anaconda-patches July 2014

anaconda-patches@lists.fedorahosted.org

14 participants
107 discussions

[blivet:master 0/8] mdadd fixing patch set
by Anne Mulhern 27 Jul '14

27 Jul '14

Only the last two patches are even remotely interesting. mulhern (8): Fix mdnominate error message. Use long messages for unittest errors. Still attempt to destroy even if remove failed. Add a test for mddetail on containers. Change Container.has_redundancy to True Add a test for activation. Revise mdadd. Add a bunch of tests for mdadd. blivet/devicelibs/mdraid.py | 30 ++-- blivet/devicelibs/raid.py | 2 +- blivet/devices.py | 14 +- tests/devicelibs_test/mdraid_interrogate_test.py | 22 ++- tests/devicelibs_test/mdraid_test.py | 196 +++++++++++++++++------ tests/pylint/pylint-false-positives | 1 + 6 files changed, 197 insertions(+), 68 deletions(-) -- 1.9.3

2 20

[master 0/8] Don't require a built tree to make dist, and other changes
by David Shea 27 Jul '14

27 Jul '14

The big change is that this gets rid of gtk-doc. gtk-doc mostly just makes things harder for us, so let's just not use it anymore. This will hopefully make things easier for those of you making and testing builds, since it means that you no longer need to run `make' before `make dist'. ./configure and `make po-pull' are still necessary. ./configure --enable-gtk-doc will fail, because there's no --enable-gtk-doc anymore. Other than that, add some things to some autotools files, remove some things, and do the tests in such a way that missing dependencies don't cause the configure script to fail. If you're just running `make dist' you don't really care whether or not gtk3-devel-3.very-latest is installed or not. gettext and gobject-introspection-devel are still required in order for autogen.sh to run, but intltool and the rest of the g* things are no longer hard requirements. David Shea (8): Dump gtk-doc. Don't distribute the gnome desktop file with translations Use the result from AC_FUNC_FORK at build time Include config.h in every C file. Add a couple of things that autoscan found Remove unused parts of the configure.ac files. Clean up the handling of CFLAGS Switch to kinder, gentler autoconf errors .gitignore | 7 ---- acinclude.m4 | 75 ++++++++++++++++++++++++++++++++++++ configure.ac | 75 +++++++----------------------------- data/liveinst/gnome/Makefile.am | 3 +- pyanaconda/isys/auditd.c | 6 +++ utils/dd/dd_extract.c | 3 ++ utils/dd/dd_list.c | 3 ++ utils/dd/rpmutils.c | 2 + widgets/Makefile.am | 4 +- widgets/acinclude.m4 | 1 + widgets/autogen.sh | 1 - widgets/configure.ac | 36 ++++++++--------- widgets/doc/AnacondaWidgets-docs.xml | 63 ------------------------------ widgets/doc/Makefile.am | 59 ---------------------------- widgets/src/BaseStandalone.c | 2 + widgets/src/BaseWindow.c | 2 + widgets/src/DiskOverview.c | 2 + widgets/src/HubWindow.c | 2 + widgets/src/LayoutIndicator.c | 2 + widgets/src/MountpointSelector.c | 2 + widgets/src/SpokeSelector.c | 2 + widgets/src/SpokeWindow.c | 2 + widgets/src/StandaloneWindow.c | 2 + widgets/src/glade-adaptor.c | 2 + widgets/src/intl.h | 2 +- widgets/src/widgets-common.c | 2 + 26 files changed, 149 insertions(+), 213 deletions(-) create mode 100644 acinclude.m4 create mode 120000 widgets/acinclude.m4 delete mode 100644 widgets/doc/AnacondaWidgets-docs.xml delete mode 100644 widgets/doc/Makefile.am -- 1.9.3

3 21

[master] zRAM follow-up follow-up
by Vratislav Podzimek 26 Jul '14

26 Jul '14

Turns out that with the squashfs.img fetched from network we need a lot more RAM. PATCH 1/4 adds a code that makes check_memory aware of that fact, PATCH 2/4 turns on zRAM up to 2 GB RAM because with squashfs.img from network it is necessary or really useful up to 2 GB. PATCH 3/4 fixes an issue I've discovered when fiding memory limits where anaconda-yum got oom-killed and the anaconda process didn't care and failed on missing authconfig in the installed system where nothing was actually installed. Finally PATCH 4/4 is just a cosmetic change removing useless numeric operations potentially leading to errors. Vratislav Podzimek (4): RAM requirements depend on squashfs.img's origin Use zRAM swap up to 2 GB of RAM Raise exception if reading lines from a killed process Do not multiply/divide RAM sizes by 1024 back and forth anaconda | 14 +++++++++++--- pyanaconda/isys/__init__.py | 9 +++++---- pyanaconda/iutil.py | 16 ++++++++++++++++ scripts/zramswapon | 4 +++- 4 files changed, 35 insertions(+), 8 deletions(-) -- 1.9.3

