5 Sep
2022
5 Sep
'22
10:50 p.m.
I think I already sent this to all the people who are directly connected
to A2RTR, but just in case I missed anybody, here it is for general
consumption -
I have moved A2RTR to an Amazon cloud server. The new IP is now
35.82.76.235, although I strongly recommend that you use the FQDN
decnet.jfcl.com instead. I've already updated the latter to point to the
new IP.
A few of you with passive, listen, connections on your end and who aren't
checking the source IP don't actually need to do anything. The new A2RTR
will just connect to you as before and you won't notice a difference.
Those who have active connections to A2RTR, or who have some kind of
source IP based filtering in place, will need to update the IP for A2RTR.
Once again, I really suggest that you use decnet.jfcl.com if at all
possible, but if not then the new IP is 35.82.76.235.
Bob
6 Sep
6 Sep
12:54 a.m.
Absolutely awesome. In your next physical, you will find your BP has dropped 10 points.
Thanks.
On Sep 5, 2022, at 4:51 PM, Robert Armstrong
<bob(a)jfcl.com> wrote:
I think I already sent this to all the people who are directly connected to A2RTR, but
just in case I missed anybody, here it is for general consumption –
I have moved A2RTR to an Amazon cloud server. The new IP is now 35.82.76.235, although
I strongly recommend that you use the FQDN decnet.jfcl.com instead. I’ve already updated
the latter to point to the new IP.
A few of you with passive, listen, connections on your end and who aren’t checking the
source IP don’t actually need to do anything. The new A2RTR will just connect to you as
before and you won’t notice a difference.
Those who have active connections to A2RTR, or who have some kind of source IP based
filtering in place, will need to update the IP for A2RTR. Once again, I really suggest
that you use decnet.jfcl.com if at all possible, but if not then the new IP is
35.82.76.235.
Bob
_______________________________________________
HECnet mailing list -- hecnet(a)lists.dfupdate.se
To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se
1:35 a.m.
Supratim Sanyal supratim(a)riseup.net wrote:
you will find your BP has dropped 10 points.
That’s good news – I could use it J
Don’t know about my BP, but it’s amazing how fast the network is in these datacenters.
I think A2RTR has, by DECnet standards at least, essentially infinite bandwidth now.
It’ll run out of CPU power for pyDECnet to forward messages long before it runs out of
bandwidth.
Bob
2:31 a.m.
If we put all routers and all emulated machines in the Amazon cloud,
the who HECnet can live in one VM...
-P
From: "bob" <bob(a)jfcl.com>
To: "The Hobbyist DECnet mailing list" <hecnet(a)lists.dfupdate.se>
Sent: Monday, September 5, 2022 7:35:37 PM
Subject: [HECnet] Re: A2RTR move
_______________________________________________
HECnet mailing list -- hecnet(a)lists.dfupdate.se
To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se
4:36 p.m.
Yep, one Amazon server could probably simulate all the DECnet machines that ever existed
once upon a time… Hopefully somebody out there still has real hardware connected to
HECnet however. I do occasionally, at least for now.
Speaking of which, I notice that IAD2 59.1020, has not connected to the new A2RTR.
STUPI and STORTR have, however. Is there a problem with IAD2?
And speaking of networks, what happened to your ANF10 connections? I’m still running
the PDP10 simh on my home machine, so those IP addresses have not changed. Don’t know if
you were using decnet.jfcl.com for those connections, but if you were then you can’t
anymore. I suppose I could create an anf10.jfcl.com for this purpose… Hang on, I’ll go
do that now…
Ok, done – I just added an anf10.jfcl.com for this purpose.
Bob
From: Peter Lothberg [mailto:roll@stupi.com]
Sent: Monday, September 5, 2022 5:31 PM
To: bob; The Hobbyist DECnet mailing list
Subject: [HECnet] Re: A2RTR move
If we put all routers and all emulated machines in the Amazon cloud,
the who HECnet can live in one VM...
-P
_____
From: "bob" <bob(a)jfcl.com>
To: "The Hobbyist DECnet mailing list" <hecnet(a)lists.dfupdate.se>
Sent: Monday, September 5, 2022 7:35:37 PM
Subject: [HECnet] Re: A2RTR move
_______________________________________________
HECnet mailing list -- hecnet(a)lists.dfupdate.se
To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se
4:47 p.m.
I'm still running real hardware. PONDUS is a 11/93 that is running 24/7.
But no, I doubt one server would even get close to simulate all ever
existing DECnet machines. But it could be an interesting exercise to try
and figure out how many machines you'd need.
DEC's internal DECnet (Easynet) was definitely going over 50.000
machines (I actually don't know how big it was at its maximum). And that
in the 80s. Then you get all the Alphas and Itaniums that came later. If
you want to emulate them, you'll get challenged for CPU and network
resources pretty quickly.
I migrated some PDP-11 systems away from DECnet to TCP/IP that was
running testing things at Saab around 2005. And that was because they
were shutting down DECnet in their facilities, but they wasn't getting
rid of the PDP-11s at that point in time. Who knows how many sites
use/used PDP-11s and DECnet around the world. I know of at least one
place where I know they were still running VAXen and PDP-11s with DECnet
in production as late as last year. I strongly suspect they are still
running their stuff.
Johnny
On 2022-09-06 16:36, Robert Armstrong wrote:
Yep, one Amazon server could probably simulate all
the DECnet
machines that ever existed once upon a time… Hopefully somebody out
there still has real hardware connected to HECnet however. I do
occasionally, at least for now.
