FBut the original issue is that cosmic images don't have
policy.json somehow
ICan confirm. RUles are present at location.
RSo it is just files that should end up in /etc that are missing?
ACheck permissions
ANothing should be missing
Idownstream issue is that at secureblue the comsic image variant does not build anymore because policy.json is missing. We suspected that ublue-os:config is not included. Currently it looks like the cosmic variant does not included this commit https://github.com/ublue-os/config/commit/bfd1b6836006ef2c83c2dd3949f6e8d04bc0354d
IChasing the issue took us to down the rabbit hole of my PR and one Issue report.
KPrior to this commit config failed to build
KSo everyone should have policy unless config was never included
KSince the previous successful build matches this
Rstrange, the secureblue cosmic build wasn't failing a few days ago
Rso something changed 
A@M2 @Kyle Gospo differences in initramfs
Amessage.txt11.48KB
Ain two random versions
Awith the same kernel
Alsinitrd --unpack 6.9.4-201.2.initramfs.img to unpackAsudo diff --brief --recursive --no-dereference 1 2 to compare
MI think the modules are biting us somehow
MLooks like packages other than the kernel are updating
Aif you have fast internet
Asudo podman mount $(sudo podman create ghcr.io/ublue-os/bazzite-deck:40-20240616) Ayou can pull 2 testing builds from the same day
Aand check if the initramfs is the same
Aalthough i think
Arebuilding akmods will invalidate it
Aexcept the gpu one, dunno why that changed
MGPU for AMD is from the image upstream image/kernel modules packages + linux-firmware packages
MIt appears that all of the config rpms are already installed
M
KSigning is the one with that file
Myeah, but they were saying udev rules weren't there as well
IWe were wrong. It turned out that the udev rules are present. The build fails because of a missing policy.json file. Current ublue-os:config has a commit that fixes thew issue. The cosmic image does not seem to have this commit included atm.
IState few hours ago. We are waiting on a rebuild of the cosmic image
J
Ithank you. The base image errored out at the uploading step. Can you rebuild it again, please?
Jyep, watching it
M@antheas another error with rechunk popped up this time with ucore
https://github.com/m2Giles/m2os/actions/runs/10335823416/job/28610865502#step:18:615
https://github.com/m2Giles/m2os/actions/runs/10335823416/job/28610865502#step:18:615
MThere is no previous image for that image.
MThink I figured out the cause. You put GCC in the same layer and it's not on image yet
MNope not it.
Mlooks like its gcc just failing to install correctly
Mokay. I'm guessing its somehow having a mixed subpackages of gcc causing the issue
IWe were able to fix our issue. /etc/ instead of /usr/etc/ in a script and it builds. I am uncertain if this change was introduced with rechunks. Thx
ALatest ver woes I think
AActually dunno
ACan't see the error from my phone I'll look at it in a bit
Aill look into it
Apulled the image and i get the error
AGitHub
Fixes a bug in the prefill algorithm that caused some packages to not be allocated and a subsequent crash. In addition, makes it so if this happens again due to a different bug an error is emitted ...
