Incus 6.17 has been released

stgraber · September 26, 2025, 4:46am

Introduction

The Incus team is pleased to announce the release of Incus 6.17!

This release comes with an early CLI for IncusOS users, a couple of nice enhancements to OVN networking, more flexibility for cluster users and a couple of new instance options.

As usual, you can try it for yourself online: Linux Containers - Incus - Try it online

Enjoy!

New features

IncusOS management commands

A new set of commands have been added under incus admin os to allow for easy configuration of remote IncusOS systems.

stgraber@castiana:~$ incus admin os system show laptop:storage
WARNING: The IncusOS API and configuration is subject to change

config: {}
state:
  drives:
  - boot: true
    bus: nvme
    capacity_in_bytes: 5.12110190592e+11
    id: /dev/disk/by-id/nvme-Samsung_SSD_970_PRO_512GB_S5JYNS0RB08237J
    model_family: ""
    model_name: Samsung SSD 970 PRO 512GB
    remote: false
    removable: false
    serial_number: S5JYNS0RB08237J
    smart:
      enabled: true
      passed: true
  pools:
  - devices:
    - /dev/disk/by-partlabel/local-data
    encryption_key_status: available
    name: local
    pool_allocated_space_in_bytes: 7.03410176e+08
    raw_pool_size_in_bytes: 4.7244640256e+11
    state: ONLINE
    type: zfs-raid0
    usable_pool_size_in_bytes: 4.7244640256e+11

This is still a very early version of that CLI and a lot is yet to come, but it can already be used for some of the common configuration steps around networking, storage and system services.

Tunnel support on OVN networks

OVN networks can now make use of the tunnel configuration keys that have long been present on traditional bridge networks.

This allows for extending an OVN network over GRE or VXLAN.

root@server04:~# incus network show ovn-vxlan
config:
  bridge.mtu: "1450"
  ipv4.address: 10.12.125.1/24
  ipv4.nat: "true"
  ipv6.address: fd42:ce36:66f9:5ec2::1/64
  ipv6.nat: "true"
  network: UPLINK
  tunnel.sta.id: "100"
  tunnel.sta.protocol: vxlan
  volatile.network.ipv4.address: 172.31.254.15
  volatile.network.ipv6.address: fd00:1e4d:637d:1234:1266:6aff:fe98:b8db
description: ""
name: ovn-vxlan
type: ovn
used_by: []
managed: true
status: Created
locations:
- server04
- server03
- server01
- server02
project: default

Documentation: OVN network - Incus documentation

Control over out-of-memory priority

A new configuration key for both containers and virtual machines has been introduced to control the process priority in the event of an out of memory event.

The configuration key is limits.memory.oom_priority with its value corresponding to kernel OOM priorities with lower values being less likely to be selected for freeing in the event of an out of memory event.

Documentation: Instance options - Incus documentation

Override-able configuration and devices on backup import

It’s now possible to apply configuration and device overrides during a backup import (incus import). This works in the same way as during an instance copy (incus copy).

stgraber@castiana:~$ incus config show a1
architecture: x86_64
config:
  image.architecture: amd64
  image.description: Alpine edge amd64 (20250925_13:00)
  image.os: Alpine
  image.release: edge
  image.requirements.secureboot: "false"
  image.serial: "20250925_13:00"
  image.type: squashfs
  image.variant: default
  volatile.base_image: aba1610b7aca5e68952511100e2652423baaa326d66d8a81efafb9385b322976
  volatile.cloud-init.instance-id: 3da65657-1355-40ec-88b0-c69cee657ddb
  volatile.eth0.host_name: veth18d71455
  volatile.eth0.hwaddr: 10:66:6a:fd:cc:96
  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.last_state.power: RUNNING
  volatile.uuid: dc392e6f-e1dd-42cd-af81-3491eebac20e
  volatile.uuid.generation: dc392e6f-e1dd-42cd-af81-3491eebac20e
devices:
  root:
    path: /
    pool: default
    size: 1GiB
    type: disk
ephemeral: false
profiles:
- default
stateful: false
description: ""
stgraber@castiana:~$ incus export a1
Backup exported successfully!
stgraber@castiana:~$ incus delete -f a1