Speaking of which, I notice that IAD2 59.1020, has not connected to
the new A2RTR. STUPI and STORTR have, however. Is there a problem
with IAD2?
And speaking of networks, what happened to your ANF10 connections?
I’m still running the PDP10 simh on my home machine, so those IP
addresses have not changed. Don’t know if you were using
decnet.jfcl.com for those connections, but if you were then you can’t
anymore. I suppose I could create an anf10.jfcl.com for this purpose…
Hang on, I’ll go do that now…
Ok, done – I just added an anf10.jfcl.com for this purpose.
Bob
*From:*Peter Lothberg [mailto:roll@stupi.com]
*Sent:* Monday, September 5, 2022 5:31 PM
*To:* bob; The Hobbyist DECnet mailing list
*Subject:* [HECnet] Re: A2RTR move
If we put all routers and all emulated machines in the Amazon cloud,
the who HECnet can live in one VM...
-P
------------------------------------------------------------------------
*From: *"bob" <bob(a)jfcl.com>
*To: *"The Hobbyist DECnet mailing list" <hecnet(a)lists.dfupdate.se>
*Sent: *Monday, September 5, 2022 7:35:37 PM
*Subject: *[HECnet] Re: A2RTR move
_______________________________________________
HECnet mailing list -- hecnet(a)lists.dfupdate.se
To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se
_______________________________________________
HECnet mailing list -- hecnet(a)lists.dfupdate.se
To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt(a)softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
5:50 p.m.
On Sep 6, 2022, at 7:36 AM, Robert Armstrong <bob(a)jfcl.com> wrote:
Yep, one Amazon server could probably simulate all the DECnet machines that ever
existed once upon a time… Hopefully somebody out there still has real hardware connected
to HECnet however. I do occasionally, at least for now.
My PDXVAX system is still running on the real Hardware. It’s a VAXstation 4000/60 with a
BA350 SCSI shelf. It’s not been online for a month or so, but that’s due to the high
temperatures we’ve been having (it currently lives in a rack in my Garage). I hope to be
able to bring it back online in the next week or two.
I have plenty of other DEC hardware that can be brought online, but I don’t normally have
it running. I need to find time to take care of the RIFA cap’s in my BA23 and BA123
chassis’ before I can bring my PDP-11’s online. That’s been on my Todo list for several
years.
Zane
6:02 p.m.
New subject: BA23 RIFAs, was Re: A2RTR move
On 9/6/22 11:50, Zane Healy wrote:
I have plenty of other DEC hardware that can be
brought online, but I
don’t normally have it running. I need to find time to take care of the
RIFA cap’s in my BA23 and BA123 chassis’ before I can bring my PDP-11’s
online. That’s been on my Todo list for several years.
The RIFA capacitors in the BA23 are:
C101: line-line, 0.47uF X, 25mm
C102: line-ground, 0.22uF Y, 20mm
C201: line-line, 0.1uF X2, 20mm
My wife's new company will begin selling capacitor replacement kits
for BA23s (and hundreds of other things) later this year.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
6:37 p.m.
New subject: BA23 RIFAs, was Re: A2RTR move
Dave McGuire <mcguire(a)neurotica.com> wrote:
My wife's new company will begin selling
capacitor replacement kits
for BA23s (and hundreds of other things) later this year.
All caps, including electrolytic, or just the RIFA filter caps?
Best plan is to replace them of course, but I just remove the RIFA caps from my BA23s.
They're in there for EMI filtering and, when you have a whole garage full of old
computer equipment of dubious FCC/EMI compliance anyway, I'm not sure these particular
supplies are a major concern.
Plus there will usually be other electrolytics in the power supply that you're going
to be forced to replace anyway, so you'll get plenty of practice that way.
Bob
6:53 p.m.
New subject: BA23 RIFAs, was Re: A2RTR move
On 9/6/22 12:37, Robert Armstrong wrote:
Just the RIFAs in the BA23 for now, as I've not come up with a list
of the other capacitors yet.
My
wife's new company will begin selling capacitor replacement kits
for BA23s (and hundreds of other things) later this year.
All caps, including electrolytic, or just the RIFA filter caps? Best plan is to replace them of course, but I just
remove the RIFA caps from my BA23s. They're in there for EMI filtering and, when you
have a whole garage full of old computer equipment of dubious FCC/EMI compliance anyway,
I'm not sure these particular supplies are a major concern.
This is true, but it's practical to go the extra mile. Or, in this
case, the proper mile to begin with.
Plus there will usually be other electrolytics in
the power supply that you're going to be forced to replace anyway, so you'll get
plenty of practice that way.
Eventually, yes, but we've not had any of the others fail yet, across
maybe thirty BA23s. As soon as I have time to develop the list(s) of
the other capacitors in the BA23, and identify appropriate replacements,
she'll add those to her database and will be able to offer them as well.
But the RIFAs are blowing up in BA23s left and right today, so that's
what she has.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
7:34 p.m.
New subject: BA23 RIFAs, was Re: A2RTR move
hecOn 9/6/2022 11:02 AM, Dave McGuire wrote:
The RIFA capacitors in the BA23 are:
C101: line-line, 0.47uF X, 25mm
C102: line-ground, 0.22uF Y, 20mm
C201: line-line, 0.1uF X2, 20mm
My wife's new company will begin selling capacitor replacement kits for BA23s (and
hundreds of other things) later this year.