2 5

[blivet] groundwork for better blivet unit test coverage
by David Lehman 25 Jul '14

25 Jul '14

The first two patches establish a framework for using regular files as fake disks so that things that depend on parted objects (most of blivet) can function without root access. While the primary motivation was testing partition allocation in a non-destructive way as a non-root user, this also enables unit testing of several areas of the devicetree, action registration, sorting and pruning, the device factories, and determining resize constraints for existing filesystems. The other two are the first tests to use the framework. David Lehman (4): Add a contextmanager to create and remove sparse tempfiles. Add a DiskFile class for testing partitioning code as a non-root user. Add a couple of tests for blivet.partitioning.DiskChunk. Add some tests for blivet.partitioning.addPartition. blivet/devices.py | 29 ++++++ blivet/util.py | 16 +++ tests/partitioning_test.py | 243 +++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 288 insertions(+) -- 1.9.3

3 9

[PATCH lorax rhel7-branch] runtime-install: Add rpm-ostree
by Radek Vykydal 25 Jul '14

25 Jul '14

From: Colin Walters <walters(a)verbum.org> rpm-ostree is a new payload type available in Anaconda, will be used by RHEL Atomic Host. Related: rhbz#1113535 Port of rpmostreepayload patches from master commit 3345a97f35e3c0af358dbb6d5ee7e0caf4eba051 --- share/runtime-install.tmpl | 2 ++ 1 file changed, 2 insertions(+) diff --git a/share/runtime-install.tmpl b/share/runtime-install.tmpl index a1d21d3..8ff8b72 100644 --- a/share/runtime-install.tmpl +++ b/share/runtime-install.tmpl @@ -7,6 +7,8 @@ installpkg anaconda anaconda-widgets installpkg tmux installpkg firstboot installpkg iscsi-initiator-utils +## Other available payloads +installpkg rpm-ostree ## kernel and firmware installpkg kernel -- 1.9.3

2 1

[PATCH master/rhel7] Add parents to ksdata for btrfs subvols (#1072060).
by Chris Lumens 24 Jul '14

24 Jul '14

Without this, the resulting kickstart file will not be valid for anaconda due to a lack of parents on any subvol line. --- blivet/devices.py | 1 + 1 file changed, 1 insertion(+) diff --git a/blivet/devices.py b/blivet/devices.py index 1028817..0110a46 100644 --- a/blivet/devices.py +++ b/blivet/devices.py @@ -5363,6 +5363,7 @@ class BTRFSSubVolumeDevice(BTRFSDevice): data.subvol = True data.name = self.name data.preexist = self.exists + data.devices = [self.parents[0]] @classmethod def isNameValid(cls, name): -- 1.9.3

1 1

[PATCH 0/6] Try harder to get API docs up on readthedocs.org
by David Lehman 24 Jul '14

24 Jul '14

I finally have some understanding of sphinx autodoc. The makefile in doc/ runs sphinx-apidoc before running sphinx-build -- that's where the API docs come from. Since readthedocs runs sphinx-build directly (without running sphinx-apidoc first) there are no API docs generated there. The obvious solution is to keep the API docs in git so they are in place for readthedocs' build. We'll have to update them if we move/add/remove modules, but that should be about it as far as maintenence. We can always run sphinx-apidoc again, change the title in the modules.rst file, add the member-order option throughout, and update them wholesale if we want to. David Lehman (6): Add API docs. Order module members by position in source instead of alphabetically. Change title of top-level module list to API. Don't remove api docs every time you make a doc target. Mock selinux when building docs. Fix sphinx formatting of clode blocks in devicefactory docstrings. .gitignore | 4 - blivet/devicefactory.py | 6 +- doc/Makefile | 1 - doc/blivet.devicelibs.rst | 102 +++++++++++++++++++++++ doc/blivet.formats.rst | 120 ++++++++++++++++++++++++++++ doc/blivet.rst | 200 ++++++++++++++++++++++++++++++++++++++++++++++ doc/conf.py | 2 +- doc/index.rst | 2 +- doc/modules.rst | 7 ++ 9 files changed, 434 insertions(+), 10 deletions(-) create mode 100644 doc/blivet.devicelibs.rst create mode 100644 doc/blivet.formats.rst create mode 100644 doc/blivet.rst create mode 100644 doc/modules.rst -- 1.9.3

