Hi folks! The Fedora 26 Beta freeze is fast approaching (it's 2017-05-
16), so it's time for a blocker status mail.
tl;dr action summary
1. https://bugzilla.redhat.com/show_bug.cgi?id=1448923
ACTION: Preferably, Firefox maintainer (mstransky) to fix build on ARM
2. https://bugzilla.redhat.com/show_bug.cgi?id=1443415
ACTION: dnf team to investigate problem and come up with a fix
3. https://bugzilla.redhat.com/show_bug.cgi?id=1446432
ACTION: QA to build a test image and verify fix
4. https://bugzilla.redhat.com/show_bug.cgi?id=144
ACTION: dracut maintainer (harald) to review patch, merge, and build
5. https://bugzilla.redhat.com/show_bug.cgi?id=1403352
ACTION: QA to run upgrade test and verify all is good
6. https://bugzilla.redhat.com/show_bug.cgi?id=1443206
ACTION: GNOME team (fmuellner) to investigate and fix remaining crashes
(#1446879...)
7. https://bugzilla.redhat.com/show_bug.cgi?id=1438046
ACTION: maintainer (mkolman) to submit update, QA to verify
8. https://bugzilla.redhat.com/show_bug.cgi?id=1348688
ACTION: lvm/blivet folks (prajnoha / vpodzime) to investigate and fix
9. https://bugzilla.redhat.com/show_bug.cgi?id=1445302
ACTION: QA to build test image and verify fix
Bug-by-bug detail
=
1. https://bugzilla.redhat.com/show_bug.cgi?id=1448923 - comps - NEW
Xfce ARM image default browser is unavailable (image fails to build)
There is apparently a build problem with Firefox 53 on ARM:
https://bugzilla.redhat.com/show_bug.cgi?id=1443938
and the maintainer decided to just go ahead and disable ARM and
build 53. The result of this is that all ARM images with Firefox
in them fail to compose, including the Xfce image, which is a
release-blocking image. We really want Firefox fixed, here, but
*technically* the blocker could be resolved by shipping a different
default browser. No-one thinks that's a good idea, though.
2. https://bugzilla.redhat.com/show_bug.cgi?id=1443415 - dnf - NEW
[TRACKING] Upgrade f25 to f26 get stuck in Cleanup
This was reported against ppc64le (which is not a release-blocking
arch) but was reported by pwhalen to also occur on armhfp (which is).
Not sure where we stand on identifying/fixing it at present. FWIW,
I have not seen this hit the openQA upgrade tests on i386/x86_64.
3. https://bugzilla.redhat.com/show_bug.cgi?id=1446432 - dnf - MODIFIED
Minimal install option install full gnome desktop
This is a problem with package selection during install (changing
the default package group results in the default group *and* the
newly-selected group both being installed). A fix is just now
heading to updates-testing, so QA will need to build an image and
check that this is fixed.
4. https://bugzilla.redhat.com/show_bug.cgi?id=144 - dracut - POST
Since curl-7.53.1-7.fc26 , curl TLS transactions in initramfs environment
are broken
A change in how curl handles CA trust resulted in TLS transactions
using curl not working from the initramfs environment. This mainly
affects the installer (it means you can't use kickstarts or updates
images from HTTPS servers). I've submitted a fix for this, it's
waiting on dracut maintainer review and merge (and package build).
5. https://bugzilla.redhat.com/show_bug.cgi?id=1403352 - freeipa - ON_QA
FreeIPA server install fails (and existing servers probably fail to start)...
This is very likely resolved, but we're leaving it open because we
still haven't got around to testing that an upgrade of an existing
FreeIPA server from Fedora 25 to Fedora 26 works correctly. I will
get on that next week.
6. https://bugzilla.redhat.com/show_bug.cgi?id=1443206 - libgweather - ON_QA
gnome-shell consistently crashes in the middle of first-login
gnome-initial-setup
The initial crash here actually has been diagnosed and fixed, but
the update was rejected because multiple testers saw a different
crash. That crash is filed as
https://bugzilla.redhat.com/show_bug.cgi?id=1446879, and we are now
waiting on a fix for *that* crash from the GNOME team, so an update
can go out that will pass testing.
7. https://bugzilla.redhat.com/show_bug.cgi?id=1438046 - initial-setup - POST
initial-setup.service: Failed to set up stdin: Inappropriate ioctl for device
A fix for this has just showed up today, and a tester verified that
a scratch build appears to work. We now need mkolman to send out a
proper build/update.
8. https://bugzilla.redhat.com/show_bug.cgi?id=1348688 - lvm2 - ASSIGNED
Anaconda cannot access LVM partitions in a LUKS-encrypted disk partition
after decryption
This is about accessing a particular combined LVM/LUKS partitioning
setup. We are waiting for a fix from the maintainers, here. A clear
explanation of the offending setup is in comment #39. I am not sure
if this is really an lvm2 issue or a blivet issue, so CCing both
maintainers.
9. https://bugzilla.redhat.com/show_bug.cgi?id=1445302 - python-blivet - ON_QA
partition on a FW RAID used as a PV is put into a list of filtered
