I reckon it could be done with a single interface.
Will take a look at the specs, a raspberry pi DECnet router would be great!
Regards
Rob
-----Original Message-----
From: owner-hecnet at Update.UU.SE [mailto:owner-
hecnet at Update.UU.SE] On Behalf Of Mark Benson
Sent: 06 June 2012 22:41
To: hecnet at Update.UU.SE
Subject: Re: [HECnet] Multinet Tunnel Connections to SG1::
On 6 Jun 2012, at 22:37, <Paul_Koning at Dell.com> wrote:
On Jun 6, 2012, at 5:34 PM, Johnny Billquist wrote:
On 2012-06-06 22:40, Rob Jarratt wrote:
...
It would be a fun exercise to turn the bridge into an actual DECnet
router.
One day, when I have a lot of time I might even think about trying
that.
Feel free. The specifications are out there, so it's definitely not an
impossible task. But I guess it will take some work. It would be very nice
if the
bridge did turn into a WAN router...
Johnny
DECnet Phase IV is pretty straightforward. The specs are all out there.
And
much of it is implemented in Linux, so you can find C sources ready to
go...
Even if sample code only does L1 routing, you're nearly there because area
routing is essentially the same thing done a second time on a second set
of
tables.
There may be mileage in building dedicated DECnet bridges/routers using
RaspberryPi boards in a suitable case running a very cut-down Linux with
an
advanced version of the bridge? I am no programming wizard but I am good
at compiling and testing stuff.
Would it require 2 Ethernet interfaces to work or would (like the bridge)
just
one work okay?
--
Mark Benson
http://DECtec.info
Twitter: @DECtecInfo
HECnet: STAR69::MARK
Online Resource & Mailing List for DEC Enthusiasts.
On 2012-06-06 23:40, Mark Benson wrote:
On 6 Jun 2012, at 22:37,<Paul_Koning at Dell.com> wrote:
On Jun 6, 2012, at 5:34 PM, Johnny Billquist wrote:
On 2012-06-06 22:40, Rob Jarratt wrote:
...
It would be a fun exercise to turn the bridge into an actual DECnet router.
One day, when I have a lot of time I might even think about trying that.
Feel free. The specifications are out there, so it's definitely not an impossible task. But I guess it will take some work. It would be very nice if the bridge did turn into a WAN router...
Johnny
DECnet Phase IV is pretty straightforward. The specs are all out there. And much of it is implemented in Linux, so you can find C sources ready to go... Even if sample code only does L1 routing, you're nearly there because area routing is essentially the same thing done a second time on a second set of tables.
There may be mileage in building dedicated DECnet bridges/routers using RaspberryPi boards in a suitable case running a very cut-down Linux with an advanced version of the bridge? I am no programming wizard but I am good at compiling and testing stuff.
Hmm. Sure. But if you go the Linux routing route, then you instead get into the problem that it don't have a point-to-point connection to another Linux box on the Internet...
Would it require 2 Ethernet interfaces to work or would (like the bridge) just one work okay?
It would (today) require two ethernet interfaces, in which it routes between. However, talking about connections across the world, you are not even sitting on any shared ethernet networks, so it will not give you what you need for HECnet until you add some point-to-point connection, using IP (and whatever) to some other destination.
Maybe something already exists to accomplish this. Not sure if the GRE implementation in Linux can be used to solve it...
Johnny
On 6 Jun 2012, at 22:37, <Paul_Koning at Dell.com> wrote:
On Jun 6, 2012, at 5:34 PM, Johnny Billquist wrote:
On 2012-06-06 22:40, Rob Jarratt wrote:
...
It would be a fun exercise to turn the bridge into an actual DECnet router.
One day, when I have a lot of time I might even think about trying that.
Feel free. The specifications are out there, so it's definitely not an impossible task. But I guess it will take some work. It would be very nice if the bridge did turn into a WAN router...
Johnny
DECnet Phase IV is pretty straightforward. The specs are all out there. And much of it is implemented in Linux, so you can find C sources ready to go... Even if sample code only does L1 routing, you're nearly there because area routing is essentially the same thing done a second time on a second set of tables.
There may be mileage in building dedicated DECnet bridges/routers using RaspberryPi boards in a suitable case running a very cut-down Linux with an advanced version of the bridge? I am no programming wizard but I am good at compiling and testing stuff.
Would it require 2 Ethernet interfaces to work or would (like the bridge) just one work okay?
--
Mark Benson
http://DECtec.info
Twitter: @DECtecInfo
HECnet: STAR69::MARK
Online Resource & Mailing List for DEC Enthusiasts.
On Jun 6, 2012, at 5:34 PM, Johnny Billquist wrote:
On 2012-06-06 22:40, Rob Jarratt wrote:
...
It would be a fun exercise to turn the bridge into an actual DECnet router.
One day, when I have a lot of time I might even think about trying that.
Feel free. The specifications are out there, so it's definitely not an impossible task. But I guess it will take some work. It would be very nice if the bridge did turn into a WAN router...
Johnny
DECnet Phase IV is pretty straightforward. The specs are all out there. And much of it is implemented in Linux, so you can find C sources ready to go... Even if sample code only does L1 routing, you're nearly there because area routing is essentially the same thing done a second time on a second set of tables.
paul
On 2012-06-06 22:40, Rob Jarratt wrote:
No, you have not misunderstood. You are just not getting the full picture.
