Windows 11 WSL and Netlab

[snuffy22]

So I have managed to get netlab relatively happy running under WSL without too much work.

Only needed a few tweaks, I might provide a proper writeup if people want it.

TLDR is..

• Ensure we tell it we do not want VirtualBox just because you detected windows.. export VAGRANT_DEFAULT_PROVIDER=libvirt
• Add to /etc/wsl.conf line

[boot]
systemd=true

• Use latest Debian image
• Ensure you have WSL version: 2.7.10.0 (gets you GRE / VXLAN)
• Cannot do MPLS.. but we cant have it all..

[ipspace]

That's great information. Thanks a million. Let me create the documentation framework (I'll add WSL and Apple/ARM information), and then you can add all the relevant information.

[ipspace]

Tiny detail:

• Ensure we tell it we do not want VirtualBox just because you detected windows.. export VAGRANT_DEFAULT_PROVIDER=libvirt

Wondering why you need to set this. libvirt was never the default Vagrant provider, so we always specify the provider explicitely.

[snuffy22]

All i know is if i didn't do that myself it would try talking to virtualbox and die horribly.. 🤷

[ipspace]

All i know is if i didn't do that myself it would try talking to virtualbox and die horribly.. 🤷

Thank you!

Could you do "netlab up" with enough verbosity to see the commands executed? Looking at the libvirt provider definition, the only command that I see could cause that is "vagrant plugin list" expecting to have VirtualBox installed (which would be "interesting").

[snuffy22]

I knew it wasn't fully broken.. It only happens when doing netlab down / netlab down --cleanup on a lab.

So it is only with netlab down the error occurs..
Pasted below is the start and the libvirt section, full output can be found in the other details expansion.

Removing unmanaged devices from topology
Setting environment variable NETLAB_ARGS_VERBOSE=3
Setting environment variable NETLAB_ARGS_CLEANUP=True
Setting environment variable NETLAB_ARGS_SNAPSHOT=netlab.snapshot.pickle
Setting NETLAB_ARGS_TOPOLOGY=/home/netlab/snuffy-netlab/tests/integration/dhcp/03-ipv4-server.yml
┌──────────────────────────────────────────────────────────────────────────────────┐
│ STOPPING libvirt nodes                                                           │
└──────────────────────────────────────────────────────────────────────────────────┘
CLI command down plugin hooks: [] hook: pre_shell_pre_stop_libvirt
CLI hook netlab.down.pre_stop_libvirt: None
CLI command down plugin hooks: [] hook: post_shell_pre_stop_libvirt
provider libvirt: executing vagrant destroy -f
run_command executing: vagrant destroy -f
Vagrant failed to initialize at a very early stage:

Vagrant is unable to use the VirtualBox provider from the Windows Subsystem for
Linux without access to the Windows environment. Enabling this access must be
done with caution and an understanding of the implications. For more information
on enabling Windows access and using VirtualBox from the Windows Subsystem for
Linux, please refer to the Vagrant documentation:

  https://www.vagrantup.com/docs/other/wsl.html
Error executing vagrant destroy -f:
  Command '['vagrant', 'destroy', '-f']' returned non-zero exit status 1.
[FATAL]   netlab down: vagrant destroy -f failed, aborting...
(py-net) netlab@LAPTOP-8380MASB:~/snuffy-netlab/tests/integration/dhcp$

Full output from netlab down

netlab down exec `vagrant destroy -f`

(py-net) netlab@LAPTOP-8380MASB:~/snuffy-netlab/tests/integration/dhcp$ netlab down -vvv --cleanup
Removing unmanaged devices from topology
Setting environment variable NETLAB_ARGS_VERBOSE=3
Setting environment variable NETLAB_ARGS_CLEANUP=True
Setting environment variable NETLAB_ARGS_SNAPSHOT=netlab.snapshot.pickle
Setting NETLAB_ARGS_TOPOLOGY=/home/netlab/snuffy-netlab/tests/integration/dhcp/03-ipv4-server.yml

