in 16.9.4 ip domain timeout is removed, as its now based on the RTD of nameservers - and that's probably good if it worked, but check out this:
08:40:00.570 CET Tue Dec 22 2020
% Unrecognized host or address, or protocol not running.
08:42:41.927 CET Tue Dec 22 2020
So roughly 150 seconds before timeout - that really sucks. Nameservers are reached via DMVPN, but I can't see how that should be the issue?
When I lookup existing hosts, it all works fine and quickly - anyone else seeing this?
Im using C1111-8P for this
Not sure i am able to understand the issue :
do you have DNS configured, and is that reachable? then it should resolve, this is more of host entry by localhost not by FQDN
Generally, depends on TTL and every 3 seconds it retries - but depends on IOS what default set here.
you need to have Local DNS also configured in case not able to reach the global one it can reach local DNS for your host entries.
Yes, DNS is configured, and working fine for stuff which actually exists - like google.com or similar. so if I do 'ping somethingnonexisting' then it will try to resolve 'somethingsnonexisting' and this part will take up to 150 seconds to fail, which I'd like to cut down to ~2-3 seconds if possible.
If I do 'ping google.com' it resolves it nice and fast.
This is mostly to avoid a hanging console when a typo is made, as waiting 150 seconds every time a typo happens is not super effective..
Its an internal DNS server(s) but its close by:
Success rate is 100 percent (5/5), round-trip min/avg/max = 20/23/24 ms
and responds super fast, whenever the query is successful - its only when I'm querying bogus stuff it takes forever (150 secs)