ENT-5725: Added vagrant provider for spawning VMs

[victormlg]

It was a bit hard to know exactly what approach to take so if you have any better ideas I am interested

This doesn't exactly work yet, because

• when creating VMs in a loop with a default name (like dev-1, dev-2, dev-3...) , the vms are created but only the first one created is accessible from the directory where lies the Vagrantfile (I'm using virtualbox)
• when creating VMs with a custom name given by an env variable (ike in the pr) the vm is just not accessible, as if it wasn't created, but is actually present in virtualbox.

I am not sure how we should provision the created vms. We could:

• just leave them unprovisioned
• like I do in the code: let the user input his own provision script
• start the VM unprovisioned, then inspect it using nt-discovery.sh and find all the information we need (like what package manager it uses, etc...) then provision it with a modular script using the collected data.

[craigcomstock]

looks like a great start! I have added some notes on things we have solved in jenkins-vms regarding various bits.

[craigcomstock]

regarding time and dns you might need to include this snippet from jenkins-vms that handles darwin/macos

https://gitlab.com/Northern.tech/CFEngine/jenkins-vms/-/blob/master/scripts/common.rb?ref_type=heads#L104

[craigcomstock]

synced folders are also tricky for older distributions like debian-9:

here are some workarounds for centos-7 and old debians:

https://gitlab.com/Northern.tech/CFEngine/jenkins-vms/-/blob/master/scripts/common.rb?ref_type=heads#L151

[larsewi]

Looks great 🚀 Some smaller comments

While reviewing your PR, I never understood what the config argument to ssh and scp is for, or why it's necessary. Could you explain?

[larsewi]

If you run cf-remote spawn two times, the IPs will conflict?

[victormlg]

For now, yes. I didn't come up with any better solution

[victormlg]

Looks great 🚀 Some smaller comments

While reviewing your PR, I never understood what the config argument to ssh and scp is for, or why it's necessary. Could you explain?

When you spawn a vagrant VM, vagrant generates a ssh config file with a private key and everything set up correctly. You can inspect it by running vagrant ssh-config. When you ssh into a vagrant machine by running vagrant ssh <name>, vagrant is simply doing ssh -F .
Connecting with vanilla ssh was a bit cumbersome because it resulted in a lot of errors and man-in-the-middle attacks warnings and such, so I thought it would be easier to connect by using the ssh config file generated by vagrant

The variable "config" is simply the path to the vagrant-ssh-config file inside the VM's directory. If it is not None, then the connection will connect using it.

[larsewi]

Looks like there are some failing test :)