extension-signing-safer-experience/" title="https://blog.mozilla.org/addons/2015/02/10/extension-signing-safer-experience/">https://blog.mozilla.org/addons/2015/02/10/extension-signing-safer-exper...</a>

Unresponsive maintainer: masahase

Hello,

Does anyone know how to contact HASEGAWA Masahiro?

FAS: masahase

He has some untouched (trivial) bug reports (example[1]). He did not respond to
a personal e-mail.

Pkgdb: <a href="https://admin.fedoraproject.org/pkgdb/packager/masahase/" title="https://admin.fedoraproject.org/pkgdb/packager/masahase/">https://admin.fedoraproject.org/pkgdb/packager/masahase/</a>

Thanks,
Michael

[1] <a href="https://bugzilla.redhat.com/show_bug.cgi?id=903740" title="https://bugzilla.redhat.com/show_bug.cgi?id=903740">https://bugzilla.redhat.com/show_bug.cgi?id=903740</a>

SystemD service stop behavior

Should "systemctl stop foo.service" stop all parent and child service processes?

Example: GlusterFS starts a service daemon (glusterd) and a brick daemon
(glusterfsd). When a user issues "systemctl stop glusterd" the service daemon is
stopped but the brick daemon is left running.

I have recently learned that it is expected behavior[1], but this seems to be a
defeat of the purpose of the stop command. What is @devel thoughts?

[1] <a href="https://bugzilla.redhat.com/show_bug.cgi?id=1022542#c1" title="https://bugzilla.redhat.com/show_bug.cgi?id=1022542#c1">https://bugzilla.redhat.com/show_bug.cgi?id=1022542#c1</a>
[2] I did not contact JoeJulian as I feel this is a Fedora issue and not a
GlusterFS issue.

Firefox and Gstreamer

Why is gstreamer explicitly disabled for Firefox 24+?

(before I make an RFE bug)

Thanks,
Michael

Contact Ian Weller

Does anyone know how to contact Ian Weller?

He has not responded to his gmail account since I e-mailed him last
week. His last fedora activity seems to be about 6 months ago.

He is the owner of the mediawiki-ParserFunctions package, which has now
been merged with the upstream mediawiki package. I need him to
orphan/retire his package.

Thanks,
Michael

Mediawiki 1.20.2 has landed

Hi,

The mediawiki package has been updated to 1.20.2 for Rawhide and 1.19.3
for F17/F18.

The mediawiki-math/nopath packages have been merged into just
"mediawiki" (with Obsoletes). There is one package that depended on the
old packages.

mediawiki-semantic
owner: jlaska

If you have your own wiki please look over the release notes and test
the packages in updates-testing. We're coming from a 2 year old version
so there are a few changes.

Thanks,
Michael

Non-responsive Maintainer: mediawiki

The mediawiki package is severely outdated. I have attempted to contact
the maintainer and received only one reply in a couple of weeks. He has
ignored my requests for co-maintainership.

I commented on the NRM bug[1] two weeks ago with no response. The bug
itself is months old, but he pops in to keep from having the package
taken. At this point it is time for someone to update the package as it
is currently a security hazard.

mediawiki package split

The current "mediawiki" package is several years old at version 1.16. A
new package, named "mediawiki119" at version 1.19 has been reviewed[1]
and imported into Fedora. I understand some packages require multiple
versions to be available, but mediawiki does not require multiple packages.

Why was "mediawiki119" given an approved review?

I only knew about the "mediawiki119" package from the NRM bug[2] on
"mediawiki". The current version, 1.16, is riddled with security holes
and will not be updated by upstream.

Emacs package guidelines introduce unnecessary deps

Upon upgrading a F17 box today I notice "desktop-file-utils" wants to
pull in "emacs-filesystem". When I investigate the cause I find it to be
the Emacs packaging guidelines.

The guidelines now force any package that has Emacs add-ons to install
them in the main package and Requires: emacs-filesystem. Emacs is not
installed by default and I do not use Emacs, nor will I ever. I'm not
sure why having sub-packages was such a negative thing. Can we bring
back sub-packages?

Good bye!

Seeing Maemo as an opportunity to have a full, Linux experience to take
with me on the go was exciting. Several years of active community
members and developers gave proof that that excitement was real.
Unfortunately all of that excitement had been extinguished once the
hardware hit a brick wall. This e-mail seems like a belated good-bye
seeing that no real activity in Maemo has happened in over a year. I
refer Maemo as both 5 and 6(MeeGo).