-Dave
Dave,
Question about C102. I've done 3 BA23 P/S and in all three C102 was an 0.22µF X2
There was a thread on the VCF forum about it and I think it originally said it was a Y but
I think it was later found to be incorrect. Have you seen P/S with a 0.22µF Y cap?
--
John H. Reinhardt
9:21 p.m.
New subject: BA23 RIFAs, was Re: A2RTR move
On 9/6/22 13:34, John H. Reinhardt wrote:
Interesting! I have that list in my notes from about ten years ago;
I'm pretty sure I got that info from the capacitors actually present in
one of the BA23s that I worked on. I'd love to know what happened there.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
The RIFA
capacitors in the BA23 are:
C101: line-line, 0.47uF X, 25mm
C102: line-ground, 0.22uF Y, 20mm
C201: line-line, 0.1uF X2, 20mm
My wife's new company will begin selling capacitor replacement kits
for BA23s (and hundreds of other things) later this year.
Question about C102. I've done 3 BA23 P/S and in all three C102 was an
0.22µF X2 There was a thread on the VCF forum about it and I think it
originally said it was a Y but I think it was later found to be
incorrect. Have you seen P/S with a 0.22µF Y cap?
5:57 p.m.
On 9/6/22 10:36, Robert Armstrong wrote:
Yep, one Amazon server could probably simulate all
the DECnet
machines that ever existed once upon a time… Hopefully somebody out
there still has real hardware connected to HECnet however. I do
occasionally, at least for now.
We have a lot of real systems at LSSM that are on HECnet when we have
them powered up:
VAX-11/750 VMS 4.7
VAX-8300 VMS 7.3
VAX-7000 model 720 VMS 7.3
VAX-6000 model 510 VMS 7.3
VAX-4000/90 VMS 7.3
VAX-4000/700a VMS 7.3
MicroVAX-2000 VMS 7.3
MicroVAX-III VMS 7.3
DEC3000-800 VMS 8.2
PDP-11/53 RSTS/E v10.1
PDP-11/70 RSX-11M-PLUS v4.6
PDP-11/44 RSTS/E v10.1
HP rx2660 VMS v8.4
Systems Concepts SC40 TOPS-20 v7
Symbolics XL1201 Genera 8.3
Our VAX-11/780 will hopefully be up and running sometime this winter,
as will the second PDP-11/70. One of the DECsystem-2020s may come up
this winter as well, depending on volunteer time availability. I just
started putting together a MicroVAX-II dual-booting VMS 4.7 and Ultrix,
and will try to get DECnet running under Ultrix on that.
Another project that we may do this winter is getting SGI 4DDN
(DECnet for IRIX) running on one of the Irises.
That's at LSSM. At home, I have a MicroVAX-II and VAX-4000 model 96
running VMS and a PDP-11/83 running RSX-11M-PLUS that are on HECnet when
I have them up and running. Soon a MicroVAX-I will join them.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
8:32 p.m.
On 2022-09-06 17:57, Dave McGuire wrote:
We have a lot of real systems at LSSM that are on
HECnet when we have
them powered up:
VAX-11/750 VMS 4.7
VAX-8300 VMS 7.3
VAX-7000 model 720 VMS 7.3
VAX-6000 model 510 VMS 7.3
VAX-4000/90 VMS 7.3
VAX-4000/700a VMS 7.3
MicroVAX-2000 VMS 7.3
MicroVAX-III VMS 7.3
DEC3000-800 VMS 8.2
PDP-11/53 RSTS/E v10.1
PDP-11/70 RSX-11M-PLUS v4.6
PDP-11/44 RSTS/E v10.1
HP rx2660 VMS v8.4
Systems Concepts SC40 TOPS-20 v7
Symbolics XL1201 Genera 8.3
Our VAX-11/780 will hopefully be up and running sometime this winter,
as will the second PDP-11/70. One of the DECsystem-2020s may come up
this winter as well, depending on volunteer time availability. I just
started putting together a MicroVAX-II dual-booting VMS 4.7 and Ultrix,
and will try to get DECnet running under Ultrix on that.
The -2020 will be tricky to get up on HECnet, though. As far as I
remember/know, it only officially had Phase III. And you need to run
TOPS-20 V4.
Not sure if Tops-10 ever had DECnet on them, but if it did, I would
suspect it's the same story there. The -2020s I ever used just ran
Tops-10 with ANF-10.
But let's hope you can get some of those other systems up and online a
bit more.
Another project that we may do this winter is
getting SGI 4DDN
(DECnet for IRIX) running on one of the Irises.
That's at LSSM. At home, I have a MicroVAX-II and VAX-4000 model 96
running VMS and a PDP-11/83 running RSX-11M-PLUS that are on HECnet when
I have them up and running. Soon a MicroVAX-I will join them.
The MVI is another interesting beast. Not sure when VMS stopped
supporting that one, but it was probably quite some time ago. Max 4MB of
memory is somewhat limiting. :-)
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt(a)softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
8:40 p.m.
On Sep 6, 2022, at 2:32 PM, Johnny Billquist
<bqt(a)softjar.se> wrote:
On 2022-09-06 17:57, Dave McGuire wrote:
TOPS-20 V4 has DECnet Phase II. I've used it for Phase II testing. PyDECnet can
handle that, at least in T1.1. That will do "intercept" service for the TOPS-20
node, making all of HECnet visible. Yes, you actually get better access than a Phase III
node would (since those are limited to in-area only).