5 13

[master] Use parallel instead of xargs
by Vratislav Podzimek 24 Jul '14

24 Jul '14

parallel is a nice utility that can save us a lot of work when running tests in parallel. By using parallel we can drop things like creating temporary files with the outputs and failure-flaging. Signed-off-by: Vratislav Podzimek <vpodzime(a)redhat.com> --- scripts/githooks/pre-push | 7 +++---- tests/glade/run_glade_tests.sh | 2 +- tests/pylint/pylint-one.sh | 8 ++++---- tests/pylint/runpylint.sh | 22 ++++------------------ 4 files changed, 12 insertions(+), 27 deletions(-) diff --git a/scripts/githooks/pre-push b/scripts/githooks/pre-push index db8c8ae..ea2e154 100755 --- a/scripts/githooks/pre-push +++ b/scripts/githooks/pre-push @@ -72,11 +72,10 @@ fi ## main ## # fork separate process for every commit, querying bugzilla takes time -num_commits=$(echo ${commits}|wc -w) -echo ${commits}|xargs -n1 --max-procs=${num_commits} .git/hooks/check_commit_msg.sh ${release} -xargs_ret=$? +echo -n ${commits}|parallel -d' ' --gnu -j0 .git/hooks/check_commit_msg.sh ${release} {} +ret=$? if [ ${ACK_FATAL} == "0" ]; then - test ${xargs_ret} -eq 0 + test ${ret} -eq 0 exit $? else exit 0 diff --git a/tests/glade/run_glade_tests.sh b/tests/glade/run_glade_tests.sh index e197180..6c69ad6 100755 --- a/tests/glade/run_glade_tests.sh +++ b/tests/glade/run_glade_tests.sh @@ -22,7 +22,7 @@ fi status=0 for check in ${srcdir}/*/check_*.py ; do - findtestfiles -name '*.glade' | xargs "${check}" "$@" + findtestfiles -name '*.glade' | parallel --gnu -j0 "${check}" "$@" {} if [ "$?" -ne 0 ]; then status=1 fi diff --git a/tests/pylint/pylint-one.sh b/tests/pylint/pylint-one.sh index 6aed441..8cf4248 100755 --- a/tests/pylint/pylint-one.sh +++ b/tests/pylint/pylint-one.sh @@ -12,8 +12,6 @@ if grep -q '# pylint: skip-file' $1; then exit 0 fi -file_suffix="$(eval echo \$$#|sed s?/?_?g)" - pylint_output="$(pylint \ --msg-template='{msg_id}:{line:3d},{column}: {obj}: {msg}' \ -r n --disable=C,R --rcfile=/dev/null \ @@ -33,6 +31,8 @@ gi.overrides.__path__[0:0] = (os.environ["ANACONDA_WIDGETS_OVERRIDES"].split(":" if [ -n "$(echo "$pylint_output" | fgrep -v '************* Module ')" ]; then # Replace the Module line with the actual filename pylint_output="$(echo "$pylint_output" | sed "s|\* Module .*|* Module $(eval echo \$$#)|")" - echo "$pylint_output" > pylint-out_$file_suffix - touch "pylint-$file_suffix-failed" + echo "$pylint_output" + exit 1 +else + exit 0 fi diff --git a/tests/pylint/runpylint.sh b/tests/pylint/runpylint.sh index 3615052..6599254 100755 --- a/tests/pylint/runpylint.sh +++ b/tests/pylint/runpylint.sh @@ -114,27 +114,13 @@ if [ -z "$FILES" ]; then egrep -v '(|/)old_tests/') fi -num_cpus=$(getconf _NPROCESSORS_ONLN) # run pylint in paralel -echo $FILES | xargs --max-procs=$num_cpus -n 1 "$srcdir"/pylint-one.sh $ARGS || exit 1 +output=$(echo -n $FILES | parallel -d' ' --gnu "$srcdir"/pylint-one.sh $ARGS {}) +exit_status=$? -for file in $(find -name 'pylint-out*'); do - cat "$file" >> pylint-log - rm "$file" -done - -fails=$(find -name 'pylint*failed' -print -exec rm '{}' \;) -if [ -z "$fails" ]; then - exit_status=0 -else - exit_status=1 -fi - -if [ -s pylint-log ]; then +if [ "$output" != '\n' -a "$output" != "" ]; then echo "pylint reports the following issues:" - cat pylint-log -elif [ -e pylint-log ]; then - rm pylint-log + echo "$output" | tee pylint-log fi exit "$exit_status" -- 1.9.3

4 4

[master/rhel7] Add platform specific group selection (#884385)
by Brian C. Lane 24 Jul '14

