From wang!elf.wang.com!ucsd.edu!tcp-digest-relay Wed Feb 13 14:52:29 1991 remote from tosspot Received: by tosspot (1.63/waf) via UUCP; Thu, 14 Feb 91 07:26:22 EST for lee Received: from somewhere by elf.wang.com id aa04307; Wed, 13 Feb 91 14:52:29 GMT Received: from ucsd.edu by uunet.uu.net (5.61/1.14) with SMTP id AA06544; Wed, 13 Feb 91 09:11:11 -0500 Received: by ucsd.edu; id AA13756 sendmail 5.64/UCSD-2.1-sun Wed, 13 Feb 91 04:30:26 -0800 for jbloom Received: by ucsd.edu; id AA13751 sendmail 5.64/UCSD-2.1-sun Wed, 13 Feb 91 04:30:21 -0800 for /usr/lib/sendmail -oc -odb -oQ/var/spool/lqueue -oi -ftcp-digest-relay tcp-digest-list Message-Id: <9102131230.AA13751@ucsd.edu> Date: Wed, 13 Feb 91 04:30:19 PST From: Advanced Amateur Radio Networking Group Reply-To: TCP-Group@ucsd.edu Subject: TCP-Group Digest V91 #40 To: tcp-group-digest@ucsd.edu TCP-Group Digest Wed, 13 Feb 91 Volume 91 : Issue 40 Today's Topics: [Vance Morrison : Re: pcroute, appletalk, (2 msgs) internet <-> packet ideas .. (2 msgs) NNTP and SMTP SMTP corruption Send Replies or notes for publication to: . Subscription requests to . Problems you can't solve otherwise to brian@ucsd.edu. Archives of past issues of the TCP-Group Digest are available (by FTP only) from UCSD.Edu in directory "mailarchives". We trust that readers are intelligent enough to realize that all text herein consists of personal comments and does not represent the official policies or positions of any party. Your mileage may vary. So there. ---------------------------------------------------------------------- Date: Tue, 12 Feb 91 3:07:36 EST From: Tony Querubin Subject: [Vance Morrison : Re: pcroute, appletalk, To: tcp-group@ucsd.edu Received: by thucydides.cs.uiuc.edu (5.61+/IDA-1.2.8) id AA02352; Thu, 7 Feb 91 10:00:28 -0600 Date: Thu, 7 Feb 91 10:00:28 -0600 From: Vance Morrison Message-Id: <9102071600.AA02352@thucydides.cs.uiuc.edu> To: tony@uhheph.phys.hawaii.edu Subject: Re: pcroute, appletalk, SLIP Hello, >OK, I've taken your advice and subnetted the entire 128.171.11.0 >network. The net is split up into 8 subnets with a subnet mask of >255.255.255.224. I am afraid you can't do this. You see the 128.171 is your class B network number, thus within that subnet, you can ONLY subnet once. That is the subnet mask for ALL hosts begining with 128.171 MUST all agree on what the subnet mask is. I suspect that you 128.171 network has already been subnetted with the mask 255.255.255.0. If that is the case, you may NOT chose a different mask for this part of the net. You MUST choose a differernt, unused subnetwork number (like 128.171.12, or 128.171.15 etc). You will need two of these, one for the localtalk net and one for the slip line. >ping the appletalk port of the PC (128.171.11.94) from the SUN, the >ICMP echo-replys come back with a from address of 128.171.11.33. Many people think this is a bug, it is not, PCroute always replys with the IP address that is 'closest' to the host. >I don't have any static routes defined in PCROUTE. I don't have >logging enabled. You will need static routing information. Also remember that the KIP box does NOT do RIP, (at least on the localtalk side) so it needs to know the fact that it can get to the SLIP network (and any other nets on the 'far' side of the localtalk), by sending to the PCrouter. (I hope KIP boxes are smart enough for this). The more I think about this problem, the more I realize how subtle this is. There are MANY things that can go wrong, and unless you know enough to bypass most of them, it is going to take a LOT of trial and error, You may even find that the KIP box simply doesn't understand 'far' networks through the localtalk, and thus it would be impossible, unless another PCrouter was setup in parallel with the KIP box. In short, we are trying to make the KIP box do something that it may not be designed to do (route to far nets through localtalk). It may work, it may not. If you need a 'sure thing' this is NOT it. If it can be done in some other way, that might be the better option. Vance ------------------------------ Date: 12 Feb 91 09:52:23 PST (Tue) From: romkey@asylum.sf.ca.us (John Romkey) Subject: [Vance Morrison : Re: pcroute, appletalk, To: tony@uhheph.phys.hawaii.edu >ping the appletalk port of the PC (128.171.11.94) from the SUN, the >ICMP echo-replys come back with a from address of 128.171.11.33. Many people think this is a bug, it is not, PCroute always replys with the IP address that is 'closest' to the host. It is a bug. From RFC 1122, Host Requirements: The IP source address in an ICMP Echo Reply MUST be the same as the specific-destination address (defined in Section 3.2.1.3) of the corresponding ICMP Echo Request message. - john romkey Epilogue Technology USENET/UUCP/Internet: romkey@asylum.sf.ca.us FAX: 415 594-1141 ------------------------------ Date: Tue, 12 Feb 91 14:28:58 CST From: andyw@aspen.cray.com (Andy Warner) Subject: internet <-> packet ideas .. To: tcp-group@ucsd.edu Since I just posted this to rec.ham-radio.packet, I thought it'd be wise to cross-post to where most of the NOS work-in-progress occurs. <================================= cut here ================================> In article <312@nachos.SSESCO.com>, elmquist@nachos.SSESCO.com (Chris Elmquist) writes: |> In article <1462@tsdiag.ccur.com> davet@tsdiag.ccur.com (Dave Tiller N2KAU) writes: |> >Regardless of the other brouhaha over the legalities of internet<->packet |> >gatewaying, here's something that I'll be