Incus-compose storage volume not found

1

I’m playing with an incus cluster, validating what I’m hoping to do with incus in the future. What I’m hoping is to be able to use incus-compose as lightweight configuration as code and define the services, volumes, and networks that’ll be in my cluster through compose.

I have a 3 node cluster with identical hardware, each one with a btrfs storage pool.

I’ve got this compose:

services:
  server:
    image: codeberg.org/forgejo/forgejo:16
    container_name: forgejo
    environment:
      - USER_UID=1000
      - USER_GID=1000
    restart: always
    volumes:
      - forgejo-data:/data
    ports:
      - '3000:3000'
      - '222:22'

volumes:
  forgejo-data:
    x-incus-compose:
      pool: default

networks:
  default:
    x-incus-compose:
      network: incusbr0

and what i’m getting running incus-compose from my laptop:

22:09 ERR Ensuring resources project=forgejo error="in an operation: instance(forgejo): Failed creating instance record: Failed initializing instance: Failed add validation for device \"vol-forgejo-data\": Failed loading custom volume: Storage volume not found"
ensure   image              codeberg.org/forgejo/forgejo:16        [done]
ensure   image              ghcr.io/lxc/incus-compose/ic-healthd:~ [done]
ensure   network            default                                [done]
ensure   storage-volume     ic-healthd                             [done]
ensure   storage-volume     forgejo-data                           [done]
ensure   instance           forgejo-ic-healthd                     [done]
ensure   instance           forgejo                                [warn: in an operation: Failed creating instance record: Failed initializing instance: ~

Based on the amount of documentation I’m finding, this seems like an unusual configuration?

2

Hey!

Thanks for the first standalone incus-compose related Forum support Post! Ohh and welcome in the forums!

services:
  server:
    image: codeberg.org/forgejo/forgejo:16
    container_name: forgejo
    environment:
      - USER_UID=1000
      - USER_GID=1000
    restart: always
    volumes:
      - forgejo-data:/data
    ports:
      - "3000:3000"
      - "222:22"

volumes:
  forgejo-data:

networks:
  default:
    external: true
    x-incus-compose:
      network: incusbr0

Note the exterrnal: true for external networks is important else you get weired results.

r3j0@server01 ~/forgejo> incus remote add --protocol oci codeberg.org https://codeberg.org
r3j0@server01 ~/forgejo> incus-compose up
ensure   image              codeberg.org/forgejo/forgejo:16        [done]
ensure   image              ghcr.io/lxc/incus-compose/ic-healthd:~ [done]
ensure   network            default                                [done]
ensure   storage-volume     forgejo-data                           [done]
ensure   storage-volume     ic-healthd                             [done]
ensure   instance           forgejo-ic-healthd                     [done]
ensure   instance           forgejo                                [done]
start    instance           forgejo-ic-healthd                     [done]
start    instance           forgejo                                [done]

I removed the pool: default as it is not needed here. Having it in doesn’t change my result.

Please give the output of incus storage list

r3j0@server01 ~/forgejo> incus storage list
+---------+--------+-------------+---------+---------+
|  NAME   | DRIVER | DESCRIPTION | USED BY |  STATE  |
+---------+--------+-------------+---------+---------+
| default | btrfs  |             | 129     | CREATED |
+---------+--------+-------------+---------+---------+
| ext1    | btrfs  |             | 4       | CREATED |
+---------+--------+-------------+---------+---------+

As a workaround you could try to create the volume on your own in advance.

incus-compose incus storage volume create default vol-forgejo-data shifted=true