I tried running the instructions for a START1-M instance at Scaleway

$ qemu-img create -f qcow2 MyDisk.qcow2 64G

Command 'qemu-img' not found, but can be installed with:

sudo apt install qemu-utils

Also, the KVM prerequisite can be installed and configured as de

In some cases, it is crucial to set performance options for a VM's disk. For that, you currently have to use virt-manager or virsh. This should be implemented into the panel.

The mentioned features are:

• [IO-Mode](https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/7/html/Virtualization_Tuning_and_Optimization_Guide/sect-Virtualization_Tuning_Optimization_Guide-BlockI

Kata shimv2 does not have any tracing yet. Will be useful to add this.

from #992:

The code in generate_manifest in the mirage tool should check that device names are unique and alphanumeric

from Solo5/solo5#372

NAME must be composed of alphanumeric characters only, and within 1..67 characters in length

A gui for ssh-copy-id maybe?

and as the tittle says, support for ed25519 keys, maybe a dropdown menu that lets you choose the ehh... cipher is it? for generating a new one. Or at least the ability to connect using them.

If you are in the ignite-spawn container, you have root and can change host files like /etc/resolv.conf using ignite <= v0.6.1 || v.0.5.3.

The Bind interface should be updated to support ro/rw or arbitrary mount options.
IMO specifying this should be required.

Document that seccomp is on and how to disable it along with a pointer on how to identify missing syscalls during development.

https://github.com/dmacvicar/terraform-provider-libvirt/blob/8cbe9ad08b26eb53ae212c28aca40f1eff5d7350/examples/v0.12/ubuntu/cloud_init.cfg#L15

I don't need no help with that of course, but this template better be fixed.

We should add a CLI argument to add extra lines of configuration when the bootstrap configuration is applied to the device.

vrnetlab has no aspiration of becoming a generic configuration management system. The focus has always been, and will continue to be, on doing the minimal amount of configuration so that the router can then be configured through NETCONF/YANG. vrnetlab configures this boots

(...)

Having said that, I can see how _acquire_ implies a transition of some kind, so perhaps it'd be better to just change the docs (solo5.h) to say that you're only allowed to call it once per handle?

Originally posted by @mato in Solo5/solo5#402

the primary instance conversion procedure in the evacuating node admin docs is ambiguous. it seems to say that, to evacuate a node, you first need to switch all the primary instances to their secondaries:

There are two steps of moving i

Ubuntu 20.04 LTS (Focal Fossa) just has been released. Add the ISO to the list of images that Ravada can download.

• Xubuntu 20.04 64 bits
• Kubuntu 20.04 64 bits
• Lubuntu 20.04 64 bits
• Mate 20.04 64 bits
• Ubuntu 20.04 64 bits

This issue #1108 can be useful as a guide if someone is encouraged to fix it.

It becomes very complicated start to use and contribute to this project, because it lacks documentation, apparently referring to unit testing, as if it were an appropriate way to start, but they are not.

Maybe for beginners like me it is complicated, but the idea is that everyone can contribute, or at least try.

I've been reading the issues about using nginx as reverse proxy, so there're people where it seems to work.
The problem is that there isn't enough info to set it up properly.
My setup:

nginx:

location / {
    # First attempt to serve request as file, then
    # as directory, then fall back to displaying a 404.
    #try_files $uri $uri/ =404;
    proxy_pass htt