Mike Shields wrote:
Mark -
I have a TOPS-20 setup on KLH10, running on a Solaris host wtih a dedicated
ethernet card for the KLH10 instance.
I'd try to help you but I honestly don't know how to configure DECnet on
TOPS-20. My goal is to get DECnet working,
At a minimum, you simply need to add the following to your system config
file (depending on your TOPS-20 version, SYSTEM:7-1-CONFIG.CMD or
similar)...
NODE MYNODE 1.123
ETHERNET 0 DECNET
DECNET ROUTER-ENDNODE
Replace "MYNODE" with your 6-letter nodename, and "1.123" with your node
number. The "0" (zero) for ETHERNET may need to be adjusted if you have
more than one network interface, but "0" is correct for the first NI0
interface. I'm fairly certain the third directive indicates that you're
an end node and are not effecting any DECnet routing yourself.
and one of the reasons I put a dedicated ethernet card in was that I read
DECnet changes the MAC address to some
specific value corresponding with the node number. For some reason I didn't
think that'd work very well with bridging/tap.
(Though it may work just fine, it was just a gut feel)
It works just fine actually. Problem at least on FreeBSD is that the
bridge ceases to function if any of the bridged interface's MAC
addresses changes without simply cycling the bridge down and up again.
Because of this I disabled the MAC address change in KLH10 and do it
manually.
Well, anyways, let me know if you want to collaborate on getting both our
systems working.
Mike
I've actually made a new discovery... looking at "netstat -g", I'm
seeing that none of the multicast addresses exist that DECnet uses and
requires! This would explain why I don't see any DECnet traffic at all.
I compiled KLH10's enaddr utility in an attempt to add the multicast
addresses manually, and I get failures for all attempts...
"SIOCADDMULTI failed - Can't assign requested address". My server's
kernel wasn't built with multicast routing enabled, which is not a
default. I'm wondering if this is the cause of the error. I'm going to
do some further testing.
Regards
Mark
I'm probably going to be running an exhibit this weekend where I connect a VAX to hecnet. I will need to set up a node temporarily with the bridge. Can I peer with someone? It would be great if you can assign me a node number in your area temporarily. Thanks.
Peace... Sridhar
Johnny Billquist wrote:
Bob Armstrong wrote:
I made this change on LEGATO too.
Not that it matters, but LEGATO thinks the adjacent node on the
circuit is
3.34, ROOSTA. From your numbering system, this must be some kind of
VAX or
Alpha.
.ncp tell roosta sho exec
Node summary as of 12-SEP-08 10:57:27
Executor node = 3.34 (ROOSTA)
State = On, Identification = DECnet for OpenVMS VAX V7.2
:-)
That's one of the reasons I wrote NML for Linux.
Incidentally, that now works on big-endian systems too. I found the bug :)
--
Chrissie
Mark Abene wrote:
Is anyone on here successfully using DECnet in TOPS-20 with the KLH10
emulator? I'm using a dedicated bridged-tap interface in FreeBSD, and
DECnet is certainly enabled in TOPS-20, but there is absolutely no
DECnet traffic on the wire. None can be seen with tcpdump. I find this
very odd. IP connectivity works fine, and I can see LAT traffic, but no
DECnet. Any suggestions?
Well, I can only say that we have one KLH10 system (PAMINA::) running here. Works fine. But it's running on a Linux box, and also have a dedicated physical interface.
No ideas why it don't work for you.
Johnny
Bob Armstrong wrote:
I made this change on LEGATO too.
Not that it matters, but LEGATO thinks the adjacent node on the circuit is
3.34, ROOSTA. From your numbering system, this must be some kind of VAX or
Alpha.
.ncp tell roosta sho exec
Node summary as of 12-SEP-08 10:57:27
Executor node = 3.34 (ROOSTA)
State = On, Identification = DECnet for OpenVMS VAX V7.2
:-)
Johnny
Bob
-----Original Message-----
From: owner-hecnet at Update.UU.SE [mailto:owner-hecnet at Update.UU.SE] On Behalf
Of Christine Caulfield
Sent: Thursday, September 11, 2008 5:31 AM
To: hecnet at Update.UU.SE
Subject: [HECnet] Node rename
I'm going to be a pest now, sorry.
My old node ZAPHOD died over a year ago. However I now have a new node
so I'm reusing the name.
This next bit might sound slightly anal but bear with me. The old Zaphod
was an Alpha so it had a node number 3.33. The new one is
a macmini running Linux (and I hope to replace fentib) so has a node
number of 3.55.
