I marked toolbox as failable when we were getting everthing working on
it (and so we would still have composes). Now that it's working, lets
mark it blocking again.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
This section is for released fedoras with updates/base repo,
for rawhide, we just pass pungi "Everything" and it does the magic.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
There's a Fedora 39 Change to increase the size of the EFI
system partition:
https://fedoraproject.org/wiki/Changes/BiggerESP
it goes from a size range of 200-600MiB to a size range of
500MiB-2GiB. The way these aarch64 disk images are built, they
get the biggest ESP possible, so they're getting a 2GiB ESP,
up from 600MiB. For these three images, that results in anaconda
calculating there isn't enough space on the root partition to
install the package set, and the image creation fails.
Per https://bugzilla.redhat.com/show_bug.cgi?id=2212121
there's at least a chance this is really how we want things, so
for now, let's increase the sizes of these images to accommodate
the bigger ESP. We can revert this if we decide that after all
the larger ESP is not needed and we want to tweak that somehow.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
Now that the Fedora + Azure change was approved, let's get Azure images
building in koji alongside the existing qcow, raw.xz, and vagrant
images.
Signed-off-by: Major Hayden <major@redhat.com>
As noted by Neal in the previous PR[0], we need UEFI-only images for
Azure. Switch to the `Fedora-30` profile to align with the existing
UEFI-only image configuration for GCP images.
[0] https://pagure.io/pungi-fedora/pull-request/1161
Signed-off-by: Major Hayden <major@redhat.com>
These files have recently been updated and their contents is now
available in a new file.
Signed-off-by: Tulio Magno Quites Machado Filho <tuliom@redhat.com>
The Cloud SIG is ready to launch Fedora on Azure and we need image built
along with rawhide composes. Although the image has a very similar
package set, it must be in VHD format.
Signed-off-by: Major Hayden <major@redhat.com>
Let's fix all the warnings thrown by pungi-config-validate. Only the
general.conf is throwing errors, but because it couldn't be validated by itself alone.
Signed-off-by: Michal Konečný <mkonecny@redhat.com>
To prevent the need to edit one thing in multiple pungi configurations let's use
it's ability to import other configuration files (See
https://docs.pagure.org/pungi/format.html#importing-other-files). This change
will create a new configuration file `general.conf` which is imported to
each other pungi configuration for Fedora.
Signed-off-by: Michal Konečný <mkonecny@redhat.com>
Cheese flatpak had a long standing issue with camera detection, which is
now finally fixed. Now that it's working again, add it to Silverblue to
keep the list of installed apps in sync with Workstation.
Signed-off-by: Kalev Lember <klember@redhat.com>
httpd from being treated as multilib, since the conflicting
i686/x86_64 packages were moved to httpd-core.
Signed-off-by: Joe Orton <jorton@redhat.com>
Recent Rawhide composes failed because they need 1M extra space
on the disk:
https://pagure.io/releng/failed-composes/issue/4478
let's be generous and give it a whole extra 1G!
Signed-off-by: Adam Williamson <awilliam@redhat.com>
There are currently no KDE applications packaged as Flatpaks in Fedora
and we are thus shipping no Fatpak by default in Fedora Kinoite.
Remove the Runtime as it is unused and will be installed on demand if
users install applications from the Fedora Flatpak remote.
This will make the installer smaller which will save download and
installation time for users that only get Flatpaks from Flathub.
This can be reverted (and fixed for aarch64) once we have KDE Flatpaks
in Fedora.
See: https://pagure.io/fedora-kde/SIG/issue/13
Signed-off-by: Timothée Ravier <tim@siosm.fr>
This uses the syntax after my recent PR:
https://pagure.io/releng/pull-request/11000
with those changes, this should actually work, instead of just
printing sad error messages like it does now!
Signed-off-by: Adam Williamson <awilliam@redhat.com>
Move it to here instead of having systemd-devel fake require
systemd-pam.
This should help with flatpak runtime packaging so that we can avoid
having to ship systemd-pam in the flatpak container.
Signed-off-by: Kalev Lember <klember@redhat.com>
Wine subpackages were probably covered by dependencies. But separate packages
like `wine-dxvk` were not covered, even though they need to be multilib. At the
same time, we don't want to use `wine*` because that would also cover e.g.
`wineglass`, which doesn't need to be multilib. The current approach seems to
work the best.
Signed-off-by: Kamil Páral <kparal@redhat.com>
This has been done in Workstation, but not yet in Silverblue.
This commit is a great example of why putting lots of items in
a single line of text is a bad idea!
Signed-off-by: Adam Williamson <awilliam@redhat.com>
The packages depends on lld, which is not shipped there. This is a
followup to #1036 now that we have `wasm32-wasi`, and we might want
`wasm64` variants in the future.
Signed-off-by: Josh Stone <jistone@redhat.com>
We now have the flatpak runtime available for aarch64, which was the
last missing piece to be able to install Fedora flatpaks on aarch64.
Signed-off-by: Kalev Lember <klember@redhat.com>
The compose ID and some paths have an extra bogus comma at the end. That
makes it difficult to consume the message.
