22 Oct
2021
22 Oct
'21
7:06 p.m.
Yes, NETPTH is part of the PANDA distribution and I had begun looking at
it about 18 months ago before I fell into a couple of deep, dark DECnet
holes, viz:
* The SETNODE rewrite is done but wants a bit of a write-up for the
new functionality.
* The extensive work on FAL and anonymous access is complete and
documentation is done.
* A number of patches have been done for DAP, but I have yet to
implement RENAME.
* DECnet SMTP remains to be debugged; there is a .PRIOU/.PRIIN
conflict between it and DEC DDT that is hampering this.
o I may change XDDT
o I have yet to see whether IDDT has the same problem
* A few monitor changes were necessary to support the above
* A patch was done to keep DECnet from filling up the error log (when
talking on HECnet)
* Research on NRT is now down to monitor changes
NETPTH appears to not be fully productized for Tops-20.? Error handling
frequently involves complaining and exiting, whereas I saw that recovery
could be done in certain cases.? It is using SCAN, which, in addition to
occupying program virtual address space, also pulls in PA1050, which
occupies another 28 pages. SCAN makes NETPTH appear kind of clunky for a
Tops-20 user.
Not the end of the world by any means, but I keep making mistakes in it
that wouldn't be happening were it to be fully native Tops-20 and
leverage the SETNODE enhancements which give you node recognition (I.E.,
"?" pick lists, abbreviation and completion).
Still, NETPTH is completely serviceable and to what degree
productization should have happened really is a matter of debate.
------------------------------------------------------------------------
On 10/21/21 10:39 AM, Peter Lothberg wrote:
Tops10/Tops20 has something caled "NETPTH" but it don't work like trace
route that depends on packet TTL expire and ICMP replies, it actually talks management to
each node and splices the path together.
[Routing path from MRC (59.41) to LEGATO (2.1)]
From Via Back Thru Cost Hops
MRC (59.41) =>NI-0-0 / TAP-0 <= OPS1VA (59.59) -1 -1
OPS1VA (59.59) =>GRE-3 / GRE-4 <= IAD2 (59.1020) -1 -1
IAD2 (59.1020) =>DMC-0 / MUL-59-1020 <= A2RTR (2.1023) -1 -1
A2RTR (2.1023) =>TAP-0 / QNA-1 <= LEGATO (2.1) -1 -1
> ------------------------------------------------------------------------
> *From*: "David Moylan"<djm at wiz.net.au>
> *To*: "hecnet"<hecnet at Update.UU.SE>
> *Sent*: Thursday, October 21, 2021 1:20:12 AM
> *Subject*: RE: [HECnet] Is MIM running?
>
> It would be awesome to implement a "traceroute" equivalent on something :-)
I assume we'd have to allow node inspection from remote, but I'm sure people
would be open for that.
>
> I used to run a Cisco router for my area, but replaced it with pyDECnet as I wanted
to mix both GRE and point-to-point tunnels. pyDECnet ticked all the boxes and was simple
to implement.
>
> The one thing that I miss, however, is the simplicity of "ping decnet" in
Cisco IOS. Right from the router I could see if something was up or down. Now I have to
jump onto one of my boxes and do an ncp loop node to confirm the same.
>
> Would be nice to have the same implemented in pyDECnet, perhaps as a command line
option, or via the web gui.
>
> If nobody else has any equipment handy, I could setup a router again at home and do
some captures.
>
> Cheers, Wiz!!