The intercept machinery is not 100% yet, because Phase II assumes no packet loss and in
Phase III and up that isn't a valid assumption. Also, intercept has to track the
connections and close them if the other end goes away. I don't have tracking yet, and
when I add that I figure I can make the intercept service also do NSP packet
retransmission on behalf of the Phase II client.
Still, the existing machinery works reasonably well. I've done file access from
TOPS-20 V4 to RSX; that works except that directory listings are somewhat messed up (dates
look weird). Haven't figured out why yet. And while rumor has it that there is a
cross-protocol remote terminal client on the support tapes, I haven't figured out how
to build those. It doesn't help that I'm an utter TOPS novice, TOPS-20
especially.
paul
We have a lot of real systems at LSSM that are
on HECnet when we have them powered up:
VAX-11/750 VMS 4.7
VAX-8300 VMS 7.3
VAX-7000 model 720 VMS 7.3
VAX-6000 model 510 VMS 7.3
VAX-4000/90 VMS 7.3
VAX-4000/700a VMS 7.3
MicroVAX-2000 VMS 7.3
MicroVAX-III VMS 7.3
DEC3000-800 VMS 8.2
PDP-11/53 RSTS/E v10.1
PDP-11/70 RSX-11M-PLUS v4.6
PDP-11/44 RSTS/E v10.1
HP rx2660 VMS v8.4
Systems Concepts SC40 TOPS-20 v7
Symbolics XL1201 Genera 8.3
Our VAX-11/780 will hopefully be up and running sometime this winter, as will the
second PDP-11/70. One of the DECsystem-2020s may come up this winter as well, depending
on volunteer time availability. I just started putting together a MicroVAX-II
dual-booting VMS 4.7 and Ultrix, and will try to get DECnet running under Ultrix on that.
The -2020 will be tricky to get up on HECnet, though. As far as I remember/know, it only
officially had Phase III. And you need to run TOPS-20 V4.
8:41 p.m.
On Sep 6, 2022, at 2:32 PM, Johnny Billquist
<bqt(a)softjar.se> wrote:
Not sure if Tops-10 ever had DECnet on them, but if it did, I would suspect it's the
same story there. The -2020s I ever used just ran Tops-10 with ANF-10.
Yes, it did, and I remember Phase IV being worked on by that team. One other interesting
tidbit I remember is that its delay estimator (in the NSP layer) used floating point --
because it could. No one else did that to my knowledge, and there isn't a whole lot
of reason for it.
paul
9:21 p.m.
On 9/6/22 14:32, Johnny Billquist wrote:
Well, we'll do the best we can, and I'll learn those limitations. I
myself was never a PDP-10 guy; I cut my teeth on PDP-8, PDP-11, and VAX
systems. I'm slowly learning TOPS-20 on the SC-40 so we can open it up
for giving accounts to people here. (and elsewhere)
Our
VAX-11/780 will hopefully be up and running sometime this
winter, as will the second PDP-11/70. One of the DECsystem-2020s may
come up this winter as well, depending on volunteer time
availability. I just started putting together a MicroVAX-II
dual-booting VMS 4.7 and Ultrix, and will try to get DECnet running
under Ultrix on that.
The -2020 will be tricky to get up on HECnet, though. As far as I
remember/know, it only officially had Phase III. And you need to run
TOPS-20 V4.
Not sure if Tops-10 ever had DECnet on them, but if it did, I would
suspect it's the same story there. The -2020s I ever used just ran
Tops-10 with ANF-10. But let's hope you can get some of those other
systems up and online a
bit more.
We're doing the best we can. I think we've done very well with what
little help we've had. I've had a soldering iron inside just about
every one of those machines.
The MVI is another interesting beast. Not sure when
VMS stopped
supporting that one, but it was probably quite some time ago. Max 4MB of
memory is somewhat limiting. :-)
Yes, and having that memory directly on the Qbus is pretty limiting
too. But still, strangely, the machine has personality. I ran one (as
a VAXstation-I) back in the 1980s. It wasn't fast, but it was a VAX,
and it was mine, when all my friends were farting around with their
Commodore 64s! B-)
VMS officially de-supported the MicroVAX-I as of, I think, v6.2, but
I don't know if support was actually removed. That said, running a VMS
release that late on a -I would be extremely painful, so I'll probably
run 4.7 on this one, to get at least some performance out of it.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
7 Sep
7 Sep
1:07 a.m.
Dave,
I ran VMS on the CANINE:: notes server at DEC. It was a MicroVAX-I with dual RC-25’s.
When you bring it up, let me know and I can have a go at tuning it. You WILL be surprised
at what you can get out of that little beast! I know - I was… :-)
-Steve
SF:IP2
On Sep 6, 2022, at 15:21, Dave McGuire
<mcguire(a)neurotica.com> wrote:
On 9/6/22 14:32, Johnny Billquist wrote:
Well, we'll do the best we can, and I'll learn those limitations. I myself was
never a PDP-10 guy; I cut my teeth on PDP-8, PDP-11, and VAX systems. I'm slowly
learning TOPS-20 on the SC-40 so we can open it up for giving accounts to people here.