stgraber@castiana:~$ incus import a1.tar.gz -c limits.cpu=4 -d root,size=5GiB
stgraber@castiana:~$ incus config show a1
architecture: x86_64
config:
  image.architecture: amd64
  image.description: Alpine edge amd64 (20250925_13:00)
  image.os: Alpine
  image.release: edge
  image.requirements.secureboot: "false"
  image.serial: "20250925_13:00"
  image.type: squashfs
  image.variant: default
  limits.cpu: "4"
  volatile.base_image: aba1610b7aca5e68952511100e2652423baaa326d66d8a81efafb9385b322976
  volatile.cloud-init.instance-id: 3da65657-1355-40ec-88b0-c69cee657ddb
  volatile.eth0.host_name: veth18d71455
  volatile.eth0.hwaddr: 10:66:6a:fd:cc:96
  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.last_state.power: RUNNING
  volatile.uuid: dc392e6f-e1dd-42cd-af81-3491eebac20e
  volatile.uuid.generation: dc392e6f-e1dd-42cd-af81-3491eebac20e
devices:
  root:
    path: /
    pool: default
    size: 5GiB
    type: disk
ephemeral: false
profiles:
- default
stateful: false
description: ""

`database-client` cluster role

A new database-client cluster role has been added.

Giving this role to a server within a cluster will ensure that this server will never get a copy of the database and will always act as a database client only.

The main use case for this role is to handle environments where some servers in the cluster are actually virtual machines running on top of physical servers that are themselves also in the cluster. In such a scenario, you want to ensure that none of those VMs ever become part of the database cluster as losing one of the physical servers would cause the loss of one or more additional database servers, potentially causing the cluster to instantly lose quorum and deadlock.

root@server04:~# incus cluster list
+----------+----------------------------------------------------+------------------+--------------+----------------+-------------+--------+-------------------+
|   NAME   |                        URL                         |      ROLES       | ARCHITECTURE | FAILURE DOMAIN | DESCRIPTION | STATUS |      MESSAGE      |
+----------+----------------------------------------------------+------------------+--------------+----------------+-------------+--------+-------------------+
| server01 | https://[2602:fc62:b:8006:216:3eff:fe1a:ed0d]:8443 | database         | x86_64       | default        |             | ONLINE | Fully operational |
+----------+----------------------------------------------------+------------------+--------------+----------------+-------------+--------+-------------------+
| server02 | https://[2602:fc62:b:8006:216:3eff:fe56:5276]:8443 | database         | x86_64       | default        |             | ONLINE | Fully operational |
+----------+----------------------------------------------------+------------------+--------------+----------------+-------------+--------+-------------------+
| server03 | https://[2602:fc62:b:8006:216:3eff:fec6:eaa8]:8443 | database-leader  | x86_64       | default        |             | ONLINE | Fully operational |
|          |                                                    | database         |              |                |             |        |                   |
+----------+----------------------------------------------------+------------------+--------------+----------------+-------------+--------+-------------------+
| server04 | https://[2602:fc62:b:8006:216:3eff:fea3:6d]:8443   | database-standby | x86_64       | default        |             | ONLINE | Fully operational |
+----------+----------------------------------------------------+------------------+--------------+----------------+-------------+--------+-------------------+
root@server04:~# incus cluster role add server04 database-client
root@server04:~# incus cluster list
+----------+----------------------------------------------------+-----------------+--------------+----------------+-------------+--------+-------------------+
|   NAME   |                        URL                         |      ROLES      | ARCHITECTURE | FAILURE DOMAIN | DESCRIPTION | STATUS |      MESSAGE      |
+----------+----------------------------------------------------+-----------------+--------------+----------------+-------------+--------+-------------------+
| server01 | https://[2602:fc62:b:8006:216:3eff:fe1a:ed0d]:8443 | database        | x86_64       | default        |             | ONLINE | Fully operational |
+----------+----------------------------------------------------+-----------------+--------------+----------------+-------------+--------+-------------------+
| server02 | https://[2602:fc62:b:8006:216:3eff:fe56:5276]:8443 | database        | x86_64       | default        |             | ONLINE | Fully operational |
+----------+----------------------------------------------------+-----------------+--------------+----------------+-------------+--------+-------------------+
| server03 | https://[2602:fc62:b:8006:216:3eff:fec6:eaa8]:8443 | database-leader | x86_64       | default        |             | ONLINE | Fully operational |
|          |                                                    | database        |              |                |             |        |                   |
+----------+----------------------------------------------------+-----------------+--------------+----------------+-------------+--------+-------------------+
| server04 | https://[2602:fc62:b:8006:216:3eff:fea3:6d]:8443   | database-client | x86_64       | default        |             | ONLINE | Fully operational |
+----------+----------------------------------------------------+-----------------+--------------+----------------+-------------+--------+-------------------+