Yes, the Cisco box will transport the DECnet frames inside something. At
the
bottom there will be IP, but most likely the DECnet frames are actually
sitting
inside UDP...
So what is the difference from the bridge program, you ask.
Well, simple, the bridge program tries to be clever about which frames to
send where, but it cannot be clever enough at times, and also there are
times when it is forced to send traffic that we might want to limit.
The reason is that the bridge is not really a part of the DECnet traffic,
but is
just relaying packets back and forth, making the two ends of the bridge
appear as if they were on the same ethernet segment.
The Cisco box, on the other hand, is a DECnet routing node. It sits on the
ethernet, and participate fully in the DECnet traffic.
Thanks. Now I get it. I didn't know that anyone other than DEC ever made
something that could route actual DECnet and therefore did not realise that
"actual DECnet router" meant "actual DECnet router" :-) Until now I thought
this was something only available in a VMS router node.
VMS not required... MIM is an area router running RSX. :-)
But yeah, "DECnet router" really meant just that.
It would be a fun exercise to turn the bridge into an actual DECnet router.
One day, when I have a lot of time I might even think about trying that.
Feel free. The specifications are out there, so it's definitely not an impossible task. But I guess it will take some work. It would be very nice if the bridge did turn into a WAN router...
Johnny
On 6/6/2012 5:26 AM, Peter Coghlan wrote:
No, you have not misunderstood. You are just not getting the full
picture. Yes, the Cisco box will transport the DECnet frames inside
something. At the bottom there will be IP, but most likely the DECnet
frames are actually sitting inside UDP...
My understanding of it is that GRE is another IP protocol on the same level
as TCP and UDP rather than using UDP.
You are correct. GRE is Layer 4 just like TCP and UDP.
-brian
On 6/4/2012 8:51 PM, Dave McGuire wrote:
On 06/04/2012 12:13 PM, Brian Hechinger wrote:
SIMH 3.9 somehow kernel paniced my solaris server!
Wow.
-brian
ps: it's older solaris, but stil!
Holy cow! I have NEVER seen that happen. (which is why I run it!)
How the heck did you do that?
Built and ran SIMH 3.9
3.8-1 works just fine (it's what i'm running now) and 3.9 runs fine in my sol11 VM.
-brian
On 6/6/2012 9:04 AM, Peter Lothberg wrote:
Could this work in a domestic environment with an ISP that
gives out dynamic IP addresses?
My head hurts.. Ask for a /56 of IPv6 space. -:)
If there are many users that need this, I will look in to it, I thnk
there was magic that could be applied.
On the side with the dhcp address it's simple. You just make the source the actual interface and not an IP.
interface Tunnel0
description Tunnel for DECnet to Dave M.
no ip address
decnet cost 10
tunnel source FastEthernet0/1
tunnel destination 50.73.179.1
tunnel path-mtu-discovery
end
interface FastEthernet0/1
description Internet Connection (Cable Modem)
ip address dhcp
ip nat outside
ip virtual-reassembly
duplex auto
speed auto
end
Now, in my case it's actually a static IP but RCN require me to actually fetch my IP via DHCP or it won't work.
As to how to handle this on the other end, I have no clue. :)
-brian
On 6/6/2012 8:48 AM, Peter Lothberg wrote:
On 2012-06-05 03:50, Peter Lothberg wrote:
For people who don't need LAT and have a Cisco box, this is way better
than the bridge program.
(If you really want LAT with other parts of HECnet, then there is no
alternative to the bridge...)
You can bridge LAT with the cisco box... -:) ((It might even HELP you
with LAT over WAN....)) (And it can act as a LAT termianl server)
Nice. I did not know that. That's even better. Will it just bridge the
protocols you ask for, or will it do them all? Hmm, I assume it's using
GRE for this?
Johnny
I -think- there where special stuff to help with LAT in WAN scenarios,
at some point there might even have been a telnet/lat gw.. I have to
go and refresh my memory..
SuperLAT! It no longer says that in the version banner though.
bart#lat ?
WORD Name of a remote system
bart(config-subif)#lat ?
enabled Enable LAT protocol translation
host Statically define LAT services
-brian
No, you have not misunderstood. You are just not getting the full picture.
Yes, the Cisco box will transport the DECnet frames inside something. At
the
bottom there will be IP, but most likely the DECnet frames are actually
sitting
inside UDP...
So what is the difference from the bridge program, you ask.
Well, simple, the bridge program tries to be clever about which frames to
send where, but it cannot be clever enough at times, and also there are
times when it is forced to send traffic that we might want to limit.
The reason is that the bridge is not really a part of the DECnet traffic,
but is
just relaying packets back and forth, making the two ends of the bridge
appear as if they were on the same ethernet segment.
The Cisco box, on the other hand, is a DECnet routing node. It sits on the
ethernet, and participate fully in the DECnet traffic.
Thanks. Now I get it. I didn't know that anyone other than DEC ever made
something that could route actual DECnet and therefore did not realise that
"actual DECnet router" meant "actual DECnet router" :-) Until now I thought
this was something only available in a VMS router node.
It would be a fun exercise to turn the bridge into an actual DECnet router.
One day, when I have a lot of time I might even think about trying that.
Regards
Rob