writing and trying out in the |> >near future: |> > |> > |> > |> >TNC-----Internet Node-----------//----------Internet Node-----TNC |> >145.01 running running 144.97 |> > NJ socket socket 'elsewhere' |> > prog prog |> > |> [some instructions on use deleted] |> |> >Any comments, good or bad? I'm looking for someone (out of local range) |> >with an internet node handy to try it with me. Any takers? (Phil Karn, are |> >you listening?) |> |> [gofar hole stuff deleted] |> |> I'd be willing to play with this extensively. We need outside |> world connectivity for our packet network here in Minnesota-- OK - I've been thinking about an idea for a while now ... I was trying to write the code before I had to go public, but life's been too short recently. Basically, it's implementation would go something like this (in NOS) :- If we get an incoming AX25 packet which we're going to digipeat, check the next callsign against a hashed table mapping callsigns onto IP addresses. If a match was found - encapsulate the whole AX25 packet in UDP and send it off to that IP address. On the other end there is a server which receives these encapsualted packets, checks the sender's IP address against it's list of IP <-> callsign mappings (as a crude check). If all is well, it figures out which interface to send the AX25 packet out of. The routes would be bi-directional and would require positive intervention by the sysops at BOTH ends before a link could be used. There are some benifits that fall out .. 1) The minimal case means that you can cross-band digi through your own machine - by specifying the loopback interface. 2) You could have (for instance) two machines in your shack with radio ports on both and be able to provide xband between ALL the ports. This might be especially useful if one was running high speed, and didn't need to be bothered with interrupts from a 1200/2400 baud metro channel... 3) Providing the link at the digipeat level means that NET/Wrong and AX25 users can exploit this easily. 4) You could (appropriate use of networks permitting) span huge distances very quickly [if you doubt this try pinging some machines 1000 miles away!] I've not thought through ALL the ramifications yet, and it would probably require the use of SSID's which has been shunned so far. Having already written cross-band digi code for NOS, I know this should be straight-forward to implement - I just haven't got round to it yet :-) :-) Any thoughts. -- andyw. (W0/G1XRL) andyw@aspen.cray.com Andy Warner, Cray Research, Inc. (612) 683-5835 <================================= cut here ================================> ------------------------------ Date: Tue, 12 Feb 91 19:04:46 -0500 (EST) From: Anders.Klemets@CS.CMU.EDU Subject: internet <-> packet ideas .. To: Andy Warner , tcp-group@ucsd.EDU > I've not thought through ALL the ramifications yet, and it would probably require > the use of SSID's which has been shunned so far. No it is not absolutely necessary to make NOS use different SSID's on different ports. It is enough to check the next callsign in the digipeater field, just as you suggest. Here is an example of how it would work. Suppose I have two ports on my station. Both ports use SM0RGV as the link level address. W3XX has a similar setup at his station. If I receive a frame on port #1 that I am supposed to digipeat, I will compare the next callsign in the digipeater field with entries in a special "wormhole table." Suppose the callsign is W3XX-2 and it matches an entry in the table. I will then replace my callsign (SM0RGV) with something that uniquely identifies my station and the port number, for instance SM0RGV-1. Then the AX.25 frame is sent off to W3XX with UDP. When W3XX receives the frame he will see "W3XX-2" and realize that this means that the frame should be sent on his port #2. However, he will replace the callsign W3XX-2 with the "mycall" callsign (W3XX) before transmitting the frame. The return path will work in a similar fashion. Of course, W3XX needs a reciprocal entry for SM0RGV-1 in its "wormhole table" to properly route an incoming AX.25 frame. Anders ------------------------------ Date: Tue, 12 Feb 91 15:52:24 EST From: "Mark Bramwell" Subject: NNTP and SMTP To: "TCP/IP User Group" I know how to pull NNTP messages from our campus server. The files are deposited in .TXT files. How can I have those messages requeued into SMTP. I have a few users on other messages who would be interested in the messages. Automatic forwarding would be nice. I would guess that you need something in the /ALIAS file, but how do you specify something that has '/' or '.' in it's name? Example: rec.video.satellite =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Mark Bramwell, VE3PZR Located in sunny London, Ontario Internet: mark@hamster.business.uwo.ca IP Address: 129.100.22.100 Packet: VE3PZR @ VE3GYQ UWO Phone: (519) 661-3714 ------------------------------ Date: Tue, 12 Feb 91 7:34:14 EST From: rutgers!tsdiag.ocpt.ccur.com!nn2z!nn2z Subject: SMTP corruption To: tcp-group I took a look at the mail thats been forwarded here by NOS 910131 and I also see that most message text is corrupted. Yes Tom something is ivery wrong. The system also crashes daily. It seems to just hang. Phil did anything change in SMTP lately ? I am back to version 900828 which is about all I can run. I will have to try the new g1emm version and see how that works. Dave ------------------------------ End of TCP-Group Digest ******************************