Documentation: About clustering - Incus documentation

Support for `parent=none` on OVN uplink networks

It’s now possible to have an OVN uplink network that only exists on a subset of the servers within the cluster. The servers that don’t have access to the physical network in question should have parent=none set on them.

With that done, OVN will only run the logical routers on servers that are physically connected to the uplink while the rest of the servers will still be able to run instances with all traffic being tunneled over to the other servers.

Documentation: OVN network - Incus documentation

Cluster groups in configuration preseed

The Incus server preseed logic has been extended to now include cluster groups.
This allows configuring the default groups as well as defining or updating additional groups during initialization.

root@server01:~# cat seed.yaml 
cluster_groups:
  - name: default
    description: Default cluster group
    config:
      instances.vm.cpu.x86_64.baseline: host
  - name: test2
    description: Test
    config:
      instances.vm.cpu.x86_64.baseline: core2duo
root@server01:~# incus admin init --preseed < seed.yaml 
root@server01:~#

Complete changelog

Here is a complete list of all changes in this release:

Full commit list

tests: Skip XFS on ZFS with Ubuntu 24.04
doc: Remove unnecessary backquotes around commands in TrueNAS doc
incusd/device/disk: Allow mounting individual files from custom volumes
tests: Test mounting individual files from custom volumes
network/incusd: Allow parent=none physical networks, bypass chassis enablement
incusd/cluster/evacuate: Clarify error
doc/instance-create: Mention TLS for the agent
devcontainer: Update Go to 1.24 and Debian to trixie
incusd/db: Add ‘database-client’ role
incusd/cluster: Add support for rebalancing nodes with ‘database-client’ role
incusd: Add logic for selecting nodes with ‘database-client’ role to process
incusd: Allow rebalancing when there are too many voters or stand-by nodes
docs: Fix markdown linting issues for tables
incusd/cluster: Rename dqlite to cowsql in header
incus: Add incus admin os command
i18n: Update translation templates
incusd/cluster: Clarify comment for cluster rebalancing
incusd/instance_patch: Fix description field not respecting PATCH semantics
Translated using Weblate (Portuguese)
client/oci: Allow OCI image names with a pinned hash
doc: Add Rocky 10 Copr Repository
shared/tls: Remove tls.Config.Time override
build(deps): bump actions/setup-go from 5 to 6
build(deps): bump actions/labeler from 5 to 6
incusd/storage/lvm: Don’t use pvcreate with cluster
incus/admin/os: Fix list commands
incus/admin/os: Don’t require remote name in debug
incus/admin/os: Rename commands for consistency
doc: Update third party tool URLs
doc: Hashicorp web servers don’t like Github Runners
incusd/api_os: Set X-IncusOS-Proxy prefix
incus/export: Fail fast if target already exists
i18n: Update translation templates
tests: Cleanup backup files after use
docs: Clarify clustered storage pools
api: instance_limits_oom
incusd/instance/validate: Add OOM priority validation
incusd/instance/config: Add limits.memory.oom_priority config key
doc: Update metadata
incusd/instance/driver_common: Add setOOMPriority shared method
incusd/instance/driver_lxc: Apply OOM priority to containers
incusd/instance/driver_qemu: Apply OOM priority to VMs
incusd/project/permissions: Forbid limits.memory.oom_priority in restricted projects
api: backup_override_config
client: Allow config/device override on backup import