┌──────────────────────────────────────────────────────────────────────────────────┐
│ CHECKING virtualization provider installation                                    │
└──────────────────────────────────────────────────────────────────────────────────┘
Checking virtualization provider installation
run_command executing: stat /usr/sbin/kvm-ok
... run result: CompletedProcess(args=['stat', '/usr/sbin/kvm-ok'], returncode=0, stdout='  File: /usr/sbin/kvm-ok\n  Size: 3482      \tBlocks: 8          IO Block: 4096   regular file\nDevice:
 8,48\tInode: 293939      Links: 1\nAccess: (0755/-rwxr-xr-x)  Uid: (    0/    root)   Gid: (    0/    root)\nAccess: 2026-07-18 11:02:02.867539465 +1000\nModify: 2022-09-19 11:22:27.000000000
+1000\nChange: 2026-06-30 17:59:45.849446423 +1000\n Birth: 2026-06-30 17:59:45.849446423 +1000\n', stderr='')
run_command executing: /usr/sbin/kvm-ok
... run result: CompletedProcess(args=['/usr/sbin/kvm-ok'], returncode=0, stdout='INFO: /dev/kvm exists\nKVM acceleration can be used\n', stderr='')
run_command executing: which virsh
... run result: CompletedProcess(args=['which', 'virsh'], returncode=0, stdout='/usr/bin/virsh\n', stderr='')
run_command executing: which vagrant
... run result: CompletedProcess(args=['which', 'vagrant'], returncode=0, stdout='/usr/bin/vagrant\n', stderr='')
run_command executing: bash -c 'vagrant plugin list|grep vagrant-libvirt'
... run result: CompletedProcess(args=BoxList(['bash', '-c', 'vagrant plugin list|grep vagrant-libvirt']), returncode=0, stdout='vagrant-libvirt (0.11.2, system)\n', stderr='')
run_command executing: virsh net-list
... run result: CompletedProcess(args=['virsh', 'net-list'], returncode=0, stdout=' Name   State   Autostart   Persistent\n----------------------------------------\n\n', stderr='')
[SUCCESS] libvirt installed and working correctly
Checking virtualization provider installation
run_command executing: containerlab version
... run result: CompletedProcess(args=['containerlab', 'version'], returncode=0, stdout='\n ⣴⡾⠛⠛⠖ ⢠⣶⠟⠛⢷⣦ ⢸⣿⣧  ⣿ ⠘⠛⢻⡟⠛⠛  ⣾⣿⡀  ⣿⡇ ⣿⣿⡄ ⢸⡇ ⣿⡟⠛⠛⠃⢸⣿⠛⠛⣷⡄ \x1b[38;2;0;201;255m⣿⡇\x1b[0m      \x1b[38;2;0
;201;255m⣿⡇\x1b[0m    \n⢸⣿     ⣿⡇   ⣿⡇⢸⣿⠹⣧⡀⣿   ⢸⡇   ⣸⡏⢹⣧  ⣿⡇ ⣿⡏⢿⣄⢸⡇ ⣿⣧⣤⣤ ⢸⣿⣀⣀⣾⠇ \x1b[38;2;0;201;255m⣿⡇⠐⠟⠛⢿⡆\x1b[0m \x1b[38;2;0;201;255m⣿⡷⠛⢿⣆\x1b[0m \n⠘⣿⣄  ⡀ ⢻⣧⡀ ⣠⣿⠃⢸⣿ ⠘⣷⣿   ⢸⡇  ⢠⣿⠷⠶⢿⡆ ⣿⡇ ⣿⡇ ⢻⣾⡇
 ⣿⡇   ⢸⣿⠉⢻⣧⡀ \x1b[38;2;0;201;255m⣿⡇⢰⡟⠛⣻⡇\x1b[0m \x1b[38;2;0;201;255m⣿⡇\x1b[0m \x1b[38;2;0;201;255m⣸⡿\x1b[0m \n ⠈⠙⠛⠛⠉  ⠉⠛⠛⠋⠁ ⠘⠛  ⠈⠛   ⠘⠃  ⠚⠃  ⠘⠛ ⠛⠃ ⠛⠃  ⠙⠃ ⠛⠛⠛⠛⠃⠘⠛  ⠙⠓ \x1b[38;2;0;201;255m⠛⠃⠈⠛⠛⠙⠃
\x1b[0m \x1b[38;2;0;201;255m⠛⠛⠛⠛⠁\x1b[0m \n\n    version: 0.75.0\n     commit: 6a686b7d7\n       date: 2026-04-27T09:35:56Z\n     source: https://github.com/srl-labs/containerlab\n rel. notes:
https://containerlab.dev/rn/0.75/\n', stderr='')
run_command executing: bash -c '[[ `containerlab version|awk '"'"'/version/ {print $2}'"'"'` > '"'"'0.74.99'"'"' ]] && echo OK'
... run result: CompletedProcess(args=BoxList(['bash', '-c', "[[ `containerlab version|awk '/version/ {print $2}'` > '0.74.99' ]] && echo OK"]), returncode=0, stdout='OK\n', stderr='')
[SUCCESS] clab installed and working correctly
CLI command down plugin hooks: [] hook: pre_shell_pre_stop_lab
CLI hook netlab.down.pre_stop_lab: None
CLI command down plugin hooks: [] hook: post_shell_pre_stop_lab

