upgrade to 22.04 on raspberry pi can leave ethernet interface unreachable if vlans are used

Bug #1973485 reported by Tim Sedlmeyer
12
This bug affects 2 people
Affects Status Importance Assigned to Milestone
linux-raspi (Ubuntu)
Fix Released
Undecided
Unassigned
Impish
Won't Fix
Undecided
Unassigned
Jammy
Fix Released
Undecided
Unassigned

Bug Description

[Impact]

The 8021q kernel module was moved from the linux-modules package to the linux-modules-extras package for the Raspberry Pi. This means that vlans cannot be configured if the linux-modules-extras package is not installed. The release notes mention that some modules have been moved and to install the linux-modules-extra package if your application doesn't work. This would be adequate warning except for one issue. Systemd-networkd will fail to configure an ip address for the raspberry pi's ethernet interface when the vlans cannot be configured leaving headless systems inaccessible. I suffered this exact scenario upon upgrading since my netplan file included vlans assigned to the Pi's ethernet interface. I would recommend either updating the release notes to specifically call out this scenario or move the 8021q module back to the linux-modules package.

[Test Case]

See above.

[Fix]

Move 8021q modules to linux-modules package.

[Where Problems Could Occur]

None expected.

Juerg Haefliger (juergh)
tags: added: kern-3290
Juerg Haefliger (juergh)
description: updated
Changed in linux-raspi (Ubuntu):
status: New → Invalid
Changed in linux-raspi (Ubuntu Jammy):
status: New → In Progress
Changed in linux-raspi (Ubuntu Impish):
status: New → In Progress
Revision history for this message
Ubuntu Kernel Bot (ubuntu-kernel-bot) wrote :

This bug is awaiting verification that the linux-raspi/5.15.0-1012.14 kernel in -proposed solves the problem. Please test the kernel and update this bug with the results. If the problem is solved, change the tag 'verification-needed-jammy' to 'verification-done-jammy'. If the problem still exists, change the tag 'verification-needed-jammy' to 'verification-failed-jammy'.

If verification is not done by 5 working days from today, this fix will be dropped from the source code, and this bug will be closed.

See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you!

tags: added: verification-needed-jammy
Revision history for this message
Launchpad Janitor (janitor) wrote :
Download full text (46.0 KiB)

This bug was fixed in the package linux-raspi - 5.15.0-1012.14

---------------
linux-raspi (5.15.0-1012.14) jammy; urgency=medium

  * jammy/linux-raspi: 5.15.0-1012.14 -proposed tracker (LP: #1979421)

  * Packaging resync (LP: #1786013)
    - debian/dkms-versions -- update from kernel-versions (main/2022.06.20)

  * jammy/linux-raspi: Update to upstream raspberrypi rpi-5.15.y (2022-05-24)
    (LP: #1975599)
    - drm/atomic-helpers: remove legacy_cursor_update hacks
    - Revert "drm/vc4: hvs: Defer dlist slots deallocation"
    - drm/vc4_hdmi: Force modeset when bpc changes
    - media: i2c: imx258: Remove unused defines
    - media: i2c: imx258: Make image geometry meet sensor requirements
    - media: i2c: imx258: Disable digital cropping on binned modes
    - media: i2c: imx258: Implement HFLIP and VFLIP controls.
    - media: i2c: imx258: Remove redundant I2C writes.
    - media: i2c: imx258: Add regulator control
    - media: i2c: imx258: Make V4L2_CID_VBLANK configurable.
    - media: i2c: imx258: Register the ctrls from fwnode properties
    - media: i2c: imx258: Add support for 24MHz clock
    - media: i2c: imx258: Add support for running on 2 CSI data lanes
    - media: i2c: imx258: Follow normal V4L2 behaviours for clipping exposure
    - media: i2c: imx258: Add get_selection for pixel array information
    - media: i2c: imx258: Allow configuration of clock lane behaviour
    - media: i2c: imx258: Correct max FRM_LENGTH_LINES value
    - media: i2c: imx258: Add support for long exposure modes
    - media: i2c: imx258: Issue reset before starting streaming
    - media: i2c: imx258: Set pixel_rate range to the same as the value
    - media: i2c: imx258: Support faster pixel rate on binned modes
    - defconfigs: Add IMX258 to the Pi defconfigs
    - dtoverlays: Add overlay for Sony IMX258 image sensor
    - Add support for the AudioInjector.net bare i2s sound card
    - vc4_hdmi: Remove VC4_HDMI_SCHEDULER_CONTROL_VERT_ALWAYS_KEEPOUT
    - configs: bcm2711_defconfig: Enable KASLR
    - config: Enable CONFIG_MODULE_COMPRESS_XZ
    - overlays: mipi-dbi-spi: width-mm and height-mm are mandatory
    - configs: (Re)Enable CONFIG_IR_TOY
    - config: Enable the NFT_SYNPROXY module
    - clk: Introduce Kunit Tests for the framework
    - clk: Always clamp the rounded rate
    - clk: Use clamp instead of open-coding our own
    - clk: Always set the rate on clk_set_range_rate
    - clk: Add clk_drop_range
    - clk: test: Test clk_set_rate_range on orphan mux
    - clk: Drop the rate range on clk_put()
    - clk: tests: Add test suites description
    - clk: tests: Add reference to the orphan mux bug report
    - clk: tests: Add tests for uncached clock
    - clk: tests: Add tests for single parent mux
    - clk: tests: Add tests for mux with multiple parents
    - clk: tests: Add some tests for orphan with multiple parents
    - clk: Take into account uncached clocks in clk_set_rate_range()
    - clk: Fix clk_get_parent() documentation
    - clk: Set req_rate on reparenting
    - clk: Skip set_rate_range if our clock is orphan
    - clk: Add our request boundaries in clk_core_init_rate_req
    - clk: Change clk_core_init_rate_req prototyp...