My numbering scheme (as if you cared) goes
01-19 PCs
20-30 hmm, I forget now. I'm sure I had some once though.
30-39 DEC systems
40-49 HP systems
50-59 Apple systems
60-69 Sun systems
So, anyway the point is that can you please renumber ZAPHOD from 3.33 to
3.55. Sorry.
Bob Armstrong wrote:
I made this change on LEGATO too.
Not that it matters, but LEGATO thinks the adjacent node on the circuit is
3.34, ROOSTA. From your numbering system, this must be some kind of VAX or
Alpha.
Yes it's a VAX.
Well, actually it's a simh 'vax'. My real ones are too noisy and/or hot
to run in the summer so I though it would be more reliable to have it
run this way. At least I can keep it going all the time now :)
--
Chrissie
Is anyone on here successfully using DECnet in TOPS-20 with the KLH10
emulator? I'm using a dedicated bridged-tap interface in FreeBSD, and
DECnet is certainly enabled in TOPS-20, but there is absolutely no
DECnet traffic on the wire. None can be seen with tcpdump. I find this
very odd. IP connectivity works fine, and I can see LAT traffic, but no
DECnet. Any suggestions?
Regards,
Mark
I made this change on LEGATO too.
Not that it matters, but LEGATO thinks the adjacent node on the circuit is
3.34, ROOSTA. From your numbering system, this must be some kind of VAX or
Alpha.
Bob
-----Original Message-----
From: owner-hecnet at Update.UU.SE [mailto:owner-hecnet at Update.UU.SE] On Behalf
Of Christine Caulfield
Sent: Thursday, September 11, 2008 5:31 AM
To: hecnet at Update.UU.SE
Subject: [HECnet] Node rename
I'm going to be a pest now, sorry.
My old node ZAPHOD died over a year ago. However I now have a new node
so I'm reusing the name.
This next bit might sound slightly anal but bear with me. The old Zaphod
was an Alpha so it had a node number 3.33. The new one is
a macmini running Linux (and I hope to replace fentib) so has a node
number of 3.55.
My numbering scheme (as if you cared) goes
01-19 PCs
20-30 hmm, I forget now. I'm sure I had some once though.
30-39 DEC systems
40-49 HP systems
50-59 Apple systems
60-69 Sun systems
So, anyway the point is that can you please renumber ZAPHOD from 3.33 to
3.55. Sorry.
--
Chrissie
Christine Caulfield wrote:
I'm going to be a pest now, sorry.
My old node ZAPHOD died over a year ago. However I now have a new node
so I'm reusing the name.
This next bit might sound slightly anal but bear with me. The old Zaphod
was an Alpha so it had a node number 3.33. The new one is
a macmini running Linux (and I hope to replace fentib) so has a node
number of 3.55.
My numbering scheme (as if you cared) goes
01-19 PCs
20-30 hmm, I forget now. I'm sure I had some once though.
30-39 DEC systems
40-49 HP systems
50-59 Apple systems
60-69 Sun systems
So, anyway the point is that can you please renumber ZAPHOD from 3.33 to
3.55. Sorry.
No problemo. :-)
I've done it on MIM now.
Johnny
I think I've found a solution for the infamous SMG bug on releases for which
actually HP does not (still) provide a patch: I've discovered that the
official published patch for OpenVMS VAX V7.3 works apparently flawlessy even
on V7.2 and maybe also on older releases like V7.1 and so on.
Obviously it cannot be installed via VMSINSTAL because the kit does check for
the correct version and release, but it's always possibile to extract the new
SMGSHR.EXE file from the kit and put it in place. To be safe, I've copied it
to the specific portion of my system root (e.g. DKA0:[SYS0.SYSLIB]) instead of
the better common one, but I do not have a cluster and SYS$SHARE does point to
both locations, so I've achieved the same result. After that I've just typed
INSTALL REPLACE SYS$SHARE:SMGSHR.EXE/OPEN/HEADER/SHARED and that's it!
I've done the same for my OpenVMS Alpha V7.2 system by using the V7.3-2 patch,
but in that case a newer release is temporarily needed to extract SMGSHR.EXE
from the PCSI kit because it's in the new compressed form that's not supported
by V7.2 (I do not know when it was introduced, I used a V8.2 to extract).
In the next few days, if I'll see that everything works without any problems,
maybe I'll rename the original SMGSHR.EXE file and replace it with the new one
so to have just one of it and in the right place.
Hope this helps!
G.
