My cloud-init works fine with Ubuntu 24.04, Debian 12/13, Alma 8 and 9, Fedora 43.
4-node LXD 6.5 cluster
But in Ubuntu 22.04 LXCs cloud-init cannot finish because the cloud-config.service and cloud-final.service depend on snapd.seeded.service and that service is failing
root@ubuntu-2204:~# systemctl status snapd.seeded.service
● snapd.seeded.service - Wait until snapd is fully seeded
Loaded: loaded (/lib/systemd/system/snapd.seeded.service; enabled; vendor preset: enabled)
Active: activating (start) since Sun 2025-10-26 08:56:14 UTC; 7min ago
Main PID: 226 (snap)
Tasks: 14 (limit: 18570)
Memory: 21.3M
CPU: 1.077s
CGroup: /system.slice/snapd.seeded.service
└─226 /usr/bin/snap wait system seed.loaded
Oct 26 08:56:14 ubuntu-2204 systemd[1]: Starting Wait until snapd is fully seeded...
My instances have working network connectivity/dns. cloud-init in ubuntu 24.04, debian 12/13, alma 8/9 and fedora works fine.
Hoping @sdeziel1 can help here 
I wanted to confirm if it works on 20.04 but something else is very broken there.
The cloud-init data isn’t even made available inside the instance.
When I launch an ubuntu 20.04 LXC with cloud-init data inherited from the default project, the data isn’t inserted into the instance. While testing I’ve had occasions where all of a sudden the cloud-init data appeared inside the instance after modifying the cloud-init config specified on the instance itself (not inherited).
Unfortunately I’ve not been able to reproduce this. I tried modifying the cloud-init user and vendor data in the 20.04 instance and it’s still not found inside the instance.
root@diamond:/home/vos# lxc exec ubuntu-2004 --project sandbox -- cat /var/lib/cloud/instance/vendor-data.txt
#cloud-config
{}
root@diamond:/home/vos# lxc config show ubuntu-2004 --project sandbox
architecture: x86_64
config:
image.architecture: amd64
image.description: ubuntu 20.04 LTS amd64 (release) (20250624)
image.label: release
image.os: ubuntu
image.release: focal
image.serial: "20250624"
image.type: squashfs
image.version: "20.04"
user.cluster: unknown
user.fqdn: ubuntu-2004.sandbox.lab.example.com
user.project: sandbox
volatile.apply_template: create
volatile.base_image: 94b4b0055aca5c16e801aa20c08aba22f387af228a3c578ab5ef9c8a342d5832
volatile.cloud-init.instance-id: 0c9e95db-89d5-4d16-b7ef-faa42c5c563e
volatile.eth2.hwaddr: 00:16:3e:49:f2:7b
volatile.eth2.name: eth0
volatile.idmap.base: "0"
volatile.idmap.current: '[{"Isuid":true,"Isgid":false,"Hostid":1000000,"Nsid":0,"Maprange":1000000000},{"Isuid":false,"Isgid":true,"Hostid":1000000,"Nsid":0,"Maprange":1000000000}]'
volatile.idmap.next: '[{"Isuid":true,"Isgid":false,"Hostid":1000000,"Nsid":0,"Maprange":1000000000},{"Isuid":false,"Isgid":true,"Hostid":1000000,"Nsid":0,"Maprange":1000000000}]'
volatile.last_state.idmap: '[]'
volatile.uuid: a0945ff5-ae3d-4123-925b-b85dc9f954f7
volatile.uuid.generation: a0945ff5-ae3d-4123-925b-b85dc9f954f7
devices: {}
ephemeral: false
profiles:
- default
- storage-local
- network-sandbox
stateful: false
description: ""
root@diamond:/home/vos#
root@diamond:/home/vos#
root@diamond:/home/vos# lxc exec ubuntu-2004 --project sandbox -- cat /var/lib/cloud/instance/vendor-data.txt
#cloud-config
{}
root@diamond:/home/vos# lxc config show ubuntu-2004 --project sandbox --expanded
architecture: x86_64
config:
cloud-init.vendor-data: |-
#cloud-config
timezone: Europe/Amsterdam
package_upgrade: false
packages:
- openssh-server
- curl
users:
- name: ansible
uid: 600
no_user_group: true
primary_group: nogroup
groups: sudo
shell: /bin/bash
sudo: "ALL=(ALL) NOPASSWD:ALL"
ssh_authorized_keys:
- ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIMj1/LsMgJf3ci0axbBJd39AlsTTzzvJ2r0w3Uw/mAHA ansible@vos.dev
- ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIJf9y9WBKU+TpLTPjkZiqCVfzA+DXRdB/Gf7dnpCCukV Semaphore
cluster.evacuate: stop
image.architecture: amd64
image.description: ubuntu 20.04 LTS amd64 (release) (20250624)
image.label: release
image.os: ubuntu
image.release: focal
image.serial: "20250624"