incus/import: Add override config and device
incusd/instances/import: Add device and config overrides
i18n: Update translation templates
po: Manual import from weblate
internal/incusos: Introduce IncusOS API client
incusd/sys/os: Make use of IncusOS client
incusd: Update for OS struct change
incusd/networks: Use IncusOS API client
incusd/storage: Add IncusOS service checks
incusd/network: Add IncusOS service checks
incusd/config: Prevent unsetting core.https_address on IncusOS
incusd: Consistent spelling of IncusOS
incusd/certificates: Add check for IncusOS
incusd/db: Add UpdateStoragePoolConfig
incusd/storage/drivers: Add SameSource field to Info struct
incusd: Update configs of all members when SameSource is true
incusd/storage: Fix unmount calls for ISO volumes
incusd/storage/lvm: Fix locking changes
incusd/instance/qemu: Re-enable vsock on Windows
shared/idmap: Skip xattrs on EINVAL
Makefile: Bump minimal Go to 1.24.7 and remove pins
gomod: Update dependencies
Makefile: Use latest go-swagger (for Go 1.25 support)
Revert “github: Disable go-swagger on Go 1.25+”
incusd/network/driver: Add support for network tunnels to OVN networks
incusd/networks: Validate config keys only for ClientTypeNormal requests
api: Add network_ovn_tunnels extension
doc: Update config
client/oci: Set the umoci logger on init
incusd/instances/oci: Properly handle environment from profiles
shared/api: Add missing YAML tag
api: init_preseed_cluster_groups
shared/api: Add support for cluster group preseeding
client: Add support for cluster group preseeding
doc/rest-api: Refresh swagger YAML
client: Implement IdenticalCertificate
incusd/cluster: Set IdenticalCertificate for intra-cluster connections
incusd/cluster: Rework tlsTransport to handle cluster certificate validation
incusd/cluster: Make use of updated tlsTransport
gomod: Update dependencies

Documentation

The Incus documentation can be found at:

Packages

There are no official Incus packages as Incus upstream only releases regular release tarballs. Below are some available options to get Incus up and running.

Installing the Incus server on Linux

Incus is available for most common Linux distributions. You’ll find detailed installation instructions in our documentation.

Homebrew package for the Incus client

The client tool is available through HomeBrew for both Linux and MacOS.

Chocolatey package for the Incus client

The client tool is available through Chocolatey for Windows users.

Winget package for the Incus client

The client tool is also available through Winget for Windows users.

https://winstall.app/apps/LinuxContainers.Incus

Support

Monthly feature releases are only supported up until the next release comes out. Users needing a longer support length and less frequent changes should consider using Incus 6.0 LTS instead.

Community support is provided at: https://discuss.linuxcontainers.org
Commercial support is available through: Zabbly - Incus services
Bugs can be reported at: GitHub · Where software is built

stgraber · September 26, 2025, 4:54pm

bensmrs · October 2, 2025, 7:52am

Hi, could you please sync the website? Thanks!

stgraber · October 2, 2025, 8:09pm

Oops, I had prepared the commit, just forgot to push it… Done now.

kiv · October 3, 2025, 12:21pm

Hi, not sure if it appeared with this version, but `export` now refuse to delete existing file if it exists : How to back up instances - Incus documentation should be updated accordingly by rewriting the warning.

stgraber · October 3, 2025, 3:25pm