stable

fwupd-1.8.12-1.fc38, libjcat-0.1.13-1.fc38, & 1 more

FEDORA-2023-f6afa6f9e5 created by rhughes a year ago for Fedora 38

This release adds the following features:

  • Add support for replaying USB devices so they can be emulated in CI
  • Allow desktop software to inhibit the system to prevent updates
  • Allow using requirements with depth=0 and no parent
  • Auto-set the CCGX remove-delay now we parse DMC subcomponents
  • Detect and warn users with the broken NVMe firmware 3B2QGXA7
  • Print errors as JSON objects when using fwupdmgr --json

This release fixes the following bugs:

  • Allow installing battery firmware updates even when the power is too low
  • Allow setting the package user agent before the client has connected
  • Correctly fall back to the compatible vendor when FDT vendor is missing
  • Detect CCGX factory mode and set a non-zero version
  • Detect fixed Insyde firmware that can actually use Capsule-on-Disk
  • Do not make any of the HWIDs setup failures fatal
  • Fix a critical warning when parsing an empty kernel cmdline
  • Fix a small memory leak when installing TPS6598x firmware
  • Fix compiling with -Dbuild=library for Flathub
  • Fix fwupdtool firmware-convert to work with image-less formats
  • Fix regression in downloading files in fwupdtool
  • Fix SMBIOS struct parsing when the tag section ends with NUL
  • Indicate HSI attributes that will only be returned for specific CPU vendors
  • Only offset the IPMI user ID when using Lenovo XCC
  • Prefer the Intel USB4 plugin over the Thunderbolt plugin when required
  • Require at least twice the capsule size in the ESP when updating
  • Save all the device flags in the pending database correctly
  • Set the device percentage and status for the duration of the update
  • Show the 4XX download failure in the CLI error output
  • Speed up regenerating the MOTD when installing composite devices
  • Use an updated shim if provided during for capsule update
  • Use strict snap confinement

This release adds support for the following hardware:

  • CalDigit Element Hub
  • CalDigit TS4 Dock

How to install

Updates may require up to 24 hours to propagate to mirrors. If the following command doesn't work, please retry later:

sudo dnf upgrade --refresh --advisory=FEDORA-2023-f6afa6f9e5

This update has been submitted for testing by rhughes.

a year ago

This update's test gating status has been changed to 'waiting'.

a year ago

This update's test gating status has been changed to 'passed'.

a year ago

This update has been pushed to testing.

a year ago

So there's one very odd failure happening for this update (in a non-gating test, which is why the gating status is passed). The test:

  1. Builds a Silverblue ostree, with packages from the update included if relevant
  2. Builds a DVD installer image with the ostree embedded, again with packages from the update included in the installer environment if relevant
  3. Installs from the image and boots
  4. Rebases to Fedora 37, reboots, checks it reached Fedora 37
  5. Rolls back the rebase to return to Fedora 38, reboots, checks it reached Fedora 38 again

For some reason, this test is consistently failing in step 5 for this update. After the rollback command succeeds, it tries to type systemctl reboot to reboot, but it doesn't work: nothing appears on screen (it's at a VT) and the system doesn't reboot.

This is only failing on this update - not on the same test run on updates before and after this one - and it's failing repeatedly, across multiple runs on both prod and staging openQA. But I can't for the life of me see how an F38 update to fwupd, libjcat and plasma-discover could have that effect. It's very bizarre. I'll see if I can manually reproduce the issue and, if so, if I can figure anything out about it, but I'm on PTO till Wednesday so wanted to give a heads-up about it first. If you want to try, you can download the installer image at https://openqa.fedoraproject.org/tests/1783388/asset/iso/01783388-FEDORA-2023-f6afa6f9e5-Silverblue-ostree-x86_64.iso (for now, it'll get garbage collected quite soon).

User Icon adamwill commented & provided feedback a year ago
karma

Aside from the weirdness above, which I can't reproduce in manual testing and have decided to just workaround in openQA, this seems to be working fine locally.

This update can be pushed to stable now if the maintainer wishes

a year ago

This update has been submitted for stable by bodhi.

a year ago

This update has been pushed to stable.

a year ago

Please login to add feedback.

Metadata
Type
bugfix
Severity
medium
Karma
1
Signed
Content Type
RPM
Test Gating
Settings
Unstable by Karma
-3
Stable by Karma
3
Stable by Time
14 days
Dates
submitted
a year ago
in testing
a year ago
in stable
a year ago
approved
a year ago

Automated Test Results