One of the googled responses I read last night, the guy has dnsmasq running. He did not know that it had been installed along with some network management software. Also, of course, there was Andy's investigation earlier when he used a different program to deliberately block 53. I think I will stick to trusting it and if there is a problem looking at lsof -i tcp:53 and udp then kill manually. I could then see what is doing the blocking
Hi Tim,
On Wed, Jul 24, 2019 at 05:00:01PM +0100, Tim Dickson wrote:
> bind_stop() {
[…]
> sleep 1
> if ps axc | grep -q named ; then
> echo "Using "killall named" on additional BIND processes..."
> /bin/killall named 2> /dev/null
> fi
> }
>
> I would guess that systemd does not have the functionality of the
> belt'n'braces bit at the end
Interesting. Clearly someone has been here before.
It seems extremely charitable to call that "functionality"! :)
It's not going to be hard to set the systemd service unit to
unconditionally kill any process called "named" when it stops the
service, other than that it's horrible and I'm thinking that no
distribution maintainer would accept such a patch.
Also there is the point that it doesn't have to be something called
"named" that binds to the port.
I'm still thinking this is a deficiency in bind9, but at least I can
add it to my list of things to check, now…
Cheers,
Andy
--
https://bitfolk.com/ -- No-nonsense VPS hosting
_______________________________________________
users mailing list
users@lists.bitfolk.com
https://lists.bitfolk.com/mailman/listinfo/users