24 Jul '14

Virtualization platforms would like to have their packages installed automatically when running inside them. This uses systemd-detect-virt to identify the platform and creates a group id to be selected by prefixing it with "platform-" so that, for example, running on kvm will result in a group named "platform-kvm" being selected if it exists. --- pyanaconda/iutil.py | 18 ++++++++++++++++++ pyanaconda/packaging/__init__.py | 11 +++++++++++ pyanaconda/packaging/dnfpayload.py | 2 +- pyanaconda/packaging/yumpayload.py | 2 +- 4 files changed, 31 insertions(+), 2 deletions(-) diff --git a/pyanaconda/iutil.py b/pyanaconda/iutil.py index 642e7b0..325270d 100644 --- a/pyanaconda/iutil.py +++ b/pyanaconda/iutil.py @@ -878,3 +878,21 @@ def xprogressive_delay(): while True: yield 0.25*(2**counter) counter += 1 + +def get_platform_groupid(): + """ Return a platform group id string + + This runs systemd-detect-virt and if the result is not 'none' it + prefixes the lower case result with "platform-" for use as a group id. + + :returns: str + """ + try: + platform = execWithCapture("systemd-detect-virt", []).strip() + except (IOError, AttributeError): + return "" + + if platform == "none": + return "" + + return "platform-"+platform.lower() diff --git a/pyanaconda/packaging/__init__.py b/pyanaconda/packaging/__init__.py index c368e0d..d235183 100644 --- a/pyanaconda/packaging/__init__.py +++ b/pyanaconda/packaging/__init__.py @@ -725,6 +725,17 @@ class PackagePayload(Payload): else: self.rpmMacros.append(('__file_context_path', '%{nil}')) + # Add platform specific group + groupid = iutil.get_platform_groupid() + if groupid and groupid in self.groups: + if isinstance(groups, list): + log.info("Adding platform group %s", groupid) + groups.append(groupid) + else: + log.warning("Could not add %s to groups, not a list.", groupid) + elif groupid: + log.warning("Platform group %s not available.", groupid) + @property def kernelPackages(self): if "kernel" in self.data.packages.excludedList: diff --git a/pyanaconda/packaging/dnfpayload.py b/pyanaconda/packaging/dnfpayload.py index 64f7433..4efc524 100644 --- a/pyanaconda/packaging/dnfpayload.py +++ b/pyanaconda/packaging/dnfpayload.py @@ -556,7 +556,7 @@ class DNFPayload(packaging.PackagePayload): return list(gids) def preInstall(self, packages=None, groups=None): - super(DNFPayload, self).preInstall() + super(DNFPayload, self).preInstall(packages, groups) self.requiredPackages = packages self.requiredGroups = groups self.addDriverRepos() diff --git a/pyanaconda/packaging/yumpayload.py b/pyanaconda/packaging/yumpayload.py index 6dca7f6..67f7a02 100644 --- a/pyanaconda/packaging/yumpayload.py +++ b/pyanaconda/packaging/yumpayload.py @@ -1332,7 +1332,7 @@ reposdir=%s def preInstall(self, packages=None, groups=None): """ Perform pre-installation tasks. """ - super(YumPayload, self).preInstall() + super(YumPayload, self).preInstall(packages, groups) progressQ.send_message(_("Starting package installation process")) self.requiredPackages = packages -- 1.9.3

2 3

[f21/master] Solidify the row separator in the welcome spoke.
by David Shea 23 Jul '14

23 Jul '14

--- data/anaconda-gtk.css | 8 ++++++++ pyanaconda/ui/gui/spokes/welcome.glade | 3 +++ 2 files changed, 11 insertions(+) diff --git a/data/anaconda-gtk.css b/data/anaconda-gtk.css index 8d71f60..89ad84a 100644 --- a/data/anaconda-gtk.css +++ b/data/anaconda-gtk.css @@ -190,3 +190,11 @@ AnacondaSpokeSelector:insensitive { color: @anaconda_insensitive_fg_color; border-color: @anaconda_insensitive_borders; } + +/* When multi-column GtkTreeViews set a row separator, the horizontal-sepataor + * is still applied to the row separator, breaking it up for each column. It + * looks kind of dumb. Provide a way to not do that. + */ +GtkTreeView.solid-separator { + -GtkTreeView-horizontal-separator: 0; +} diff --git a/pyanaconda/ui/gui/spokes/welcome.glade b/pyanaconda/ui/gui/spokes/welcome.glade index b2a6cc5..0180e63 100644 --- a/pyanaconda/ui/gui/spokes/welcome.glade +++ b/pyanaconda/ui/gui/spokes/welcome.glade @@ -316,6 +316,9 @@ <property name="AtkObject::accessible-name" translatable="yes">Languages</property> </object> </child> + <style> + <class name="solid-separator"/> + </style> </object> </child> </object> -- 2.0.0

2 1

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

anaconda-patches July 2014