(and elsewhere)
Our
VAX-11/780 will hopefully be up and running sometime this winter, as will the second
PDP-11/70. One of the DECsystem-2020s may come up this winter as well, depending on
volunteer time availability. I just started putting together a MicroVAX-II dual-booting
VMS 4.7 and Ultrix, and will try to get DECnet running under Ultrix on that.
The
-2020 will be tricky to get up on HECnet, though. As far as I remember/know, it only
officially had Phase III. And you need to run TOPS-20 V4.
Not sure if Tops-10 ever had DECnet on them, but if it did, I would suspect it's the
same story there. The -2020s I ever used just ran Tops-10 with ANF-10. But let's hope you can get some of those
other systems up and online a bit more.
We're doing the best we can. I think we've done very well with what little help
we've had. I've had a soldering iron inside just about every one of those
machines.
The MVI is another interesting beast. Not sure
when VMS stopped supporting that one, but it was probably quite some time ago. Max 4MB of
memory is somewhat limiting. :-)
Yes, and having that memory directly on the Qbus is pretty limiting too. But still,
strangely, the machine has personality. I ran one (as a VAXstation-I) back in the 1980s.
It wasn't fast, but it was a VAX, and it was mine, when all my friends were farting
around with their Commodore 64s! B-)
VMS officially de-supported the MicroVAX-I as of, I think, v6.2, but I don't know if
support was actually removed. That said, running a VMS release that late on a -I would be
extremely painful, so I'll probably run 4.7 on this one, to get at least some
performance out of it.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
_______________________________________________
HECnet mailing list -- hecnet(a)lists.dfupdate.se
To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se
2:38 a.m.
Ah, thanks Steve, will do! I've got to get the MicroVAX-II out of
the way first, I hope to finish that up by this weekend, or at least
next week.
If anyone has any VR-260 monitors lying dormant, we're looking for a
couple for the museum. Working preferred, but we'll fix them if not.
-Dave
On 9/6/22 19:07, Steve Davidson wrote:
Dave,
I ran VMS on the CANINE:: notes server at DEC. It was a MicroVAX-I with dual RC-25’s.
When you bring it up, let me know and I can have a go at tuning it. You WILL be surprised
at what you can get out of that little beast! I know - I was… :-)
-Steve
SF:IP2
--
Dave McGuire, AK4HZ
New Kensington, PA
On Sep 6, 2022, at 15:21, Dave McGuire
<mcguire(a)neurotica.com> wrote:
On 9/6/22 14:32, Johnny Billquist wrote:
Well, we'll do the best we can, and I'll learn those limitations. I myself was
never a PDP-10 guy; I cut my teeth on PDP-8, PDP-11, and VAX systems. I'm slowly
learning TOPS-20 on the SC-40 so we can open it up for giving accounts to people here.
(and elsewhere)
_______________________________________________
HECnet mailing list -- hecnet(a)lists.dfupdate.se
To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se Our
VAX-11/780 will hopefully be up and running sometime this winter, as will the second
PDP-11/70. One of the DECsystem-2020s may come up this winter as well, depending on
volunteer time availability. I just started putting together a MicroVAX-II dual-booting
VMS 4.7 and Ultrix, and will try to get DECnet running under Ultrix on that.
The
-2020 will be tricky to get up on HECnet, though. As far as I remember/know, it only
officially had Phase III. And you need to run TOPS-20 V4.
Not sure if Tops-10 ever had DECnet on them, but if it did, I would suspect it's the
same story there. The -2020s I ever used just ran Tops-10 with ANF-10. But let's hope you can get some of those
other systems up and online a bit more.
We're doing the best we can. I think we've done very well with what little
help we've had. I've had a soldering iron inside just about every one of those
machines.
The MVI is another interesting beast. Not sure
when VMS stopped supporting that one, but it was probably quite some time ago. Max 4MB of
memory is somewhat limiting. :-)
Yes, and having that memory directly on the Qbus is pretty limiting too. But still,
strangely, the machine has personality. I ran one (as a VAXstation-I) back in the 1980s.
It wasn't fast, but it was a VAX, and it was mine, when all my friends were farting
around with their Commodore 64s! B-)
VMS officially de-supported the MicroVAX-I as of, I think, v6.2, but I don't know
if support was actually removed. That said, running a VMS release that late on a -I would
be extremely painful, so I'll probably run 4.7 on this one, to get at least some
performance out of it.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
_______________________________________________
HECnet mailing list -- hecnet(a)lists.dfupdate.se
To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se 
10:19 p.m.
Well, I'm open for questions. My only problem is I'm so swamped that my
turnaround time is frequently abysmal...
Do you know how many sections the SC40 supports? The KL supports 32.
The reason I'm asking is because supporting DECnet in Kermit turned out
to be a far bigger job than I ever expected and my section zero virtual
address space started feeling a little cramped due to all the .PSECT's
I'm using.
