This commit attempts to add tests that confirm the nameserver is getting
passed in correctly in multiple ways.
- via ip=${ip}::${gateway}:${netmask}:${hostname}:${devname}:none:${nameserver}
- via nameserver=${nameserver}
- specifically see bug https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/issues/391
It also tries to verify the nameserver got propagated into either
resolv.conf (RHEL 8 and Fedora less than 32) or systemd-resolved (F33+).
I saw some weird behavior where the reboot was actually managing
to start the VM as well, which caused the `virsh start` to fail.
Let's just make things clearer by using an explicit `shutdown` and
then `start`.
Now that RHCOS uses Ignition spec V3 let's drop the use of
`ign-converter`. We leave the important bits in there as a comment
in case the user wants to uncomment and run a test against <4.6.
Now that we are using persistent NIC naming in FCOS we'll update the
test to also use those interface names. See:
https://github.com/coreos/fedora-coreos-tracker/issues/484
Also add a test that makes sure that net.ifnames=0 works as desired.
This makes it so we can call coreos-network-testing.sh with either
a rhcos qcow or a fcos qcow. Depending on the name of the file the
script will act differently.
This is how I've been doing network testing for the NM in
the initrd PR (https://github.com/coreos/fedora-coreos-config/pull/310).
One day I'd like to convert this into some more sophisticated
test that can be run in our CI but AFAIU right now that's not
possibly because our CI is limited to qemu unprivileged networking.
For now let's put this $somewhere so that we can pull it out
at least when we suspect problems.
