Hi Matthew,

matthew green wrote:
> 
> jason thorpe fixed this somewhat about a month ago (with me
> looking over his shoulder :-) and extended the getnet*(3)
> functions to use the resolv.conf "lookup file bind" stuff,
> but there are still cases where this will wait for the
> timeout.

Hmm, can you elaborate on this point a little more?
In what cases will route still hang? I guess in the
cases where the needed information is not found in the
neccessary file (/etc/hosts?). IMHO, this is a perfectly
valid behaviour. Same things happens if you invoke 
nslookup and you dns-server is not answering.

Either you have a working dns, then you should use dns,
or you don't, in which case you'll have to use the files
for lookup. Or you don't do any lookup at all.

> i'm not aware of any other way to stop this from waiting for
> the timeout than ripping the DNS code out of the getnet*(3)
> functions....  i'd hope that i'm missing something here ..

Well, this might be annoying, but I don't see a general fix for
the missing dns-server problem, besides shortening the timeout
period, in which case you'll have unneeded failures.

Or do you think that the name-server problem is special in
conjuntion with the route command? Maybe we can have a -n switch
for route like we have it for netstat (that is, if we don't
have this already, I can't check right now)?

  Guenther