What I did was move all the program help and text into section 1, viz:
@g kermit
@i mem
152. pages, Entry vector loc 614522 len 3
Section 0 R, W, E, Private
2 KERMIT.EXE.2559 1 R, E
6 KERMIT.EXE.2559 2 R, E
10 KERMIT.EXE.2559 3 R, E
11 KERMIT.EXE.2559 4 R, CW, E
20 KERMIT.EXE.2559 5 R, E
30 KERMIT.EXE.2559 6 R, E
35-36 KERMIT.EXE.2559 7-10 R, CW, E
42 KERMIT.EXE.2559 11 R, CW, E
100-101 KERMIT.EXE.2559 12-13 R, CW, E
103-104 KERMIT.EXE.2559 14-15 R, CW, E
106 KERMIT.EXE.2559 16 R, CW, E
500-615 KERMIT.EXE.2559 17-134 R, E
Section 1 R, W, E, Private
1001-1055 KERMIT.EXE.2559 135-211 R, E
1300-1315 KERMIT.EXE.2559 212-227 R, E
1500 KERMIT.EXE.2559 230 R, E
So that's 60. pages in section one. One doesn't want to 'bump' into
things... I also have guard pages to catch those pesky off-by-one errors.
------------------------------------------------------------------------
On 9/6/22 3:21 PM, Dave McGuire wrote:
Well, we'll do the best we can, and I'll learn those limitations. I
myself was never a PDP-10 guy; I cut my teeth on PDP-8, PDP-11, and
VAX systems. I'm slowly learning TOPS-20 on the SC-40 so we can open
it up for giving accounts to people here. (and elsewhere)
6 Sep
6 Sep
9:41 p.m.
Johnny Billquist <bqt(a)softjar.se> wrote:
The -2020 will be tricky to get up on HECnet, though. As far as I
remember/know, it only officially had Phase III. And
you need to run
TOPS-20 V4.
Not sure if Tops-10 ever had DECnet on them, but if it
did, I would
suspect it's the same story there. The -2020s I
ever used just ran
Tops-10 with ANF-10.
Oh, come on - be serious... TOPS-10 had DECnet Phase-IV, both on the KL and the KS
(2020) machines. There are even some such machines on HECnet today!!
$ set host venti
SIMH KS10 VENTI 7.04 NET 12:36:12 TTY10 system 4999
Connected to Node LEGATO Line # 0
AUTHORIZED USERS ONLY!
Please LOGIN
.net
[ANF10 network: connected to VENTI(20), located at VENTI(20), 3 nodes]
Node VENTI (20) SIMH KS10 VENTI 7.04 NET 18-Nov-21
Node TWONKY (37) TWONKY KS10 TOPS10 7.04 28-Nov-21
Node TWONK2 (40) TWONK2 KS10 TOPS10 7.04 28-Nov-21
[DECnet network: local node VENTI, 17 reachable nodes in area 2]
(2.99) (2.152) (2.153) (2.170) (2.199) A2RTR BOINGO GLDRTR HUSKY LEGATO
MASTIF REACH THEARK TOMMYT VENTI VENTI2 ZITI
.KJOB
.ini
SIMH KS10 VENTI 7.04 NET 12:36:21 TTY10 system 4999
Connected to Node LEGATO Line # 0
AUTHORIZED USERS ONLY!
Please LOGIN
.log 13,730
Job 14 SIMH KS10 VENTI 7.04 NET TTY0
Password:
[LGNLAS Last access to [13,730] succeeded on 6-Sep-122:07:27:46]
12:38 6-Sep-122 Tuesday
.set host mim
[Connected to RSX-11M+ system MIM::, using CTERM protocol]
[Type ^\,<RET> to return]
?NRTRMT Required message types not all supported
.
It even understands how to talk to other areas, although MIM doesn’t seem very
friendly.
The 2020 machines never had Ethernet, at least not in any official capacity. Perhaps
that’s what you are confused about? A 2020 only supported synchronous links and could
only be an end node.
Bob
9:45 p.m.
On 9/6/22 15:41, Robert Armstrong wrote:
The 2020 machines never had Ethernet, at least not
in any official
capacity. Perhaps that’s what you are confused about? A 2020 only
supported synchronous links and could only be an end node.
So no just sticking a DEUNA board set in that BA11-K? Damn. Is that
"just" a matter of device drivers?
One of LSSM's -2020s has a CH11 controller. (Chaosnet)
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
9:59 p.m.
So no just sticking a DEUNA board set in that BA11-K?
It's a long story. The hardware (a DEUNA in the 2020 UBA/UNIBUS box) will work, but
the software never supported it. It was mostly a memory issue. The KS10 is limited to
512KW of physical memory and space is very tight on the 2020.
Supposedly Mark Crispin had Ethernet working with TOPS20 on his 2020, but that was never
released as a product and the status of that software today is, I think, unknown.
Bob
11:23 p.m.
On 2022-09-06 21:59, Robert Armstrong wrote:
The DEUNA/DELUA is possible to stick in there. Jonny Eriksson was
working on support, but it was more than a memory limitation issue. The
DEUNA/DELUA was never design with the 18-bit transfers in mind. But as
far as I know, he got IP and UDP running over ethernet on a KS with Tops-10.
So no just
sticking a DEUNA board set in that BA11-K?
It's a long story. The hardware (a DEUNA in the 2020 UBA/UNIBUS box) will work,
but the software never supported it. It was mostly a memory issue. The KS10 is limited
to 512KW of physical memory and space is very tight on the 2020. Supposedly Mark Crispin had Ethernet working with
TOPS20 on his 2020, but that was never released as a product and the status of that
software today is, I think, unknown.
That is something I don't think I ever heard. But maybe I just missed
it. But I seem to remember that MRC worked on Phase IV for KS, and got
that working. But not using ethernet as far as I know. But I really
should try to find where that was discussed. Should be available online
somewhere...
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt(a)softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
7 Sep
7 Sep
12:36 a.m.
