If you’re running the F33 prerelease, you should know that the build from Sept 15 (33.20200915.n.0 (2020-09-15T21:09:07Z), basecommit 32b2b7d9f0d681f4b973eca8e84778b2b0f6709fdf06180ad2614e9de2fdcadd) fails to boot.
Here’s my experience dealing with this (perhaps useful for others to see?):
My boot process hung at the splash screen, so I rebooted and hit escape (after decrypting the drives) to reveal the boot messages:
I rebooted and held down shift to get the menu and selected my previous (successful) deployment, 33.20200913.n.0 (2020-09-13T07:59:41Z), then did an
rpm-ostree reset and
rpm-ostree upgrade to get a “clean” version of 33.20200915.n.0 (2020-09-15T21:09:07Z) without overlays. The boot still failed, but with different failing processes.
At this point, I realize that there’s nothing I can do but wait for a fix… and, since I’m using Silverblue, I can at least roll back to an old version.
I rebooted my laptop and held shift again to boot 33.20200913.n.0 (base commit 7de80b1a27968e257a31180f4ae046f5e7acc3c6fd978b90122a5551320b5db9).
BTW: It seems this build has failed several tests, but still went out as an update for us anyway. I know this is a (pre)beta, but aren’t we still supposed to have test gating for things like this?
For reference, here’s a copy/paste from the failed tests on https://www.happyassassin.net/nightlies.html :
- openQA: 667299 (install_default_upload)
- openQA: 667300 (desktop_terminal)
- openQA: 667301 (desktop_browser)
- openQA: 667302 (install_default)
- openQA: 667303 (release_identification)
- openQA: 667304 (base_reboot_unmount)
- openQA: 667305 (base_system_logging)
- openQA: 667306 (base_services_start)
- openQA: 667307 (base_selinux)
- openQA: 667308 (base_service_manipulation)
- openQA: 667309 (desktop_background)