Running build of Fedora CoreOS 35.20220116.3.0 ends in emergency mode
pawcykca opened this issue · comments
Bug Report
Environment
What operating system is being used to run coreos-assembler?
Fedora CoreOS 35.20220116.3.0
What operating system is being assembled?
Fedora CoreOS 35.20220116.3.0
Is coreos-assembler running in Podman or Docker?
Podman
If Podman, is coreos-assembler running privileged or unprivileged?
privileged
Expected Behavior
Running image by 'cosa run -c' don't end in emergency mode.
Actual Behavior
Running image by 'cosa run -c' ends in emergency mode.
Entering emergency mode. Exit the shell to continue.
Type "journalctl" to view system logs.
You might want to save "/run/initramfs/rdsosreport.txt" to a USB stick or /boot
after mounting them and attach it to a bug report.
Press Enter for maintenance
(or press Control-D to continue):
Possible root cause:
:/root# journalctl -u afterburn-network-kargs.service --no-pager
-- Journal begins at Thu 2023-07-20 11:13:23 UTC, ends at Thu 2023-07-20 11:13:24 UTC. --
Jul 20 11:13:23 localhost afterburn[215]: Error: error: The argument '--default-value <default-value>' requires a value but none was supplied
Jul 20 11:13:23 localhost afterburn[215]: USAGE:
Jul 20 11:13:23 localhost afterburn[215]: afterburn exp rd-network-kargs --cmdline --default-value <default-value>
Jul 20 11:13:23 localhost afterburn[215]: For more information try --help
:/root# journalctl -u ostree-prepare-root --no-pager
-- Journal begins at Thu 2023-07-20 11:13:23 UTC, ends at Thu 2023-07-20 11:13:24 UTC. --
Jul 20 11:13:23 localhost ostree-prepare-root[218]: preparing sysroot at /sysroot
Jul 20 11:13:23 localhost ostree-prepare-root[218]: ostree-prepare-root: Couldn't find specified OSTree root '/sysroot//ostree/boot.1/fedora-coreos/454a550751511c63f4c154a96d15069e716f79c6f5dc804d5206835bbf567443/0': No such file or directory
Reproduction Steps
- export COREOS_ASSEMBLER_CONTAINER=quay.io/coreos-assembler/coreos-assembler:v0.1X.0
- mkdir pku-35.20220116.3.0
- cd pku-35.20220116.3.0
- cosa init --commit 8ee7600e41a0da41cda0576845f8a0b03f2d37e5 https://github.com/coreos/fedora-coreos-config
- sed -i 's/latest/4629653/' src/config/fedora-coreos-pool.repo
- cosa fetch
- cosa build
- cosa run -c
Other Information
This works for me few weeks ago - I was able to build Fedora CoreOS 35.20220116.3.0 and run image without errors.
COSA image tag and FOSA Commit are checked on this site: https://builds.coreos.fedoraproject.org/browser?stream=stable&arch=x86_64
Since yesterday all attempts end in emergency mode when I run the built image,
Build of next version of Fedora CoreOS 35.20220131.3.0 works.
Hey @pawcykca - most likely the old versions of the coreos-pool repo are getting garbage collected. Fedora 35 is End Of Life so you probably won't find anyone that spends time to try to make this work. Sorry.
The reported issue was not a lack of packages in coreos-pool repo needed to build the image, but the incorrect work of the built image.
In that case I have no idea. The package set should be the same since there are lockfiles in the repo, but it's hard to know what version of COSA was used to build stable back then (I mean the commit ID of cosa is there, but the pushed container is garbage collected).
I doubt we would be able to enable you to do what you were doing (building F35 FCOS), but if you explain what your goal is we might be able to help with that.
Basically, I wanted to understand what I'm doing wrong, if building this image was working fine just a few days ago, and now without any modifications to the build environment it doesn't work.
Building images works on newer versions of FCOS, so it may not make sense to delve further into this topic.
Assuming your cosa image is the same as it was a few days ago, I don't know.
I'm surprised you found an old distrepo version to pull from, we cleaned up the f35 packages some time ago. Those could get cleaned up at any time I would say.