Error: Failed to create network 'lxdbr0': Failed to automatically find an unused IPv4 subnet, manual configuration required

hsbawa · June 30, 2018, 3:07am

LXD init on Ubuntu VM returns following error:
Error: Failed to create network 'lxdbr0': Failed to automatically find an unused IPv4 subnet, manual configuration required

lxc info:

config: {}
api_extensions:

storage_zfs_remove_snapshots

container_host_shutdown_timeout

container_stop_priority

container_syscall_filtering

auth_pki

container_last_used_at

etag

patch

usb_devices

https_allowed_credentials

image_compression_algorithm

directory_manipulation

container_cpu_time

storage_zfs_use_refquota

storage_lvm_mount_options

network

profile_usedby

container_push

container_exec_recording

certificate_update

container_exec_signal_handling

gpu_devices

container_image_properties

migration_progress

id_map

network_firewall_filtering

network_routes

storage

file_delete

file_append

network_dhcp_expiry

storage_lvm_vg_rename

storage_lvm_thinpool_rename

network_vlan

image_create_aliases

container_stateless_copy

container_only_migration

storage_zfs_clone_copy

unix_device_rename

storage_lvm_use_thinpool

storage_rsync_bwlimit

network_vxlan_interface

storage_btrfs_mount_options

entity_description

image_force_refresh

storage_lvm_lv_resizing

id_map_base

file_symlinks

container_push_target

network_vlan_physical

storage_images_delete

container_edit_metadata

container_snapshot_stateful_migration

storage_driver_ceph

storage_ceph_user_name

resource_limits

storage_volatile_initial_source

storage_ceph_force_osd_reuse

storage_block_filesystem_btrfs

resources

kernel_limits

storage_api_volume_rename

macaroon_authentication

network_sriov

console

restrict_devlxd

migration_pre_copy

infiniband

maas_network

devlxd_events

proxy

network_dhcp_gateway

file_get_symlink

network_leases

unix_device_hotplug

storage_api_local_volume_handling

operation_description

clustering

event_lifecycle

storage_api_remote_volume_handling

nvidia_runtime
api_status: stable
api_version: “1.0”
auth: trusted
public: false
auth_methods:

tls
environment:
addresses:
architectures:

x86_64

i686
certificate: |
-----BEGIN CERTIFICATE-----
-----END CERTIFICATE-----
certificate_fingerprint: xxxxxxxxxxxxxxxx
driver: lxc
driver_version: 3.0.1
kernel: Linux
kernel_architecture: x86_64
kernel_version: 4.4.0-127-generic
server: lxd
server_pid: 1815
server_version: 3.0.1
storage: “”
storage_version: “”
server_clustered: false
server_name: virtualserver01

simos · June 30, 2018, 1:56pm

It happened to me as well the other day on a fresh installation.
Lxd init was not able to figure out the IP for lxdbr0 and gave up. It would be good to figure out the exact details in order to fix in LXD init.

I am on mobile now; you can create manually lxdbr0 with lxc network, the add the interface to the default LXD profile, using lxc profile add…

hsbawa · July 1, 2018, 4:18am

Thanks @simos.

Even with manual create, getting same error.

root@myserver:~# lxc network create lxdbr0
Error: Failed to automatically find an unused IPv4 subnet, manual configuration required
root@myserver:~#

simos · July 1, 2018, 5:24am

You need to specify the rest of the details manually I think you need to add ipv4.address=10.10.10.1/24.

hsbawa · July 2, 2018, 3:21am

@simos Thanks. That worked. I am still curious why it didn’t work with lxd init.

I am guessing error was because of device name conflict with eth0 (as default) which is already in use (along with eth1), in my case.

Just a thought: if internal subnet and device name to be attached to, can be added as an option for LXD init?

simos · July 2, 2018, 7:54am

Hi!

Here is the code that lxd init will use to find a random available IPv4 network,

github.com

lxc/lxd/blob/master/lxd/networks_utils.go#L357




		// Check if we have a route to our new subnet
		if lineNet.Contains(subnet.IP) {
			return true
		}
	}


	return false
}


func networkRandomSubnetV4() (string, error) {
	for i := 0; i < 100; i++ {
		cidr := fmt.Sprintf("10.%d.%d.1/24", rand.Intn(255), rand.Intn(255))
		_, subnet, err := net.ParseCIDR(cidr)
		if err != nil {
			continue
		}


		if networkInRoutingTable(subnet) {
			continue
		}

You can see that it tries with 100 random subnets before it gives up.
For each randomly created network, it checks

If the network is in the routing table
If the network can be pinged

and does that for both IPv4 and IPv6.