┌──────────────────────────────────────────────────────────────────────────────────┐
│ STOPPING clab nodes                                                              │
└──────────────────────────────────────────────────────────────────────────────────┘
CLI command down plugin hooks: [] hook: pre_shell_pre_stop_clab
CLI hook netlab.down.pre_stop_clab: None
CLI command down plugin hooks: [] hook: post_shell_pre_stop_clab
provider clab: executing sudo -E containerlab destroy --cleanup -t clab-augment.yml
run_command executing: sudo -E containerlab destroy --cleanup -t clab-augment.yml
11:15:16 INFO Parsing & checking topology file=clab-augment.yml
11:15:16 INFO Parsing & checking topology file=clab-augment.yml
11:15:16 INFO Destroying lab name=dhcp
11:15:17 INFO Removed container name=clab-dhcp-dut
11:15:17 INFO Removing host entries path=/etc/hosts
11:15:17 INFO Removing SSH config path=/etc/ssh/ssh_config.d/clab-dhcp.conf
... run result: CompletedProcess(args=['sudo', '-E', 'containerlab', 'destroy', '--cleanup', '-t', 'clab-augment.yml'], returncode=0)
post-stop hook for Containerlab, cleaning up any bridges
CLI command down plugin hooks: [] hook: pre_shell_post_stop_clab
CLI hook netlab.down.post_stop_clab: None
CLI command down plugin hooks: [] hook: post_shell_post_stop_clab


┌──────────────────────────────────────────────────────────────────────────────────┐
│ STOPPING libvirt nodes                                                           │
└──────────────────────────────────────────────────────────────────────────────────┘
CLI command down plugin hooks: [] hook: pre_shell_pre_stop_libvirt
CLI hook netlab.down.pre_stop_libvirt: None
CLI command down plugin hooks: [] hook: post_shell_pre_stop_libvirt
provider libvirt: executing vagrant destroy -f
run_command executing: vagrant destroy -f
Vagrant failed to initialize at a very early stage:

Vagrant is unable to use the VirtualBox provider from the Windows Subsystem for
Linux without access to the Windows environment. Enabling this access must be
done with caution and an understanding of the implications. For more information
on enabling Windows access and using VirtualBox from the Windows Subsystem for
Linux, please refer to the Vagrant documentation:

  https://www.vagrantup.com/docs/other/wsl.html
Error executing vagrant destroy -f:
  Command '['vagrant', 'destroy', '-f']' returned non-zero exit status 1.
[FATAL]   netlab down: vagrant destroy -f failed, aborting...
(py-net) netlab@LAPTOP-8380MASB:~/snuffy-netlab/tests/integration/dhcp$

Here is the 'up' shrunk down for reference..

netlab up with a libvirt device

(py-net) netlab@LAPTOP-8380MASB:~/snuffy-netlab/tests/integration/dhcp$ netlab up -vvv -p clab -d eos 03-ipv4-server.yml

┌──────────────────────────────────────────────────────────────────────────────────┐
│ CREATING configuration files                                                     │
└──────────────────────────────────────────────────────────────────────────────────┘
CLI command create plugin hooks: [] hook: output
Removing unmanaged devices from topology
[INFO]    Rendered initial template for c1 into c1/initial
[INFO]    Rendered dhcp template for c1 into c1/dhcp
[INFO]    Rendered initial template for c2 into c2/initial
[INFO]    Rendered dhcp template for c2 into c2/dhcp
[INFO]    Rendered normalize template for dut into dut/normalize
[INFO]    Rendered initial template for dut into dut/initial
[INFO]    Rendered dhcp template for dut into dut/dhcp
Removing unmanaged devices from topology
[CREATED] provider configuration file: Vagrantfile
[INFO]    Creating configuration file for secondary provider clab
[CREATED] provider configuration file: clab-augment.yml
[CREATED] pickled transformed topology data into netlab.snapshot.pickle
Removing unmanaged devices from topology
[GROUPS]  group_vars for all
[GROUPS]  group_vars for modules
[GROUPS]  group_vars for linux
[HOSTS]   host_vars for c1
[HOSTS]   host_vars for c2
[GROUPS]  group_vars for eos
[HOSTS]   host_vars for dut
[CREATED] minimized Ansible inventory hosts.yml
[CREATED] Ansible configuration file: ansible.cfg
CLI command create plugin hooks: [] hook: post_output
Removing unmanaged devices from topology
Setting NETLAB_ARGS_TOPOLOGY=/home/netlab/snuffy-netlab/tests/integration/dhcp/03-ipv4-server.yml
CLI command up plugin hooks: [] hook: pre_shell_pre_probe
CLI hook netlab.up.pre_probe: None
CLI command up plugin hooks: [] hook: post_shell_pre_probe

