Troubleshooting Slicer¶

The best place to get help with the Home Edition is via Discord. For Pro users, email support is available, check the welcome email for details.

Doesn't boot right or has a networking issue - perhaps GitHub keys aren't being imported?¶

Look in the log file outputted from slicer. So if you are running i.e. k3s as the host group and VM 1 isn't booting:

sudo cat /var/log/slicer/k3s-1.txt

If you can get into the VM via the SOS console, then run the following:

sudo journalctl -u mount-config --no-pager

Look for networking issues, or a bad routing.

If your networking equipment is forcing the microVMs to use IPv6, but you do not have IPv6 connectivity, then you could try to disable IPv6.

The host is easier to fix and may mean you can leave the microVMs as they are.

sudo sysctl -w net.ipv6.conf.all.disable_ipv6=1
sudo sysctl -w net.ipv6.conf.default.disable_ipv6=1
sudo sysctl -w net.ipv6.conf.lo.disable_ipv6=1

To make it permanent:

Then add the following lines to /etc/sysctl.conf:

net.ipv6.conf.all.disable_ipv6 = 1
net.ipv6.conf.default.disable_ipv6 = 1
net.ipv6.conf.lo.disable_ipv6 = 1

Alternatively, you could try disabling only within the microVM on first boot:

config:
    host_groups:
    - name: k3s
      userdata: |
        #!/bin/bash
        sysctl -w net.ipv6.conf.all.disable_ipv6=1
        sysctl -w net.ipv6.conf.default.disable_ipv6=1
        sysctl -w net.ipv6.conf.lo.disable_ipv6=1

        echo "net.ipv6.conf.all.disable_ipv6 = 1" |tee -a /etc/sysctl.conf
        echo "net.ipv6.conf.default.disable_ipv6 = 1" |tee -a /etc/sysctl.conf
        echo "net.ipv6.conf.lo.disable_ipv6 = 1" |tee -a /etc/sysctl.conf

        # Cause the SSH keys to re re-imported from GitHub on the next boot

        rm -rf /home/ubuntu/.ssh/github_keys_imported

        # Optionally, reboot the VM to re-import the keys from GitHub
        # reboot

If you're having issues reaching GitHub for your SSH keys, you can set them manually in the config or userdata.

The problem may be fixed by upgrading Slicer¶

You can upgrade the Slicer binary by running the instructions at the end of the installation page.

I can't connect over SSH¶

Have a look at the VM's serial console, written to: /var/log/slicer/vm-1.txt where 1 is the VM number, and vm is the host group name.

The VM can't connect to the Internet¶

This can occur when there are old or stable routes in place, for instance, if you added a route for 192.168.137.0/24 on another host, but are now running Slicer on your own workstation.

Check routes with:

sudo ip route

Then delete one with i.e. sudo ip route del 192.168.137.0/24.

I've run out of disk space¶

There are three places to perform a prune/clean-up.

Check there are not a lot of unused .img files from various launches of VMs

sudo find / -name *.img

Delete the ones you no longer require. Beware of deleting .img files for VMs that you still need.

If you've been working with custom images, prune them from the containerd library:

sudo ctr -n slicer i ls

sudo ctr -n slicer i ls -q |xargs sudo ctr -n slicer i rm

If you are using a snapshotter like ZFS or Devmapper, the above command will delete all images and their snapshots.

So to be more selective, you can delete individual images by name:

For instance:

sudo ctr -n slicer i rm docker.io/library/ubuntu:22.04

Remove the /var/run folder for Slicer

The /var/run and /var/log folders contain logs, sockets, temporary disks, and extracted Kernel files. This can build up over time. /var/run is generally ephemeral, and removed on each reboot.

sudo rm -rf /var/run/slicer
sudo rm -rf /var/log/slicer

For the nuclear option, delete all of the containerd's data, this will remove all images, snapshots, including any from containers that you run via Docker, if that's also installed.

sudo systemctl stop containerd
sudo rm -rf /var/lib/containerd
sudo systemctl restart containerd