L2TPv3 Cisco and LAT Protocol ?

Would be rather surprising if it did not. I would though want to do an evaluation before spending a lot of money. There can be very, very few people using LAT now so you could always run into some bug or other.

If the routers terminating the tunnels were also running DEC protocols I would be especially wary.

Out of interest what are you using LAT for? It was already substantially obsolete when I first heard of it in about 1996:)

Will this look like it is bridged? LAT isn't routable. Unless L2TPv3 config makes the routers involved transparent to the LANs involved, I wouldn't think you be too successful.

Why?

Local Area Transport. It wasn't devised for WAN communications; it was devised to be a lightweight, efficient, local area trans- port. I use it for X11 on the local lan; also, for terminal and serial devices. I maintained a device driver for some telecom- munications switching gear that was using LAT for device control too.

Just to be clear I don't consider myself very strong on this area but there does not seem to be anyone else so I thought I would try.

My understanding is that L2TPv3 produces not even a bridge but a (virtual) wire.

Consider the case of CDP. If a port running L2TP was to receive a CDP packet what should it do with it? Should it deliver the frame to the CPU for CDP processing of transport it over the link?

There are two answers to this.

1. If you are a custome and you expect a service provider to deliver a wire then you want the CDP to pass over the link.
2. If you are say using L2TP within your own network you might just want local resolution of CDP.

I am not sure but I suspect that behaviour 2 is what happens.

I was concerned that a similar dichotomy might be presented to the DEC stack developers or that some oversight might be possible. In view of the small number of likely users of LAT on L2TP it seemed not unlikely that any bugs would remain undiscovered for some time.

Ah yes! I have just remembered. I think you can manage (telnet to - sorry) a router with MOP. I completely forget how that relates to LAT. Hence my caution.

All the DEC stuff may have gone now of course. Don't know - and don't care really:))

Well, I predict in increasing level of boredom in the future:) Although of course in some sense VMS (or its wayward daughter?) is probably the most widely used operating system on the planet today.

What I am saying is that I think that this will work but that you should try it out.

If the MAC (Ethernet addresses) can be visible, there's hope.

MOP is in the DECnet stack. Under Phase IV, the routing was accomplished by changing the MAC address AA-00-04-00-XX-YY where XX:YY were computed as a function of the DECnet area and node number. AB-... was a broadcast. MOP function code is 06-01/02

I'm plenty busy.

Wayward daughter? RSX?

I have heard of some IBM type people being paid obscene amounts of money to support obsolete systems (TCAM?). There seems a decent chance of an action replay with VMS.

Good luck anyway.

I was thinking of NT and its descendants.

I understand that Microsoft hired the VMS architect from DEC and that he was responsible for NT. I forget his name. Cutler or something like that. Dear god, googles and I was right!

You're funny.

To associate the crap that is WEENDOZE with the elegance that is VMS is insulting.

David Cutler and it was only to give M$ crap some semblance of being a real OS. We all know better now, don't we.