Changed in linux-raspi (Ubuntu Jammy):
status: In Progress → Fix Released
Juerg Haefliger (juergh)
Changed in linux-raspi (Ubuntu Impish):
status: In Progress → Won't Fix
Revision history for this message
Launchpad Janitor (janitor) wrote :
Download full text (42.2 KiB)

This bug was fixed in the package linux-raspi - 5.19.0-1001.3

---------------
linux-raspi (5.19.0-1001.3) kinetic; urgency=medium

  * kinetic/linux-raspi: 5.19.0-1001.3 -proposed tracker (LP: #1982325)

  * Packaging resync (LP: #1786013)
    - [Packaging] update Ubuntu.md
    - debian/dkms-versions -- update from kernel-versions (main/master)

  * Missing overlays/README (LP: #1954757)
    - SAUCE: Install overlays/README

  * dtoverlay=uart4 breaks Raspberry Pi 4B boot (LP: #1875454)
    - SAUCE: arm: dts: Add 'brcm,bcm2835-pl011' compatible for uart2-5

  * kinetic/linux-raspi: Update to upstream raspberrypi rpi-5.19.y (2022-07-19)
    (LP: #1982090)
    - arm64: setup: Fix build warning
    - clk: Drop the rate range on clk_put()
    - clk: Skip clamping when rounding if there's no boundaries
    - clk: Introduce clk_get_rate_range()
    - clk: Mention that .recalc_rate can return 0 on error
    - clk: Clarify clk_get_rate() expectations
    - clk: tests: Add test suites description
    - clk: tests: Add reference to the orphan mux bug report
    - clk: tests: Add tests for uncached clock
    - clk: tests: Add tests for single parent mux
    - clk: tests: Add tests for mux with multiple parents
    - clk: tests: Add some tests for orphan with multiple parents
    - clk: Take into account uncached clocks in clk_set_rate_range()
    - clk: Fix clk_get_parent() documentation
    - clk: Set req_rate on reparenting
    - clk: Change clk_core_init_rate_req prototype
    - clk: Move clk_core_init_rate_req() from clk_core_round_rate_nolock() to its
      caller
    - clk: Introduce clk_hw_init_rate_request()
    - clk: Add our request boundaries in clk_core_init_rate_req
    - clk: Switch from __clk_determine_rate to clk_core_round_rate_nolock
    - clk: Introduce clk_core_has_parent()
    - clk: Stop forwarding clk_rate_requests to the parent
    - clk: Zero the clk_rate_request structure
    - clk: Test the clock pointer in clk_hw_get_name()
    - clk: Introduce the clk_hw_get_rate_range function
    - clk: qcom: clk-rcg2: Take clock boundaries into consideration for gfx3d
    - clk: tests: Add some tests for clk_get_rate_range()
    - clk: tests: Add missing test case for ranges
    - drm/vc4: Adopt the dma configuration from the HVS or V3D component
    - Support RPi DPI interface in mode6 for 18-bit color
    - drm/vc4: Add FKMS as an acceptable node for dma ranges.
    - drm/vc4: A present but empty dmas disables audio
    - drm/vc4: Add debugfs node that dumps the current display lists
    - drm/vc4: Add all the HDMI registers into the debugfs dumps
    - drm/vc4: Add the 2711 HVS as a suitable DMA node
    - vc4_hdmi: Report that 3d/stereo is allowed
    - vc4: Clear unused infoframe packet RAM registers
    - drm/vc4: Change the default DPI format to being 18bpp, not 24.
    - vc4/drm: Avoid full hdmi audio fifo writes
    - vc4/drm: vc4_plane: Remove subpixel positioning check
    - drm/vc4: Correct pixel order for DSI0
    - drm/vc4: Register dsi0 as the correct vc4 encoder type
    - drm/vc4: Fix dsi0 interrupt support.
    - drm/vc4: Add correct stop condition to vc4_dsi_encoder_disable iteration
    - drm/atomic: Don't fixup modes...

Changed in linux-raspi (Ubuntu):
status: Invalid → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.