Someone can add some print statements in the code to see where it fails (either the random networks are not random, either the check if the network is already in the routing table fails, either the pinging of the subnet has some weird issue and LXD gets replies from anything), either one of IPv4 or IPv6 is not available on the host.

The code that uses these functions is

github.com

lxc/lxd/blob/master/lxd/networks_config.go#L183


						return fmt.Errorf("Maximum MTU for a VXLAN FAN bridge is 1450")
					}
				}
			}
		}
	}


	return nil
}


func networkFillAuto(config map[string]string) error {
	if config["ipv4.address"] == "auto" {
		subnet, err := networkRandomSubnetV4()
		if err != nil {
			return err
		}


		config["ipv4.address"] = subnet
	}


	if config["ipv6.address"] == "auto" {

so if you run lxd init, and press Enter all the way while there is no, for example, IPv6 support, then you get the error because in one of the Enters you select auto for IPv6.

lxd init can also accept a pre-seed file, therefore you can add there the network instead of asking LXD to find one randomly. Or, when you use the wizard lxd init, do not press Enter to accept auto but specify the network there. If, for example, there is no IPv6 support, type none instead.

Therefore, in your case, find the reason why auto fails in LXD init (network subnet configuration) for IPv4 and IPv6, and supply the appropriate values either at lxd init or with a LXD preseed file.

In the worst case, you can select none for the networking and then create the network later and finally add it to the lxd profile. In that way, you can get a nice 10.10.10.1 network.

stgraber · July 3, 2018, 6:26pm

I suspect the issue you ran into is that you had a route for 10.0.0.0/8 on your machine, effectively marking the entirety of the 10.0.0.0/8 RFC1918 space as directly attached.

In such a case, LXD cannot safely find any subnet for you to use, making manual configuration the only safe option. That’s pretty uncommon though as having a 255.0.0.0 netmask is a terrible idea in general since it means you may have up to 16 million devices talking to you directly (with all the possible collisions and needless ARP/broadcast that this would cause).

hsbawa · July 4, 2018, 6:08am

@stgraber You are correct. I checked network interface (eth0) and it has exact same configuration for the route.

512YB · April 14, 2021, 1:52pm

Tried to do so, it worked, but containers attached to such bridge doesn’t see the internet. I can only ping the addresses of host’s enp2s0 from containers.
So adding ipv4.address=w.x.y.z to the lxc network create ... doesn’t really help. Any ideas?

tunc · October 14, 2021, 1:39pm

add ipv4.nat=true to network

Salih · November 18, 2023, 5:58pm

I just had the same issue on a testing VPS Ubuntu 22.04 and reading @stgraber post at Network management with LXD (2.3+) | Stéphane Graber's website

I first checked the ip addr

ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: ens3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
    link/ether 00:00:59:2f:a3:05 brd ff:ff:ff:ff:ff:ff
    altname enp0s3
    inet ------------------ brd ------------- scope global ens3
       valid_lft forever preferred_lft forever
    inet 10.47.163.5/8 brd 10.255.255.255 scope global ens3:1
       valid_lft forever preferred_lft forever
    inet6 ----------------------------------- scope global
       valid_lft forever preferred_lft forever
    inet6 ----------------------------------- scope link
       valid_lft forever preferred_lft forever
3: docker0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default
    link/ether 02:42:30:7f:97:94 brd ff:ff:ff:ff:ff:ff
    inet 172.17.0.1/16 brd 172.17.255.255 scope global docker0
       valid_lft forever preferred_lft forever

then created a “lxdbr0” network device

lxc network create lxdbr0 ipv6.address=auto ipv4.address=10.47.163.6/24 ipv4.nat=true

then I procceded with

sudo lxd init

which asked the network related questions I changed the default answer

Would you like to create a new local network bridge? (yes/no) [default=yes]: no
Would you like to configure LXD to use an existing bridge or host interface? (yes/no) [default=no]: yes
Name of the existing bridge or host interface: lxdbr0

And the configuration went trhorugh succesfully.

Just a question:
Is there a problem using the same IP subnet with inet 10.47.163.x ?
Any suggestions ?