🚀 go-pugleaf

From: andyw@aspen.cray.com (Andy Warner)
Date: Mon, 15 Jul 91 10:27:23 CDT
Subject: sendmail.cf question..

Is there anyone else out there doing battle with sendmail.cf files
on a unix platform that is in ampr.org and a n other domain ?

I know about comp.sys.sendmail, but rather than trying to explain
the problem from scratch, I thought I'd ask here first to see what's
been done so far..

I've experimented with several things so far, including, the heretic
idea of producing those non-RFC lines:

R: biggest MSYS system in the upper mid-west blah, blah, blah

Just to stop people bitching about the voluminous headers (we've
probably all been THERE at least once). Anyway, I've tried various
things in the sendmail.cf file, some good, some bad; is there anyone
else out there who's invested any time in this area too ? Want to
share what you've learned ?
-- 
andyw.	(W0/G1XRL)

andyw@aspen.cray.com	Andy Warner, Cray Research, Inc.	(612) 683-5835


----
From: Dave Goodwin <GOODWIN%SMCVAX.BITNET@mitvma.mit.edu>
Date: Mon, 15 Jul 91 13:22 EDT
Subject: PA0GRI Error

I received the following error message on two separate occasions last week.
Anyone know what it's trying to tell me :^)

PANIC: Timer loop at (some hex address)

Thanks in advance...

==========================================================================
Dave Goodwin                                  |
Assistant Director, Academic Computing        | To err is human.
VAX System Manager                            | To really foul things up,
----------------------------------------------| You need a computer.
BitNet/CREN : Goodwin@Smcvax                  |
Ma Bell     : (802)655-2000 X2220             |
Pony Express: Saint Michael's College         |           -Who Knows?
              Department of Computer Services |
              Winooski Park                   |
              Colchester, VT 05439            |
Ham Radio   : N1HRO                           |
==========================================================================


----
From: Mark H. Levine <yba@space.mit.edu>
Date: Mon, 15 Jul 91 17:24:23 EDT
Subject: Re: FYI.. htonl and patent #4,956,809

I echo Phil's comment even more emphatically on the byte-order  
patent.

I was at BBN (Cambridge. MA) in 1980 and in the days before TCP/IP  
when the Arpanet used NCP, there was already a well-established  
"network byte order" defined for transmission.  The FTP program you  
are all familiar with on tcp-group predates TCP/IP as well  
(surprise), and it worked for PDP-11s, PDP-10s, Honeywell 600s,  
PDP-20s, and later C-70s, among others, where the _size_ of bytes as  
well as their order differed.

The nice thing is that all the BBN Arpanet work was ARPA funded, and  
hence available under the Freedom of Information Act (although most  
is quite freely available from even a cursory literature search).

I suspect that the "prior art" is actually much older, and dates to  
the second day a computer with a different order was made available  
on a network....

The fellow trying to enforce this patent should have a fine for  
nuisance lawsuit slapped on him or her post haste!  It frightens me  
this even needed to be announced....


----
From: Ben Thornton <ben@val.uucp>
Date: Mon, 15 Jul 91 14:27:50 -0500
Subject: Before you do that release...


  I'm not sure if anyone else has caught it (am I the only one running
the PPP code?), but the function userlookup() was deleted from files.c
some time back for some reason.  This function is referenced when you 
enamble the PPP conditionals.  I had to dig out a previous version of
NOS to correct the problem.  It is called from a module in PPPPAP.C.

Ben Thornton
WD5HLS


----
From: "James C Walker(Cap'n)" <70007.4651@CompuServe.COM>
Date: 15 Jul 91 21:36:39 EDT
Subject: POP docs ?

What is the best source for documents on POP ? If it's on internet, is there
and alternate host, as in bbs (Compuserve users no longer have good ftp
access).
 
Specifically, if a pop session is completed, the mail is written to
mbox.pop, what happens to that mail ? Is it left in the smtp queue at the
remote station ? Does mbox.pop get rewritten back into the *.txt file in
\spool\mail ? Do any mailers (like BM) recognize this ?
 
Cap



----
From: wkt@csadfa.cs.adfa.oz.au (Warren Toomey)
Date: Tue, 16 Jul 91 12:58:52 EST
Subject: A Paper on `Roaming' Internet Hosts

I just obtained a paper about the type of mobile internetworking that
Phil Howard & I have been discussing in the tcp mailing list. Here are
the title and abstract:

		IP-based Protocols for Mobile Networking
	    John Ionnidis, Dan Duchamp, Gerald Q. Maguire Jr.
		  ji, djd, maquire @cs.columbia.edu

Abstract

We consider the problem of providing network access to hosts whose physical
location changes with time. Such hosts cannot depend on traditional forms of
network connectivity and routing because their location, and hence the route
to reach them, cannot be deduced from their network address. In this paper
we explore the concept of providing continuous access to mobile computers,
and present a set of IP-based protocols that achieve that goal. They are
primarily targeted at supporting a campus environment with mobile computers,
but also extend gracefully to accomodate hosts moving between different
networks. The key feature is the dependence on ancillary machines, the
Mobile Support Stations (MSSs), to track the location of the Mobile Hosts.
Using a combination of caching, forwarding pointers, and timeouts, a minimal
amount of state is kept in each MSS. The state information scales well, reacts
quickly to changing topologies, and does not place an overwhelming burden on
the network.

