🚀 go-pugleaf

On Sunday, in article
     <20120805.0858.12080059snz@i-m-t.demon.co.uk>
     v$af$ppint@i-m-t.demon.co.uk "ppint. at pplay" wrote:

>         - hi; in article, <20120804.1418.930196snz@twoplaces.co.uk>,
>             simon@twoplaces.co.uk "Simon Turner" clarified [0]:
> >                     "ppint. at pplay" wrote:
> >>           simon@twoplaces.co.uk "Simon Turner" elaborated:

> >...; the packet driver interface connects to packet driver
> >software that is itself talking to a network card (this is what my
> >build uses to access DSL);
> 
>         - the packet driver interface is what'd be needed, to use a
>         network card with dos driver, to use a dos desktop for broad-
>         band (still my longer-term objective) ?

Probably, yes; GremNOS does have some direct support for a handful of
network cards built-in, but it's not at all clear that you would be able
to find one of those cards.  You would probably have better luck finding
packet driver software for whatever card you do manage to find.

(Hmmm.  I've just been struck by a feeling that I've got a spare 3Com
3C509 card kicking around somewhere; I'll see if I can find it.)

> >Nope.  That's a local DNS *client* querying a remote (Demon) server;
> >only hardcore techies want to run their own local DNS servers.
> 
>         - lol - there's three or four people i know, i'd not put it past.

Oh, I know people who do run their own local DNS servers; but I can't
imagine any of them using GremNOS to do so! 8-)

> >If there are bits missing from Lite that you want to use again, I can do a
> >build that re-enables them (at the expense of a larger memory footprint)...
> 
>         - many thanks: once i'm safely through this forced emigration,
>         i may ask you (unless it'd be simpler for you to do a pplg_03e,
>         as it were, using the work you've already done for pplt_03e);

Creating a new custom build is trivial; it takes a couple of minutes.
(All I need to do is copy CONFIG.H, and possibly SMTPSCFG.H and/or
NNTPCCFG.H, to a new directory and modify them accordingly, by changing
"#undef" to "#define" or vice versa for the featuers in question.)

Would you like to me re-enable the shell stuff in ppLite (call it
ppBantam 8-) and do you a quick build you can play with?

>         have been using clockwork frog's view tsr whilst online to e.g.
>         try and check a sender vs. my mailkill file, but it's not got a
>         search function and it stops everything else:

Is this just to read the mailkill file, or to edit it as well?  If the
former, I just though I'd check you're aware that GremNOS can display
files with "more filename" (still without a search capability, but it
sounds as though the view TSR isn't much better)...

But being able to shell out, with multitasking, would probably work
better.

> >>      - i've a suspicion that c+w may write to administrator@, rather
> >>      than to postmaster@, if ever they wish to contact us post-forced
> >>      mail emigration.
> >
> >Interesting thought; you may be right.  I've set up a separate user
> >account to receive mail addressed to administrator@ and postmaster@, to
> >keep them away from the catch-all that will attract all the stuff to
> >bogus recipient addresses.
> 
>         - *nods*; but iiuc c+w've not mentioned in all of this that
>         we should (rfc/use-for "should") *all* open a user a/c for
>         postmaster, nor made it an existing user a/c by default - so
>         the catch-all a/c default named administrator@ is the only
>         a/c they can be (almost) certain we'll all have.

Indeed.  Certainly there have been lots of people in the forums saying
"what do you mean by accepting mail to postmaster? I'm so confused" or
words to that effect; they may well abandon the use of postmaster.

> >If "unknown" is in your FTPUSERS file, you wouldn't have suffered Pete's
> >fate...
> 
>         - i just bunged an extra line for it in the alias file - and'd
>         not got around to removing it, when it didn't work for the c+w
>         forgery.

It might be worth leaving it in there, to ensure you still get your mail
if you accidentally misconfigure your "pop3 collect" command (or, once
the auto-recipient-detection code is finished, if it is unable to work
out who the recipient was).

I don't know about the rest of you, but a quick grep of the logs shows
that I've never received any mail to unknown@.  (Except from me, when I
was testing the POP3 collection code!)

> >> >No worries: with POP3 collection and "pop3 bounce off", the mailkilled
> >> >messages will simply vanish in a puff of logic.
> >> 
> >>      - *g*  - thanks for the str.  have you read much/any sheckley?
> >
> >Not that I'm aware of; is this something I should correct?
> 
>         - oh yes!  (though not absolutely everything by him): one of
>         sf's all-time greatest humourists, [...]

Oooh, he sounds worth hunting down.  Thanks!

> >> >I don't know what the current system does if you reject the mail it's
> >> >trying to send to you: I assume it sends a DSN to the poor sap whose
> >> >address has been forged as the envelope-from, but perhaps not.  (If you
> >> >want to find out, I can send a message to an invalid username @imt and
> >> >tell you if I get a DSN back!)
> >> 
> >>       - feel welcome, should you wish to discover.
> >
> >Test message sent to weeble at i-m-t; I'll report back!
> 
>         - spotted in the queueue on the punt-3 cluster before i down-
>         loaded this batch of news from the firedrake; my immediate re-
>         action was "i'd better add a line to the alias file, first",
>         before i realised what it'd be and quashed the automatic urge.

It did, indeed, generate a DSN that was sent back to me.  (Actually,
there were two: I sent another message to myself and weeble at i-m-t
when I was testing how the current system handles mail that is Bcc'ed to
two different Demon customers in the same SMTP transaction[*].)

>         - again, thanks for the work you're putting in on all this;

A pleasure.

>         let me know if you'd like an eye kept open for any of the
>         sheckleys - or any other fantasy &/or sf, for that matter.

Thanks -- I'll have a quick rummage around here, but may well take you
up on the offer!

>         [0] - ghee!

ROFL!


[*] I was wondering whether mailing-list operators with multiple Demonic
subscribers send them individual messages or a single one with several
different recipients at different hostnames, and if you can use the
"Received:" header fields to find out.  The answer is that you can use
said fields: if there is only one recipient at your hostname, but the
first (i.e. last, reading top-to-bottom) Demon-created "Received:" field
doesn't contain "... for user@yourhost.dcu", then it was sent to
multiple Demon customers in a single SMTP transaction.

-- 
Simon Turner                                    DoD #0461
simon@twoplaces.co.uk
    Trust me -- I know what I'm doing!          -- Sledge Hammer