The last great stand, a.k.a. "Community Awards", was nothing more than a
front to give away hardware mostly to friends.

Community Awards

Hello,

I'd like to enter the award contest for an N9.

TMO: mooninite
<a href="http://talk.maemo.org/member.php?u=25907" title="http://talk.maemo.org/member.php?u=25907">http://talk.maemo.org/member.php?u=25907</a>
<a href="http://maemo.org/profile/view/mooninite/" title="http://maemo.org/profile/view/mooninite/">http://maemo.org/profile/view/mooninite/</a>

I've helped develop the FLAC metadata plugin when the N900 was released.
I have also helped test and QA (provide karma) for various packages that
includes community kernel packages (kernel-power, etc).

Tips on how to handle tmpfile changes

The vnstat service has traditionally run as the root user. This was
fixed in Fedora 16 and higher to run as the vnstat user, but the same
fix was just introduced[1] into Fedora 15. There is a problem with this
fix in that it requires "systemd-tmpfiles" to be run to create the new
/run/vnstat directory required to store the pid file. This new directory
also required the /etc/vnstat.conf file to be changed as the PidFile
variable tells vnstat where to create the pid file.

calls to gasyncqueue_new fail

I'm trying to get a GTK app I wrote for a regular desktop to run on the
N900.

When I compile the app in the SDK, everything goes fine. When I attempt
to run the app on the phone it aborts after attempting to call
g_async_queue_new(). The call is done 5 lines in to the program and it
works normally on a regular desktop.

Is there something Maemo-specific that I need to do to allow async queues?

Thanks,
Michael

Trouble building for rawhide on F15

It's been some time since I've had to build a Fedora package, but when I
tried today I'm having some trouble.

$ rpm -q fedpkg
fedpkg-0.5.9.2-2.fc15.noarch
$ git branch
* master
$ fedpkg build
Could not initiate build: Unknown build target: dist-rawhide

Any ideas?

My Perception of the Maemo/MeeGo Community

This is a very long, blog-like message. Thanks to those who read through
it. I am posting here instead of my personal blog as I want to create a
discussion instead of a reaction.

I feel I should explain my thoughts of the Maemo/MeeGo community from
the perspective of a US resident. I realize my opinions are mine and
mine alone. I welcome any response, particularly those who live in the
US. My perception of the Maemo/MeeGo community is that those directly
involved (Nokia employees, Maemo council members, Maemo workers) have a
self-entitled, elitist attitude.

Who can buy an N9?

No one in the USA or UK can.

Looks like my Maemo/MeeGo dream starts and ends with the N900.

Is there anyone even on this mailing list that is in one of the
countries you can buy one?

So... where's the discussion?

The Nokia N9 announcement came and went.

I don't see any chatter on this list or the MeeGo list.

Am I that out of the loop?

Michael

Where do we go now?

Nokia has backed out of MeeGo and today they have backed out of QT. This
leaves the Maemo distribution solely in the hands of the community. I
would still like to contribute to Maemo (or MeeGo if a build is
released), but I have some concerns.

Will Maemo.org and its infrastructure be online for the foreseeable
future? I have a few Maemo apps (GTK-only) that I'm in the process of
developing, but if maemo.org and its repositories are going to be taken
offline I don't see a need to continue development.

What about the Maemo Council?

PHP floating point bug possibly misinterpreted

PHP recently underwent a bit of a security crisis when many media
outlets disclosed the presence of a floating point bug. (RH bug here[1])

I feel I should bring attention to a PHP developer response[2] to this
bug. Warning: There is a bit of emotion inside. They feel it is a gcc
bug in FP handling, which seems (in my limited knowledge) to be
accurate. The gcc fix was provided, not by changing default behavior,
but by a compiler flag "-fexcess-precision=standard".

Java allows root access without permission?

Fedora 13 x86_64 with OpenJDK (not Sun) installed.

I was required to login to a web site today to configure a VPN and the
site installed a Cisco VPN Java applet. When it was finished installing
and was running I noticed the processes where running as root and had
installed into /opt. I had not given it my root password or any
permission to do so. It did not install any RPM package. It was all
driven by a Java applet.

$ ps -efw #id 502 is me
502 4791 2668 0 11:16 ?