Failed 10.10 upgrade - how to recover?

[achelon]

Hi!

First post, be kind!

I did an upgrade to Debian 10.10 yesterday with the usual

Code: Select all

apt-get update
apt-get dist-upgrade

Unfortunately the upgrade process bailed out half-way through and I do not know how to recover from the presumably inconsistent state. I foolishly did not snapshot the VM beforehand.

Code: Select all

...
Setting up linux-image-amd64 (4.19+105+deb10u12) ...
Setting up shim-helpers-amd64-signed (1+15.4+5~deb10u1) ...
Installing for x86_64-efi platform.
grub-install: error: cannot find EFI directory.
dpkg: error processing package shim-helpers-amd64-signed (--configure):
 installed shim-helpers-amd64-signed package post-installation script subprocess returned error exit status 1
Setting up linux-headers-4.19.0-17-amd64 (4.19.194-1) ...
Setting up linux-headers-amd64 (4.19+105+deb10u12) ...
dpkg: dependency problems prevent configuration of shim-signed:amd64:
 shim-signed:amd64 depends on shim-helpers-amd64-signed (>= 1+15.4+2~deb10u1); however:
  Package shim-helpers-amd64-signed is not configured yet.

dpkg: error processing package shim-signed:amd64 (--configure):
 dependency problems - leaving unconfigured
Processing triggers for man-db (2.8.5-2) ...
Processing triggers for libc-bin (2.28-10) ...
Errors were encountered while processing:
 shim-helpers-amd64-signed
 shim-signed:amd64
E: Sub-process /usr/bin/dpkg returned an error code (1)

Any idea what to do next?

Regards,
Achelon

[p.H]

Can you check that you have a directory /boot/efi and the EFI system partition is mounted on it ?

[achelon]

Yep I checked and it looks same as always. Only thing unusual about this server is encrypted boot but this has never been an issue with updates before:

Disk layout is:

Code: Select all

NAME                     MAJ:MIN RM   SIZE RO TYPE  MOUNTPOINT
sda                        8:0    0    25G  0 disk
├─sda1                     8:1    0   512M  0 part
├─sda2                     8:2    0   244M  0 part
│ └─boot_crypt           254:6    0   242M  0 crypt /boot
└─sda3                     8:3    0  24.3G  0 part
  └─sda3_crypt           254:0    0  24.3G  0 crypt
    ├─aquinas--vg-root   254:1    0   4.5G  0 lvm   /
    ├─aquinas--vg-var    254:2    0   1.8G  0 lvm   /var
    ├─aquinas--vg-swap_1 254:3    0     3G  0 lvm
    ├─aquinas--vg-tmp    254:4    0   392M  0 lvm   /tmp
    └─aquinas--vg-home   254:5    0  14.5G  0 lvm   /home
sdb                        8:16   0   300G  0 disk
└─data_crypt_dif         254:7    0 282.3G  0 crypt
  └─data_crypt           254:8    0 282.3G  0 crypt /mnt/Test

EFI is located at:

Code: Select all

(1:2019)$ sudo ls -R /boot/efi/
/boot/efi/:
EFI

/boot/efi/EFI:
debian

/boot/efi/EFI/debian:
BOOTX64.CSV  fbx64.efi	grub.cfg  grubx64.efi  mmx64.efi  shimx64.efi

[p.H]

lsblk does not show an EFI partition (sda1 ? oversized) mounted on /boot/efi.

PS : do not use sudo when it is not needed.

[achelon]

Oh crap, I missed that - thanks. After fixing that, what can I do to correct the inconsistency?

[achelon]

PS : do not use sudo when it is not needed.

It is needed:

Code: Select all

(1:2000)$ ls /boot/efi/
ls: cannot open directory '/boot/efi/': Permission denied

[p.H]

Try

Code: Select all

apt-get --fix-broken upgrade

[achelon]

Thanks dude. Fixed!

Code: Select all

(1:2002)$ sudo apt-get --fix-broken upgrade
Reading package lists... Done
Building dependency tree
Reading state information... Done
Calculating upgrade... Done
The following packages were automatically installed and are no longer required:
  linux-headers-4.19.0-14-amd64 linux-headers-4.19.0-14-common linux-image-4.19.0-14-amd64
Use 'sudo apt autoremove' to remove them.
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
2 not fully installed or removed.
After this operation, 0 B of additional disk space will be used.
Do you want to continue? [Y/n] y
Setting up shim-helpers-amd64-signed (1+15.4+5~deb10u1) ...
Installing for x86_64-efi platform.
Installation finished. No error reported.
Setting up shim-signed:amd64 (1.36~1+deb10u1+15.4-5~deb10u1) ...
Installing for x86_64-efi platform.
Installation finished. No error reported.
Secure Boot not enabled on this system.

[stevepusser]

Code: Select all

 foolishly did not snapshot the VM beforehand.

Though it's solved for now, this comment deserves more 'splaning, IMO.

[achelon]

Though it's solved for now, this comment deserves more 'splaning, IMO.

I guess I'm happy to 'splain.

The update/upgrade process for Debian has been so trouble free for so long I had become complacent about manually taking VM snapshots before running updates. Furthermore, though I had long wished to implement some sort of scripted (and thus automatic) snapshot rotation policy (daily/weekly/monthly) so that there was no reliance on manual snapshots (which you can forget or become complacent about) I have not found a satisfactory way of implementing this in ESXi when like mine, the ESXi box has no vcenter attached.