Since I still do not know what needs to be entered to get lokinet’s DNS working after reboot (after one week).
I would very thankful if you just simply drop me the simple one-liner solution for my simple question even if this proofs that I still need to learn a lot about DNS, Qubes DNS workflow and how to manipulate it. That is fine for me, I still want to learn more about QOS
I tested the official troubleshooting recommendations (for normal Linux distros) with your recommendation:
echo ‘nameserver 127.3.2.1’ > /etc/resolvconf/resolv.conf.d/head
I also did some tests with
echo ‘nameserver 127.3.2.1’ > /etc/resolv.conf
echo ‘nameserver 127.3.2.1’ > /etc/resolvconf/resolv.conf.d/original
As already stated here:
and here:
I guess this symlink is the issue here but I have no clue where to start here. I never used / worked with symlink before but what I understood it is simply a link which points to a directory.
I tried your second recommendation: using bind-dirs
As descripted in the docs I made a 50_user.conf and added
/etc/systemd/system
with that I hoped to keep the symlink persistent (which was made during the installation routine). I also did this test with a lokinet-gw and a lokinet-ws setup instead of the previous standalone Qube tests. Both still without success.
Accidentally, I saw @qubist nice Guide on:
He also used your recommended /rw/config/rc.local
approach:
echo 'nameserver 127.0.0.1' > /etc/resolv.conf
But he used it in a much more complex content and I guess in my case the issue is strongly linked to the way lokinet implemented its DNS.
@qubist maybe you have an additional hint / solution for me. As described here the issue can be reproduced very quickly (in 5 mins).