On 2022-09-06 23:23, Johnny Billquist wrote:
On 2022-09-06 21:59, Robert Armstrong wrote:
The DEUNA/DELUA is possible to stick in there. Jonny Eriksson was
working on support, but it was more than a memory limitation issue. The
DEUNA/DELUA was never design with the 18-bit transfers in mind. But as
far as I know, he got IP and UDP running over ethernet on a KS with
Tops-10.
Found a relevant thread:
http://alt.sys.pdp10.narkive.com/ceLh26Dc/decnet-for-ks10-tops-20-only
Seems MRC did the Phase IV for TOPS-20 on KS, but no ethernet. bygg did
ethernet for KS but only for IP under Tops-10. Some hints and hooks seem
to have existed for ethernet, but no code seem to actually have existed.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt(a)softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
So no
just sticking a DEUNA board set in that BA11-K?
It's a long story. The hardware (a DEUNA in the 2020 UBA/UNIBUS
box) will work, but the software never supported it. It was mostly a
memory issue. The KS10 is limited to 512KW of physical memory and
space is very tight on the 2020. Supposedly Mark Crispin had Ethernet working
with TOPS20 on his
2020, but that was never released as a product and the status of that
software today is, I think, unknown.
That is something I don't think I ever heard. But maybe I just missed
it. But I seem to remember that MRC worked on Phase IV for KS, and got
that working. But not using ethernet as far as I know. But I really
should try to find where that was discussed. Should be available online
somewhere...
6 Sep
6 Sep
11:08 p.m.
I admit that my Tops-10 knowledge is limited. But I very clearly
remember MRC unofficially getting Phase IV running under TOPS-20 on a
KS. This was work that was somehow supported/requested by DEC if I
remember things. But it never became an official release.
MRC's comments were basically that he barely managed to get it running,
with some limitations, and not much resources left for much else.
With Tops-10, like I said, I only ever used them with ANF-10 networks.
And yeah, I know that KS never officially supported ethernet. In theory
the Unibus ethernet controllers should be possible to use, but I know
that Jonny Eriksson (bygg) was working on that, and had things half
working. I think I might even have seen the code, but it never was finished.
Johnny
On 2022-09-06 21:41, Robert Armstrong wrote:
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt(a)softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
Johnny
Billquist <bqt(a)softjar.se <mailto:bqt@softjar.se>> wrote:
The -2020 will be tricky to get up on HECnet, though. As far as I
remember/know, it only officially had Phase III.
And you need to run
TOPS-20 V4.
Not sure if Tops-10 ever had DECnet on them, but
if it did, I would
suspect it's the same story there. The -2020s
I ever used just ran
Tops-10 with ANF-10.
Oh, come on - be serious... TOPS-10 had DECnet Phase-IV, both on the
KL and the KS (2020) machines. There are even some such machines on
HECnet today!!
$ set host venti
SIMH KS10 VENTI 7.04 NET 12:36:12 TTY10 system 4999
Connected to Node LEGATO Line # 0
AUTHORIZED USERS ONLY!
Please LOGIN
.net
[ANF10 network: connected to VENTI(20), located at VENTI(20), 3 nodes]
Node VENTI (20) SIMH KS10 VENTI 7.04 NET 18-Nov-21
Node TWONKY (37) TWONKY KS10 TOPS10 7.04 28-Nov-21
Node TWONK2 (40) TWONK2 KS10 TOPS10 7.04 28-Nov-21
[DECnet network: local node VENTI, 17 reachable nodes in area 2]
(2.99) (2.152) (2.153) (2.170) (2.199) A2RTR BOINGO GLDRTR HUSKY
LEGATO
MASTIF REACH THEARK TOMMYT VENTI VENTI2 ZITI
.KJOB
.ini
SIMH KS10 VENTI 7.04 NET 12:36:21 TTY10 system 4999
Connected to Node LEGATO Line # 0
AUTHORIZED USERS ONLY!
Please LOGIN
.log 13,730
Job 14 SIMH KS10 VENTI 7.04 NET TTY0
Password:
[LGNLAS Last access to [13,730] succeeded on 6-Sep-122:07:27:46]
12:38 6-Sep-122 Tuesday
.set host mim
[Connected to RSX-11M+ system MIM::, using CTERM protocol]
[Type ^\,<RET> to return]
?NRTRMT Required message types not all supported
.
It even understands how to talk to other areas, although MIM doesn’t
seem very friendly.
The 2020 machines never had Ethernet, at least not in any official
capacity. Perhaps that’s what you are confused about? A 2020 only
supported synchronous links and could only be an end node.
Bob
_______________________________________________
HECnet mailing list -- hecnet(a)lists.dfupdate.se
To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se
7 Sep
7 Sep
10:29 p.m.
Yes, MRC did that by getting Tops-20 version 5 running on a KS10, which
he said was possible with some amount of effort.
You'd have to hide symbols (that code exists in 4.2) and then modify all
the .PSECT's to be in section zero. That would be a tedious amount of
tinkering because DEC never finished relocatable .PSECT's and you've got
to move everything around by hand. Then you have to remove any section
specific instructions (XMOVEI, XHHLI, Etc.) and other such hackery.
It's been on my list of things to look at, but I want to put ANF-10
(TSK:) into Tops-20 first as well as put MIT Chaos back in.
