Unable to start instances

Incus
1

I upgraded a home machine from incus 6.10 to 6.17.

After restart instances are stopped and log flooded with messages like this one:

incusd[1013]: time="2025-10-20T11:54:18+02:00" level=warning msg="Failed auto start instance attempt" attempt=3 err="Failed to start device \"eth0\": Failed to delete ingress qdisc {LinkIndex: 47, Handle: ffff:0, Parent: ingress, Refcnt: 0}: operation not supported" instance=vpn maxAttempts=3 project=default
2

What distro is that on and can you show incus config show --expanded for one of those instances?

3

it is Debian trixie, with a custom kernel.

It used to work fine until I upgraded. I saw you changed from iproute2 to netlink API, maybe my kernel is missing something.

incus config show --expanded vpn
architecture: x86_64
config:
  boot.autostart: "true"
  image.architecture: amd64
  image.description: Debian bookworm amd64 (20250507_06:23)
  image.os: Debian
  image.release: bookworm
  image.serial: "20250507_06:23"
  image.type: disk-kvm.img
  image.variant: default
  limits.cpu: "2"
  limits.memory: 512MiB
  limits.memory.hugepages: "true"
  security.secureboot: "false"
  volatile.base_image: 527cd792049205acd30e4a968fafff46eb4b5704f93099e201dbaffe19f301ba
  volatile.cloud-init.instance-id: 380c515f-48c4-4f5e-91e7-e9c6f32b85c7
  volatile.eth0.hwaddr: 10:66:6a:c1:3d:b1
  volatile.last_state.power: RUNNING
  volatile.uuid: 0c14a204-8d1a-43a9-b1bf-e4003e31be38
  volatile.uuid.generation: 0c14a204-8d1a-43a9-b1bf-e4003e31be38
  volatile.vm.definition: pc-q35-9.0
  volatile.vm.rtc_adjustment: "-2"
  volatile.vm.rtc_offset: "-1"
  volatile.vsock_id: "1426524327"
devices:
  eth0:
    name: eth0
    nictype: bridged
    parent: lan
    type: nic
  root:
    path: /
    pool: default
    type: disk
ephemeral: false
profiles:
- default
stateful: false
description: ""
4

Yeah, I’m not seeing anything that should be problematic in your instance config.
What kernel version are you running?

5

From what I could find, the tc/qdisc stuff appears to depend on at least the CONFIG_NET_CLS config:

stgraber@dakara:~$ grep CONFIG_NET_CLS /boot/config-6.16.11-zabbly+ 
CONFIG_NET_CLS=y
CONFIG_NET_CLS_BASIC=m
CONFIG_NET_CLS_ROUTE4=m
CONFIG_NET_CLS_FW=m
CONFIG_NET_CLS_U32=m
CONFIG_NET_CLS_FLOW=m
CONFIG_NET_CLS_CGROUP=m
CONFIG_NET_CLS_BPF=m
CONFIG_NET_CLS_FLOWER=m
CONFIG_NET_CLS_MATCHALL=m
CONFIG_NET_CLS_ACT=y
6

Indeed, that was it. Thank you for your time!

It is now solved :smiley: