[Bug 2007827] Re: flash-kernel failure when upgrading f-k anad kernel in the same cycle

[Dave Jones]

Okay, attempt number 3! After another long dive into dpkg states,
triggers, and the interactions around the kernel and initrd I think I've
finally got a version that should:

1. Never install a kernel which isn't fully installed (unless --force is
specified)

2. Avoids the race on waiting for a kernel to become fully installed
during an apt run (when it discovers it's being asked to install a non-
fully-installed kernel, it defers itself by retriggering)

3. Avoids duplicated runs (it does use the full version list from
"linux-version list", but defers in the case that the latest isn't fully
installed)

It's currently building the aforementioned PPA
(https://launchpad.net/~waveform/+archive/ubuntu/flash-kernel) and the
test procedure should be the same:

* Install a stock lunar preinstalled server image
* sudo add-apt-repository ppa:waveform/flash-kernel
* sudo apt install linux-xilinx-zynqmp flash-kernel
* Ensure that upgrade proceeds without error
* sudo reboot
* Ensure system boots with new kernel version (and initrd!) successfully

I've tested this several times with the raspi images (that displayed a
related, but different symptom) in circumstances that do replicate the
race condition without the fix and things are successful there, so I'm
hoping that's also the case on Xilinx.

-- 
You received this bug notification because you are a member of Ubuntu
Sponsors Team, which is subscribed to the bug report.
https://bugs.launchpad.net/bugs/2007827

Title:
  flash-kernel failure when upgrading f-k anad kernel in the same cycle

Status in flash-kernel package in Ubuntu:
  New

Bug description:
  [Impact]
  In version 3.104ubuntu15 of flash-kernel, when both f-k and the kernel are upgraded in the same cycle, depending on the ordering of dpkg trigger execution, f-k may find the content of /boot "inconsistent" causing it to fail and return error exit status 1.

  Erorr message:
  Processing triggers for man-db (2.10.2-1) ...
  Processing triggers for flash-kernel (3.104ubuntu15) ...
  flash-kernel: installing version 5.15.0-1018-xilinx-zynqmp
  Initrd required for FIT method
  dpkg: error processing package flash-kernel (--configure):
   installed flash-kernel package post-installation script subprocess returned error exit status 1
  Processing triggers for linux-image-5.15.0-1018-xilinx-zynqmp (5.15.0-1018.20) ...
  /etc/kernel/postinst.d/initramfs-tools:
  update-initramfs: Generating /boot/initrd.img-5.15.0-1018-xilinx-zynqmp

  flash-kernel gets the latest kernel version by "linux-version list".
  When flash-kernel was triggered to generate fitimage, the kernel version is "5.15.0-1018" and the initrd for it wasn't ready. So, flash-kernel failed to generate the fitimage.

  A subsequent run of "apt install -f" fixed things because, by that
  point, the kernel's own trigger had executed, ensuring that update-
  initramfs had been run. In the case that f-k is run "prematurely" and
  finds itself in this situation (/boot/kernel-$[ver} exists, but
  /boot/initrd-${ver}) doesn't), it should probably bail out silently
  under the assumption that whatever is responsible for it will rectify
  the situation and trigger f-k again (as happens in the kernel postinst
  hooks).

  [Test Case]
  1. Flash an old image (with an out of date kernel and flash-kernel)
  2. sudo apt-get update
  3. sudo apt install flash-kernel with the fix and linux packages
  4. Upgrade should proceed without issue

  [Regression Potential]
  As with the previous flash-kernel uploads, it is possible that a breakage in the changed code can lead to issues with upgrading kernels (due to f-k being executed via a trigger at the end) or with Xilinx devices in the field not upgrading correctly. I will test all the changes extensively though.

  Related issues:
  LP: #1861292 flash-kernel failure during kernel upgrade

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/flash-kernel/+bug/2007827/+subscriptions