image.type: squashfs
image.version: "20.04"
user.cluster: unknown
user.fqdn: ubuntu-2004.sandbox.lab.example.com
user.project: sandbox
volatile.apply_template: create
volatile.base_image: 94b4b0055aca5c16e801aa20c08aba22f387af228a3c578ab5ef9c8a342d5832
volatile.cloud-init.instance-id: 0c9e95db-89d5-4d16-b7ef-faa42c5c563e
volatile.eth2.hwaddr: 00:16:3e:49:f2:7b
volatile.eth2.name: eth0
volatile.idmap.base: "0"
volatile.idmap.current: '[{"Isuid":true,"Isgid":false,"Hostid":1000000,"Nsid":0,"Maprange":1000000000},{"Isuid":false,"Isgid":true,"Hostid":1000000,"Nsid":0,"Maprange":1000000000}]'
volatile.idmap.next: '[{"Isuid":true,"Isgid":false,"Hostid":1000000,"Nsid":0,"Maprange":1000000000},{"Isuid":false,"Isgid":true,"Hostid":1000000,"Nsid":0,"Maprange":1000000000}]'
volatile.last_state.idmap: '[]'
volatile.uuid: a0945ff5-ae3d-4123-925b-b85dc9f954f7
volatile.uuid.generation: a0945ff5-ae3d-4123-925b-b85dc9f954f7
devices:
eth2:
network: sandbox
type: nic
root:
path: /
pool: local
size: 15GiB
type: disk
ephemeral: false
profiles:
- default
- storage-local
- network-sandbox
stateful: false
description: ""
root@diamond:/home/vos# lxc exec ubuntu-2004 --project sandbox -- cat /var/lib/cloud/instance/vendor-data.txt
#cloud-config
{}
root@diamond:/home/vos# lxc exec ubuntu-2004 --project sandbox -- cat /var/lib/cloud/instance/user-data.txt
#cloud-config
{}
Some steps to make the issue clearly visible:
I’m launching the 3 ubuntu LTS instances with identical LXD config.
cloud-init config is empty in 20.04, never reaching done in 22.04 and works in 24.04
root@diamond:/home/vos# lxc launch ubuntu:20.04 u1 -p default -p storage-local -p network-sandbox --project sandbox
Launching u1
root@diamond:/home/vos# lxc exec u1 --project sandbox -- cloud-init status
status: done
root@diamond:/home/vos# lxc exec u1 --project sandbox -- id ansible
id: ‘ansible’: no such user
root@diamond:/home/vos# lxc launch ubuntu:22.04 u2 -p default -p storage-local -p network-sandbox --project sandbox
Launching u2
root@diamond:/home/vos# lxc exec u2 --project sandbox -- cloud-init status
status: running
root@diamond:/home/vos# lxc exec u2 --project sandbox -- id ansible
uid=600(ansible) gid=65534(nogroup) groups=65534(nogroup),27(sudo)
root@diamond:/home/vos# lxc launch ubuntu:24.04 u3 -p default -p storage-local -p network-sandbox --project sandbox
Launching u3
root@diamond:/home/vos# lxc exec u3 --project sandbox -- cloud-init status
status: done
root@diamond:/home/vos# lxc exec u3 --project sandbox -- id ansible
uid=600(ansible) gid=65534(nogroup) groups=65534(nogroup),27(sudo)
root@diamond:/home/vos# lxc exec u2 --project sandbox -- cloud-init status
status: running
root@diamond:/home/vos# lxc exec u1 --project sandbox -- cat /var/lib/cloud/instance/vendor-data.txt
#cloud-config
{}
root@diamond:/home/vos# lxc exec u2 --project sandbox -- cat /var/lib/cloud/instance/vendor-data.txt
#cloud-config
package_upgrade: false
<..>
root@diamond:/home/vos# lxc exec u3 --project sandbox -- cat /var/lib/cloud/instance/vendor-data.txt
#cloud-config
package_upgrade: false
<..>
When I create an instance and populate either user or vendor data it sometimes does work but i am no longer able to reproduce it.
root@diamond:/home/vos# lxc delete u4 --project sandbox --force
root@diamond:/home/vos# lxc init ubuntu:20.04 u4 -p default -p storage-local -p network-sandbox --project sandbox
Creating u4
root@diamond:/home/vos# lxc config edit u4 --project sandbox
root@diamond:/home/vos# lxc config show u4 --project sandbox
architecture: x86_64
config:
cloud-init.user-data: |
#cloud-init
package_upgrade: true
image.architecture: amd64
image.description: ubuntu 20.04 LTS amd64 (release) (20250624)
image.label: release
image.os: ubuntu
image.release: focal
<..>
root@diamond:/home/vos# lxc start u4 --project sandbox
root@diamond:/home/vos#
root@diamond:/home/vos# lxc exec u4 --project sandbox -- cloud-init status
status: done
root@diamond:/home/vos# lxc exec u4 --project sandbox -- cat /var/lib/cloud/instance/user-data.txt
#cloud-config
{}