Signed-off-by: Lubomír Sedlář <lsedlar@redhat.com>
In 57985b08 nightly was updated, but not all the variants were updated
in the second sync. The first sync runs with --exclude repodata, then
the second one includes that and also does a --delete-after, but we only
updated the first one. This should clean up old artifacts.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
This update adds the various pieces for the UEFI side of things
and removes the old appliance build process including a few cleanups.
It also updates the sync process and cleans up a few bits here
while at it.
Signed-off-by: Peter Robinson <pbrobinson@gmail.com>
aarch64 isn't hardlinked with the rest of primary
rawhide and branched aren't hardlinked together
* other bits more than likely aren't hardlinked.
So, lets try and hardlink content after we sync it here.
This may be too i/o intensive or slow to keep, but we can
try it for a few days and see how bad it is.
Signed-off-by: Kevin Fenzi kevin@scrye.com
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
This is to fix the composes sync to PDC as both the live_media and
live_images have same metadata and that is causing the sync issues
Signed-off-by: Mohan Boddu <mboddu@bhujji.com>
We really need to figure out whats grown here, but in the mean time we
just need to get a compose out, so increase the size here.
The aarch64 one went past 4gb.
see: https://koji.fedoraproject.org/koji/taskinfo?taskID=56316220
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
It was made failable in #382, back in 2017, as part of the
Modular Server effort that was eventually abandoned, with the
idea being that Modular Server would be blocking instead. But
when Modular Server was abandoned, we never reverted this change
to make Server non-failable again...
Signed-off-by: Adam Williamson <awilliam@redhat.com>
When not specified, subvariant defaults to the variant. This leads to
both armhfp appliance and image_build based raw.xz image to have
identical metadata. A file with such issue can not be loaded by
productmd parser.
Due to another issue (that should be fixed in next Pungi release), the
paths to image directories are not listed in composeinfo.json file. The
compose-partial-copy script therefore has to find out if images should
be copied from the image metadata. And since there's an error in it, it
doesn't see the images and doesn't copy them.
Signed-off-by: Lubomír Sedlář <lsedlar@redhat.com>
This reverts commit 4c5ce9fb5d.
This is a one time thing to test it out. We will add it back once the
change is approved
Signed-off-by: Mohan Boddu <mboddu@bhujji.com>
This will reduce both the BOOT.iso and the DVD medium size
This modification relates to the following change proposal:
https://fedoraproject.org/wiki/Category:Changes/OptimizeSquashFS
The new option should work for Pungi version 4.24 and above
Jira: RHELCMP-693
Signed-off-by: Bohdan Khomutskyi <bkhomuts@redhat.com>
It turns out koji doesn't know about `gce` even though ImageFactory
does. Really all we need is the output disk image to be a tar.gz so
let's try with the `tar-gz` format in koji, which looks like it was
intended for use with GCP.
Signed-off-by: Dusty Mabe <dusty@dustymabe.com>
We added this so we could eventually use the VMDK to optimize our
uploads to AWS but it turns out the VMDK created by imagefactory
isn't in the right format anyway.
```
SNAPSHOTTASKDETAIL 0.0 VMDK deleted ClientError: Disk validation failed [Unsupported VMDK File Format]
```
Let's drop them for now.
Signed-off-by: Dusty Mabe <dusty@dustymabe.com>
Usage of this option should reduce both the BOOT.iso and the DVD medium size.
This is because the LiveOS/squashfs.img will be compressed better by using
a bigger dictionary size. 1MiB is the maximum allowed for SquashFS currently.
This modification relates to the following change proposal:
https://fedoraproject.org/wiki/Category:Changes/OptimizeSquashFS
The result of it should be similar as in the following example DVD:
https://khomutsky.com/fedora-dvd/Fedora-Workstation-x86_64-31-1.9.bkhomuts-squashfs-ext4-xz-1M.iso
I have tested the above mentioned DVD myself to confirm that the installation
succeeds. You can see other test results in the change proposal.
Note, that this is only one part of the change proposal.
A second part is documented at this URI: https://pagure.io/pungi-fedora/pull-request/871
The new option requires Pungi 4.2.4 that will be released soon.
Jira: RHELCMP-693
Signed-off-by: Bohdan Khomutskyi <bkhomuts@redhat.com>
This will reduce both the BOOT.iso and the DVD medium size
This modification relates to the following change proposal:
https://fedoraproject.org/wiki/Category:Changes/OptimizeSquashFS
The new option should work for Pungi version 4.24 and above
Jira: RHELCMP-693
Signed-off-by: Bohdan Khomutskyi <bkhomuts@redhat.com>
This will reduce both the BOOT.iso and the DVD medium size
This modification relates to the following change proposal:
https://fedoraproject.org/wiki/Category:Changes/OptimizeSquashFS
The new option should work for Pungi version 4.24 and above
Jira: RHELCMP-693
Signed-off-by: Bohdan Khomutskyi <bkhomuts@redhat.com>
The flatpak embedding process needs to run a local DBus session so that
flatpak can talk to flatpak-oci-authenticator to authenticate (no-op)
to registry.fedoraproject.org.
https://pagure.io/fedora-lorax-templates/pull-request/47
Signed-off-by: Owen W. Taylor <otaylor@fishsoup.net>
Since they had to be seperated, koji tried to generate two images of
same name and it failed. We decided to revert it and check with pungi
folks for a better solution.
https://pagure.io/releng/failed-composes/issue/1372
Revert "Removing ppc64le and s390x vmdk cloud images"
This reverts commit 46b1711d1d.
