VM won't boot after power cut / restart on Incus 6.17

Incus
1

Hello,

I’m facing issue with VM that don’t boot after power cut by mistake

the VM freeze to

 Booting \`Rocky Linux (5.14.0-427.24.1.el9_4.x86_64) 9.4 (Blue Onyx)' 

!!!! X64 Exception Type - 0E(#PF - Page-Fault)  CPU Apic ID - 00000000 !!!! 
ExceptionData - 0000000000000003  I:0 R:0 U:0 W:1 P:1 PK:0 SS:0 SGX:0 
RIP  - 000000007529A920, CS  - 0000000000000038, RFLAGS - 0000000000210006 
RAX  - 00000000752B3000, RCX - 00000000752B3000, RDX - 0000000000024000 
RBX  - 000000007C9A9500, RSP - 000000007EE9C228, RBP - 000000007D437A18 
RSI  - 0000000000000000, RDI - 00000000752B3000 
R8   - 00000000752D7000, R9  - 000000007AC8B6FD, R10 - 000000007C42EDB9 
R11  - 0000000000000077, R12 - 000000007E9EC018, R13 - 000000007D106000 
R14  - 000000007C9A8768, R15 - 000000007C9A8770 
DS   - 0000000000000030, ES  - 0000000000000030, FS  - 0000000000000030 
GS   - 0000000000000030, SS  - 0000000000000030 
CR0  - 0000000080010033, CR2 - 00000000752B3000, CR3 - 000000007EC01000 
CR4  - 0000000000000668, CR8 - 0000000000000000 
DR0  - 0000000000000000, DR1 - 0000000000000000, DR2 - 0000000000000000 
DR3  - 0000000000000000, DR6 - 00000000FFFF0FF0, DR7 - 0000000000000400 
GDTR - 000000007E9E1000 0000000000000047, LDTR - 0000000000000000 
IDTR - 000000007E025018 0000000000000FFF,   TR - 0000000000000000 
FXSAVE_STATE - 000000007EE9BE80 
!!!! Find image based on IP(0x7529A920) (No PDB)  (ImageBase=0000000001091170, EntryPoint=0000000001091997) !!!!

I didn’t change anything in VM configuration and incus version is

Client version: 6.17
Server version: 6.17

VM config:

architecture: x86_64
config:
  image.architecture: amd64
  image.description: Rockylinux 9 amd64 (20240717_02:06)
  image.os: Rockylinux
  image.release: '9'
  image.requirements.cdrom_agent: 'true'
  image.serial: '20240717_02:06'
  image.type: disk-kvm.img
  image.variant: cloud
  limits.cpu: '2'
  limits.memory: 4GiB
  volatile.base_image: 673712181ebe86f060a0170257841c7507b773e4a1e2df3d502d3cd26a3b5aab
  volatile.cloud-init.instance-id: 99bc53d7-60e4-41f8-ab52-c7a47a32f233
  volatile.eth0.host_name: tap7a0b7a13
  volatile.eth0.hwaddr: 00:16:3e:9b:ca:4f
  volatile.last_state.power: RUNNING
  volatile.last_state.ready: 'false'
  volatile.uuid: 6c547028-d137-4405-bd60-c4813f9af9bd
  volatile.uuid.generation: 6c547028-d137-4405-bd60-c4813f9af9bd
  volatile.vm.definition: pc-q35-9.0
  volatile.vm.rtc_adjustment: '-13'
  volatile.vm.rtc_offset: '-1'
  volatile.vsock_id: '2861442233'
devices:
  agent:
    source: agent:config
    type: disk
ephemeral: false
profiles:
  - default
stateful: false
description: ''
created_at: '2024-07-18T01:08:25.789231116Z'
name: bridge46-wordpress-blog
status: Running
status_code: 103
last_used_at: '2025-09-27T14:01:04.776600152Z'
location: none
type: virtual-machine
project: default
2

Actually I don’t know if this related to recent versions of incus, VM packages or Power cut

I know power cut happen many times before and I didn’t face this issue

3

Got the same issue here, so I’m suspecting EDK2.

We bumped the version of EDK2 following some cherry-picks to try to handle RedHat based guests, but apparently the cherry-picks weren’t sufficient…

1 Like
4

I think I found the issue in our packaging of the workaround.
I just pushed a new daily build now, after that’s built and is confirmed to fix it here, I’ll push it to stable too.

1 Like
5

fix confirmed, I just upgraded (somehow apt display it as downgrade) to 1:0~debian13~202509281046 and it resolve the issue

thank you @stgraber

6

Yep, confirmed it working here too on the daily builds. I’m now pushing this to stable too.

7

The stable update has been pushed now and should be hitting users within the next 30 minutes

1 Like