Bind and systemd-resolved

[Reindl Harald]

Am 18.04.22 um 07:26 schrieb Leroy Tennison via bind-users:
> When I attempt “dig -t AXFR office.example.com -k 
> Kexample_dns.+157+18424.key” on the DNS server (Bind 9.11) sudoed to 
> root I get:
> 
> ;; Couldn't verify signature: expected a TSIG or SIG(0)
> ; Transfer failed.
> 
> This is an Ubuntu 18.04 system and /etc/systemd/resolved.conf has 
> DNS=127.0.0.1 since the DNS server is running on it.  Systemd-resolved 
> has been restarted afterward.  I've tried using an actual interface 
> address but it doesn't help.  It seems dig tries to use 127.0.0.53 due 
> to its being in /etc/resolv.conf and that fails even though dig for 
> forward/reverse lookups works.
> 
> If I add @127.0.0.1 to the above it works.  Is there a way to get this 
> to work without having to do that and not setting up the entire network 
> configuration using systemd.  I realize it's not a big effort to add 
> @127.0.0.1 but the reason for the issue is obscure, the error message is 
> misleading and my distaste for systemd is sufficient enough that I would 
> prefer avoiding it as much as possible.  Thanks for any input

so why don't you just disable systemd-resolved?

i run Fedora everywhere in production and on workstations, have masked 
it and after "chattr +i /etc/resolv.conf" nothing messes up resolv.conf 
(even without resolvd existing it would have the immutable flag to 
prevent the dhcp client fpr the WAN interface assign the broken ISP 
resolvers)

[root at srv-rhsoft:~]$ systemctl status systemd-resolved.service
○ systemd-resolved.service
      Loaded: masked (Reason: Unit systemd-resolved.service is masked.)
      Active: inactive (dead)