Signed-off-by: Mohan Boddu <mboddu@bhujji.com>
See https://github.com/ostreedev/ostree/pull/1984 where OSTree was
changed to try to set group write on directories. We need to set a more
permissive umask to take advantage of it.
This is all done to support writing to the OSTree repo from OpenShift
projects that run as a random UID, but have ftpsync (gid:263) in their
supplemental groups. For more context see: https://pagure.io/releng/issue/8811#comment-629051
Signed-off-by: Mohan Boddu <mboddu@bhujji.com>
WARNING: Config option release_is_layered was removed and has no effect;
remove it. It's layered if there's configuration for base product.
WARNING: Config option runroot was removed and has no effect; remove it.
Please specify 'runroot_method' if you want to enable runroot, otherwise run
things locally.
WARNING: Config option bootable was removed and has no effect; remove it.
Setting buildinstall_method option if you want a bootable installer.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
Switch from F30 runtime to F31 as all the apps that are preinstalled
have now been rebuilt against the new runtime.
Signed-off-by: Kalev Lember <klember@redhat.com>
We are now seeing aarch64 and x86_64 failures with the default 2gb.
Ideally one of the minimzation efforts could help us here.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
aarch64 cloud is blocking, but it's hitting:
https://bugzilla.redhat.com/show_bug.cgi?id=1722181
and failing the composes. Due to other issues we haven't had composes
for a while before that either, so it's getting to be a long time.
Hopefully some workaround will at least appear in a few days.
If not, we can revert this and start failing again for a while.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
The koji runroot used for to create the OSTree installer has no access
to external networks, so use a special hostname for registry.fedoraproject.org
that suppresses the redirects to the CDN that would normally happen for
blob downloads.
Signed-off-by: Owen W. Taylor <otaylor@fishsoup.net>
Use a newly added Pungi option (https://pagure.io/pungi/pull-request/1278),
so that flatpak is in the Silverblue ostree-installer runroot, which is
needed for embedding Flatpaks in the Silverblue image.
Signed-off-by: Owen W. Taylor <otaylor@fishsoup.net>
The last change set the template variables for embedding Flatpaks,
but didn't actually include the necessary template.
Signed-off-by: Owen W. Taylor <otaylor@fishsoup.net>
A long-term goal of Silverblue has been moving applications that
were previously part of the Workstation image into Flatpaks.
At this point, we many of the applications that were removed from
Silverblue available as Fedora Flatpaks, and we have the necessary
Anaconda support, so we should update the Silverblue images to include
the runtime and applications.
This is currently x86_64 only, since we don't have a working OSBS
cluster to build containers for aarch64 or ppc64le.
Note that Fedora Flatpaks are single-stream, so this embeds the
F30 runtime and applications into the Rawhide installer image. When
the applications are released for F31, the embedded runtime will
need to be updated.
Requires: https://pagure.io/fedora-lorax-templates/pull-request/40
Signed-off-by: Owen W. Taylor <otaylor@fishsoup.net>
Partly revert commit e25f2baaa1
instead of pulling in dracut* (dracut and all it's subpackages), just
pull in the main one. This should hopefully fix
https://bugzilla.redhat.com/show_bug.cgi?id=1734179
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
Similar to what we did for Cloud (6eecb53) we need to stop building
Workstation repos so we can speed up our compose times (see [1]).
We already switched to sourcing from the Everything repo (39683dd).
Now we just need to stop building the repo. This commit makes it so
that we will no longer build the yum repo for Workstation.
[1] https://pagure.io/releng/issue/7403
Signed-off-by: Dusty Mabe <dusty@dustymabe.com>
For some reason, the DVD is being produced with all of the content
from this compose, rather than the smaller subset listed in
fedora-disk-server-kickstart.ks, so we should reduce this compose
to match the intended contents of the DVD. Network-enabled
installations will still use the Everything repo.
Signed-off-by: Stephen Gallagher <sgallagh@redhat.com>
Should fix:
WARNING: Config option pdc_url was removed and has no effect; Koji is
queried instead.
WARNING: Config option pdc_develop was removed and has no effect; Koji
is queried instead.
WARNING: Config option pdc_insecure was removed and has no effect; Koji
is queried instead.
in rawhide composes.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
Certain applications are using the container env variable to
know if they are running in a container or not. This is making sure
we have this define in the fedora base image..
Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1399507
Signed-off-by: Clement Verna <cverna@tutanota.com>
2019-07-10 20:53:23 +02:00
10 changed files with 900 additions and 607 deletions
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
