[dns-operations] Best practices for Linux/UNIX stub resolver failover

[Chris Adams]

Once upon a time, Brett Carr <Brett.Carr at nominet.org.uk> said:
> If I stop the resolver running on 192.168.1.244 and then generate a query I see the following in a tcpdump

That tests just one failure mode: the system is running but the resolver
is refusing connections.  That particular case has never been a problem,
because the server will immediately send back a "connection refused"
packet and the client can go on to the next server.

Try setting the first entry in /etc/resolv.conf to an IP address that is
not in use, simulating a server down, a resolver running but not
responding, etc.  The client will wait, send restries, etc., which takes
a while to time out before moving to the next server.
-- 
Chris Adams <cma at cmadams.net>