┌──────────────────────────────────────────────────────────────────────────────────┐
│ CHECKING virtualization provider installation                                    │
└──────────────────────────────────────────────────────────────────────────────────┘
Checking virtualization provider installation
run_command executing: stat /usr/sbin/kvm-ok
... run result: CompletedProcess(args=['stat', '/usr/sbin/kvm-ok'], returncode=0, stdout='  File: /usr/sbin/kvm-ok\n  Size: 3482      \tBlocks: 8          IO Block: 4096   regular file\nDevice:
 8,48\tInode: 293939      Links: 1\nAccess: (0755/-rwxr-xr-x)  Uid: (    0/    root)   Gid: (    0/    root)\nAccess: 2026-07-18 11:02:02.867539465 +1000\nModify: 2022-09-19 11:22:27.000000000
+1000\nChange: 2026-06-30 17:59:45.849446423 +1000\n Birth: 2026-06-30 17:59:45.849446423 +1000\n', stderr='')
run_command executing: /usr/sbin/kvm-ok
... run result: CompletedProcess(args=['/usr/sbin/kvm-ok'], returncode=0, stdout='INFO: /dev/kvm exists\nKVM acceleration can be used\n', stderr='')
run_command executing: which virsh
... run result: CompletedProcess(args=['which', 'virsh'], returncode=0, stdout='/usr/bin/virsh\n', stderr='')
run_command executing: which vagrant
... run result: CompletedProcess(args=['which', 'vagrant'], returncode=0, stdout='/usr/bin/vagrant\n', stderr='')
run_command executing: bash -c 'vagrant plugin list|grep vagrant-libvirt'
... run result: CompletedProcess(args=BoxList(['bash', '-c', 'vagrant plugin list|grep vagrant-libvirt']), returncode=0, stdout='vagrant-libvirt (0.11.2, system)\n', stderr='')
run_command executing: virsh net-list
... run result: CompletedProcess(args=['virsh', 'net-list'], returncode=0, stdout=' Name   State   Autostart   Persistent\n----------------------------------------\n\n', stderr='')

[SUCCESS] libvirt installed and working correctly

Checking virtualization provider installation
run_command executing: containerlab version
... run result: CompletedProcess(args=['containerlab', 'version'], returncode=0, stdout='\n ⣴⡾⠛⠛⠖ ⢠⣶⠟⠛⢷⣦ ⢸⣿⣧  ⣿ ⠘⠛⢻⡟⠛⠛  ⣾⣿⡀  ⣿⡇ ⣿⣿⡄ ⢸⡇ ⣿⡟⠛⠛⠃⢸⣿⠛⠛⣷⡄ \x1b[38;2;0;201;255m⣿⡇\x1b[0m      \x1b[38;2;0
;201;255m⣿⡇\x1b[0m    \n⢸⣿     ⣿⡇   ⣿⡇⢸⣿⠹⣧⡀⣿   ⢸⡇   ⣸⡏⢹⣧  ⣿⡇ ⣿⡏⢿⣄⢸⡇ ⣿⣧⣤⣤ ⢸⣿⣀⣀⣾⠇ \x1b[38;2;0;201;255m⣿⡇⠐⠟⠛⢿⡆\x1b[0m \x1b[38;2;0;201;255m⣿⡷⠛⢿⣆\x1b[0m \n⠘⣿⣄  ⡀ ⢻⣧⡀ ⣠⣿⠃⢸⣿ ⠘⣷⣿   ⢸⡇  ⢠⣿⠷⠶⢿⡆ ⣿⡇ ⣿⡇ ⢻⣾⡇
 ⣿⡇   ⢸⣿⠉⢻⣧⡀ \x1b[38;2;0;201;255m⣿⡇⢰⡟⠛⣻⡇\x1b[0m \x1b[38;2;0;201;255m⣿⡇\x1b[0m \x1b[38;2;0;201;255m⣸⡿\x1b[0m \n ⠈⠙⠛⠛⠉  ⠉⠛⠛⠋⠁ ⠘⠛  ⠈⠛   ⠘⠃  ⠚⠃  ⠘⠛ ⠛⠃ ⠛⠃  ⠙⠃ ⠛⠛⠛⠛⠃⠘⠛  ⠙⠓ \x1b[38;2;0;201;255m⠛⠃⠈⠛⠛⠙⠃
\x1b[0m \x1b[38;2;0;201;255m⠛⠛⠛⠛⠁\x1b[0m \n\n    version: 0.75.0\n     commit: 6a686b7d7\n       date: 2026-04-27T09:35:56Z\n     source: https://github.com/srl-labs/containerlab\n rel. notes:
https://containerlab.dev/rn/0.75/\n', stderr='')
run_command executing: bash -c '[[ `containerlab version|awk '"'"'/version/ {print $2}'"'"'` > '"'"'0.74.99'"'"' ]] && echo OK'
... run result: CompletedProcess(args=BoxList(['bash', '-c', "[[ `containerlab version|awk '/version/ {print $2}'` > '0.74.99' ]] && echo OK"]), returncode=0, stdout='OK\n', stderr='')

