RE: [h-e-w] gnuserv and loopback IP address

[John J. Lee]

I checked, and a gnuserv process is running after emacs starts up.  So it
seems to be the internet socket connection that's failing.  I've tried
this with and without the MS loopback adapter installed.

(my description of the other symptoms is appended)

I don't know the registry: how would I find out where the right entry is?


John

On Fri, 14 Jun 2002, John J. Lee wrote:
> On Thu, 13 Jun 2002, Guy Gascoigne-Piggford wrote:
> [responding to the statement that the MS loopback adapter is not required
> to use gnuserv]
> > That's not completely true.  If you have a single adapter on NT and you
> > are disconnected from the network then you stack goes into a
> > non-operative mode and so 127.0.0.1 stops working.  You can get around
> > this by disabling the auto net detection (there's a reg entry that I've
> > forgotten to do this) or by adding a loopback adapter which dodges the
> > issue by always giving you a network connection that is working.
>
> Well, I've now tried with and without the loopback adapter installed, and
> I still can't get gnuserv to do anything useful:
>
> D:\john\info>gnuclientw test
>
> where test is a text file in the current working directory, returns to the
> prompt immediately, and nothing seems to happen.  A few tens of seconds
> later, a new emacs process (not just a new frame) starts up, visiting the
> *scratch* buffer.  If no emacs process was running to start with, the
> *scratch* buffer contains the 'Welcome to GNU Emacs' message, with the
> usual 'Loading image...done' message in the minibuffer; otherwise, the
> *scratch* buffer contains the usual ';; This buffer is for notes you don't
> want to save' text, with the (abnormal) message 'Server subprocess
> exited' in the minibuffer.
>
> Help!
>
> I'm using emacs 21.1.  I do have an ethernet card, but it's not connected
> to anything, and I'm not certain it's configured correctly since I've
> never tried to use it from Windows.  Manually running the
>
> (require 'gnuserv)
> (gnuserv-start)
>
> in my .emacs gives t in both cases, with some kind of confirmatory message
> from one of them (too fast to read easily).  The library is on the load
> path, and my .emacs is certainly being read, and gives no errors on
> startup.