Any typos are mine. I have only a PostScript copy of this file, but I will
leave it (118K) in the `packet' dir on minnie.cs.adfa.oz.au for anon ftp. The
file will be called sigcom91.ps. Remember, 0700Z to 2300Z for minnie.

Cheers all,

	Warren vk1xwt.


----
From: "James C Walker(Cap'n)" <70007.4651@CompuServe.COM>
Date: 15 Jul 91 21:36:39 EDT
Subject: POP docs ?

What is the best source for documents on POP ? If it's on internet, is there
and alternate host, as in bbs (Compuserve users no longer have good ftp
access).
 
Specifically, if a pop session is completed, the mail is written to
mbox.pop, what happens to that mail ? Is it left in the smtp queue at the
remote station ? Does mbox.pop get rewritten back into the *.txt file in
\spool\mail ? Do any mailers (like BM) recognize this ?
 
Cap



----
From: wd0efl@wd0efl.UCSD.EDU (Brian K. Teravskis)
Date:  Mon, 15 Jul 1991 23:44:07 -0500
Subject: NOS compiling problems and TC++ V1.0


I am resending this message because I am not sure that the first one
that I sent out made it due to some gateway problems on this end.


I'm rather new at this, and I am wondering if someone would help me
with some compiling problems I am having with TC++ V1.0 and the source
code src_0605.

I get a lot of errors like the following:

Error bmutil 720: type mismatch in parameter 'm' in function mbox

I have also had problems in the past with a properly compiled version
of NOS that turns out to be stone deaf to the serial line...Everything
else works fine, but when it comes time to read/write to the serial
buffer nothing happens..

Any help would be appreciated.

Brian Teravskis  -- WD0EFL

AX25 :  WD0EFL @ N0DAI.MN.USA.NA
Internet : wd0efl@g1xrl.vware.mn.org


I would like to add that when I got the deaf NOS is when I told TC++
to ignore type mismatch errors as errors.

Again thanks for any help.

Brian   WD0EFL


----
From: crompton@NADC.NADC.NAVY.MIL (D. Crompton)
Date: Tue, 16 Jul 91 01:03:28 EDT
Subject: Re: New(er er) NOS 910618

Gerard,

 The problem is in the new MKNAME module. I put back the module I had
made for 1.7d - exactly as Mike prescribed and it works fine. 

Doug


----
From: crompton@NADC.NADC.NAVY.MIL (D. Crompton)
Date: Tue, 16 Jul 91 00:15:40 EDT
Subject: Re: New(er er) NOS 910618

Gerard,

 Your version 1.7e compiles fine, but it locks my system up. I am
not sure but it appears to be associated with sending/receiving mail.
It also happens at initial startup. I get lockup at MORE of stat
screen or at first net prompt. Sometimes it is complete power off
lockup and sometimes the keyboard buffer full (beep) variety.

This is the first time I have had lockup problems with nos on my system
in a long long time. I went back to last version and had no problems.
So I can confirm that it is not my hardware.

Compile was with same config.h I used previously. BCC 2.0 compiler.
Temp files appear to work OK when I can keep it working long enough.

I can confirm that Mike's suggested mods worked fine when applied to
version 1.7d. You made some small changes but other changes also so 
it might be good to look over what you did.

To those that are having problems with earlier PA0GRI code I do not
have an answer. Up until now - since Kelvin's early code - I have
had VERY few problems. I have not seen an error message for quite
awhile. I exercise the system quite a lot. Others in our group do
also with no complaints. Everyone runs the system with a little
different configuration though. Most of the problems can be traced
to not enough memory. I like to see 80-100K core left - 50K minimum.
If you are running much below that you are flirting with trouble.

Doug


----
From: Antonio Querubin <tony@mpg.phys.hawaii.edu>
Date: Tue, 16 Jul 91 0:05:22 HST
Subject: NNTP message expiration

Is there any facility in NOS for automatically expiring messages received by
the NNTP client after so many days?


Antonio Querubin  
tony@mpg.phys.hawaii.edu / ah6bw@uhm.ampr.org / querubin@uhunix.bitnet


----