package grub-efi-amd64 2.02-2ubuntu8.3 failed to install/upgrade: installed grub-efi-amd64 package post-installation script subprocess returned error exit status 1

Bug #1788630 reported by Leonardo Müller
16
This bug affects 3 people
Affects Status Importance Assigned to Milestone
grub2 (Ubuntu)
Confirmed
Undecided
Unassigned

Bug Description

On the update of grub-efi-amd64, it's now warning about unsigned kernels, which it was not doing before, and failing. However, before this update, the unsigned kernels were booting properly with Secure Boot enabled. I don't know what would make them unbootable now. The following message appears on upgrade:

 ┌───────────────────────────┤ unsigned kernels ├────────────────────────────┐
 │ │
 │ Cannot upgrade Secure Boot enforcement policy due to unsigned kernels │
 │ │
 │ Your system has UEFI Secure Boot enabled in firmware, and the following │
 │ kernels present on your system are unsigned: │
 │ │
 │ 4.17.18-041718-lowlatency │
 │ 4.17.17-041717-lowlatency │
 │ │
 │ │
 │ These kernels cannot be verified under Secure Boot. To ensure your │
 │ system remains bootable, GRUB will not be upgraded on your disk until │
 │ these kernels are removed or replaced with signed kernels. │

Using the Ubuntu's default kernel is not an option to me, as it has bugs that weren't solved and make my notebook fail to boot.

ProblemType: Package
DistroRelease: Ubuntu 18.04
Package: grub-efi-amd64 2.02-2ubuntu8.3
Uname: Linux 4.17.17-041717-lowlatency x86_64
ApportVersion: 2.20.9-0ubuntu7.2
Architecture: amd64
Date: Thu Aug 23 11:30:50 2018
ErrorMessage: installed grub-efi-amd64 package post-installation script subprocess returned error exit status 1
InstallationDate: Installed on 2017-06-13 (435 days ago)
InstallationMedia: Xubuntu 17.04 "Zesty Zapus" - Release amd64 (20170412)
InvalidGrubScript: /etc/grub.d/40_custom
ProcCmdLine: BOOT_IMAGE=/boot/vmlinuz-4.17.17-041717-lowlatency root=UUID=6b4ae5c0-c78c-49a6-a1ba-029192618a7a ro quiet kvm.ignore_msrs=1 kvm.halt_poll_ns=0 kvm.halt_poll_ns_grow=0 intel_iommu=on iommu=pt i915.enable_gvt=1 resume=UUID=0946602f-3ca2-4379-9012-7a5171928de7 ro mtrr_gran_size=2M mtrr_chunk_size=64M cgroup_enable=memory swapaccount=1
Python3Details: /usr/bin/python3.6, Python 3.6.5, python3-minimal, 3.6.5-3ubuntu1
PythonDetails: /usr/bin/python2.7, Python 2.7.15rc1, python-minimal, 2.7.15~rc1-1
RelatedPackageVersions:
 dpkg 1.19.0.5ubuntu2
 apt 1.6.3ubuntu0.1
SourcePackage: grub2
Title: package grub-efi-amd64 2.02-2ubuntu8.3 failed to install/upgrade: installed grub-efi-amd64 package post-installation script subprocess returned error exit status 1
UpgradeStatus: Upgraded to bionic on 2017-10-20 (307 days ago)

Revision history for this message
Leonardo Müller (leozinho29-eu) wrote :
Revision history for this message
Launchpad Janitor (janitor) wrote :

Status changed to 'Confirmed' because the bug affects multiple users.

Changed in grub2 (Ubuntu):
status: New → Confirmed
Revision history for this message
Sandi Vujaković (elsandosgrande) wrote :

Since you say that you can't boot Ubuntu up on your computer without the newer mainline kernels, but Ubuntu does not supply those kernels, I will assume you used some boot parameter or another less than desirable workaround for the issue until you installed the mainline kernels after installing Ubuntu.

I am running mainline kernels on my laptop because of the wireless chip (rtl8723be). With the LTS (4.15.xx.xx) kernels, it has weak WiFi signal reception by default, but it is sort of remedied by passing a parameter (rtl8723be.ant_sel=1) in the appropriate line of /etc/default/grub before splash. However, that then leads to the same problems with Bluetooth (can't play any sound through a Bluetooth speaker unless it's touching a part of the laptop near the antenna). That issue is solved in newer mainline kernels. I also need another parameter to boot the laptop properly, even with mainline (iommu=soft). I am saying this as an example in case you didn't use boot parameters, but would be willing to try (better back up in case you don't get the right parameter on the first try, since there is a small chance that files regarding the boot sequence will get corrupted, had to reinstall because I tried to see if the newer kernels did not need iommu=soft, after the third try some boot files were dead, but /home was fine).

Tl;dr I have reasons to run mainline.

The way I got past it was to temporarily boot into the default kernel for Ubuntu (LTS), got right up to the router because I couldn't be bothered to type the parameter in, removed all of the mainline kernels (tried with them there, but grub still threw a hissy fit), went through with the upgrade, reinstalled the newest kernel from UKUU, restarted with that kernel and patted myself on the back.

I hope this helps! (Newbie commenter prays that it's not too long for whatever policy there might be for comments.)

Revision history for this message
Leonardo Müller (leozinho29-eu) wrote :

My problem is with the wireless card QCA9377. It has a 50% boot success rate. One time it boots, the next time it fails. On my case, I'm able to, say, install Ubuntu successfully, but after shutting the computer down and booting it again it would fail, so there is a window to act.

Unfortunately I have found no way to work around this issue, the only solution being to install the mainline kernel. I have helped to investigate, even bisected the kernel, but the patch alone was not enough, as it seems the problem is complex and correctable only with many patches.

After the r8169 bug I reported that was fixed to me but create an enormous regression to other users, I feel uneasy of asking to fix bugs I'm having, as they could introduce serious regressions to other users.

I already use many boot parameters, mainly to enable functionalities I desire, not to work around bugs, fortunately. I use:

quiet kvm.ignore_msrs=1 kvm.halt_poll_ns=0 kvm.halt_poll_ns_grow=0 intel_iommu=on iommu=pt i915.enable_gvt=1 resume=UUID=0946602f-3ca2-4379-9012-7a5171928de7 ro mtrr_gran_size=2M mtrr_chunk_size=64M cgroup_enable=memory swapaccount=1

As I want to use Intel GVT-g, cgroups and hibernation. There are options to ath10k, but found none that worked to me.

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.