[SUCCESS] clab installed and working correctly
run_command executing: vagrant box list
... run result: CompletedProcess(args=['vagrant', 'box', 'list'], returncode=0, stdout='bento/ubuntu-24.04 (libvirt, 202508.03.0)\ndebian/bookworm64  (libvirt, 12.20260519.1)\n', stderr='')
libvirt: validating node c1 image bento/ubuntu-24.04
libvirt: validating node c2 image bento/ubuntu-24.04
clab: validating node dut image ceos:4.36.1F
run_command executing: docker image ls --format json ceos:4.36.1F
... run result: CompletedProcess(args=['docker', 'image', 'ls', '--format', 'json', 'ceos:4.36.1F'], returncode=0, stdout='{"Containers":"0","CreatedAt":"2026-06-30 18:17:30 +1000 AEST","Create
dSince":"2 weeks ago","Digest":"\\u003cnone\\u003e","ID":"b47f5f2e6197","Repository":"ceos","SharedSize":"N/A","Size":"3.72GB","Tag":"4.36.1F","UniqueSize":"N/A"}\n', stderr='')

┌──────────────────────────────────────────────────────────────────────────────────┐
│ STARTING libvirt nodes                                                           │
└──────────────────────────────────────────────────────────────────────────────────┘
CLI command up plugin hooks: [] hook: pre_shell_pre_start_lab
CLI hook netlab.up.pre_start_lab: None
CLI command up plugin hooks: [] hook: post_shell_pre_start_lab
CLI command up plugin hooks: [] hook: pre_shell_pre_start_libvirt
CLI hook netlab.up.pre_start_libvirt: None
CLI command up plugin hooks: [] hook: post_shell_pre_start_libvirt
pre-start hook for libvirt
run_command executing: vagrant status --machine-readable
run_command executing: virsh net-destroy vagrant-libvirt
run_command executing: virsh net-undefine vagrant-libvirt
run_command executing: sudo ip link delete libvirt-mgmt
[CREATED] creating libvirt management network vagrant-libvirt
run_command executing: virsh net-define /tmp/tmpmvbyq3ho
... run result: CompletedProcess(args=['virsh', 'net-define', '/tmp/tmpmvbyq3ho'], returncode=0, stdout='Network vagrant-libvirt defined from /tmp/tmpmvbyq3ho\n\n', stderr='')
provider libvirt: executing vagrant up --provider libvirt
run_command executing: vagrant up --provider libvirt
Bringing machine 'c1' up with 'libvirt' provider...
Bringing machine 'c2' up with 'libvirt' provider...

[ipspace]

I knew it wasn't fully broken.. It only happens when doing netlab down / netlab down --cleanup on a lab.

Thanks a million for chasing this gremlin!

This is even worse. The VMs are up, they were started with the libvirt provider, and yet Vagrant tries to reach VirtualBox. Disgusting :(

[snuffy22]

This code snippet into netsim/providers/libvirt.py fixed it

 def pre_stop_lab(self, topology: Box) -> None:
    log.print_verbose('pre-stop hook for libvirt')
    os.environ["VAGRANT_DEFAULT_PROVIDER"] = "libvirt"            # Ensure to set libvirt as provider to use

[ipspace]

Perfect. Thanks a million! Have no problem with that being in the code ;) Please submit a PR.