On 9/6/22 5:08 PM, Johnny Billquist wrote:
I admit that my Tops-10 knowledge is limited. But I
very clearly
remember MRC unofficially getting Phase IV running under TOPS-20 on a
KS. This was work that was somehow supported/requested by DEC if I
remember things. But it never became an official release.
MRC's comments were basically that he barely managed to get it
running, with some limitations, and not much resources left for much
else.
With Tops-10, like I said, I only ever used them with ANF-10 networks.
And yeah, I know that KS never officially supported ethernet. In
theory the Unibus ethernet controllers should be possible to use, but
I know that Jonny Eriksson (bygg) was working on that, and had things
half working. I think I might even have seen the code, but it never
was finished.
Johnny
On 2022-09-06 21:41, Robert Armstrong wrote:
Johnny
Billquist <bqt(a)softjar.se <mailto:bqt@softjar.se>> wrote:
The -2020 will be tricky to get up on HECnet, though. As far as I
remember/know, it only officially had Phase III.
And you need to run
TOPS-20 V4.
Not sure if Tops-10 ever had DECnet on them, but
if it did, I would
suspect it's the same story there. The -2020s
I ever used just ran
Tops-10 with ANF-10.
Oh, come on - be serious... TOPS-10 had DECnet Phase-IV, both on
the KL and the KS (2020) machines. There are even some such machines
on HECnet today!!
$ set host venti
SIMH KS10 VENTI 7.04 NET 12:36:12 TTY10 system 4999
Connected to Node LEGATO Line # 0
AUTHORIZED USERS ONLY!
Please LOGIN
.net
[ANF10 network: connected to VENTI(20), located at VENTI(20), 3 nodes]
Node VENTI (20) SIMH KS10 VENTI 7.04 NET 18-Nov-21
Node TWONKY (37) TWONKY KS10 TOPS10 7.04 28-Nov-21
Node TWONK2 (40) TWONK2 KS10 TOPS10 7.04 28-Nov-21
[DECnet network: local node VENTI, 17 reachable nodes in area 2]
(2.99) (2.152) (2.153) (2.170) (2.199) A2RTR BOINGO GLDRTR
HUSKY LEGATO
MASTIF REACH THEARK TOMMYT VENTI VENTI2 ZITI
.KJOB
.ini
SIMH KS10 VENTI 7.04 NET 12:36:21 TTY10 system 4999
Connected to Node LEGATO Line # 0
AUTHORIZED USERS ONLY!
Please LOGIN
.log 13,730
Job 14 SIMH KS10 VENTI 7.04 NET TTY0
Password:
[LGNLAS Last access to [13,730] succeeded on 6-Sep-122:07:27:46]
12:38 6-Sep-122 Tuesday
.set host mim
[Connected to RSX-11M+ system MIM::, using CTERM protocol]
[Type ^\,<RET> to return]
?NRTRMT Required message types not all supported
.
It even understands how to talk to other areas, although MIM
doesn’t seem very friendly.
The 2020 machines never had Ethernet, at least not in any official
capacity. Perhaps that’s what you are confused about? A 2020 only
supported synchronous links and could only be an end node.
Bob
_______________________________________________
HECnet mailing list -- hecnet(a)lists.dfupdate.se
To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se
10:25 p.m.
I can personally attest that portions of the Tops-10 DECnet
implementation are quite robust. I have several hours worth of tests
that I have run against Kermit-10 on VENTI:: (Tops-*10*) from VENTI2::
(Tops-*20*). That cross platform regression testing has enabled me to
find several bugs and I am hopeful of doing something similar against
Ultrix at some point.
Why 'portions'? The reason I started the NRT implementation for Tops-29
Kermit was because the regression tests I was running on Tops-20 DAP
were blowing Tops-10 FAL right out of the water; bad enough that Bob had
to reboot the machine.
So I needed some other mechanism to pull files off for debugging and...
there you are....
On 9/6/22 3:41 PM, Robert Armstrong wrote:
Johnny Billquist <bqt(a)softjar.se> wrote:
The -2020 will be tricky to get up on HECnet, though. As far as I
remember/know, it only officially had Phase III.
And you need to run
TOPS-20 V4.
Not sure if Tops-10 ever had DECnet on them, but
if it did, I would
suspect it's the same story there. The -2020s
I ever used just ran
Tops-10 with ANF-10.
Oh, come on - be serious... TOPS-10 had DECnet Phase-IV, both on
the KL and the KS (2020) machines. There are even some such machines
on HECnet today!!
$ set host venti
SIMH KS10 VENTI 7.04 NET 12:36:12 TTY10 system 4999
Connected to Node LEGATO Line # 0
AUTHORIZED USERS ONLY!
Please LOGIN
.net
[ANF10 network: connected to VENTI(20), located at VENTI(20), 3 nodes]
Node VENTI (20) SIMH KS10 VENTI 7.04 NET 18-Nov-21
Node TWONKY (37) TWONKY KS10 TOPS10 7.04 28-Nov-21
Node TWONK2 (40) TWONK2 KS10 TOPS10 7.04 28-Nov-21
[DECnet network: local node VENTI, 17 reachable nodes in area 2]
(2.99) (2.152) (2.153) (2.170) (2.199) A2RTR BOINGO GLDRTR HUSKY
LEGATO
MASTIF REACH THEARK TOMMYT VENTI VENTI2 ZITI
.KJOB
.ini
SIMH KS10 VENTI 7.04 NET 12:36:21 TTY10 system 4999
Connected to Node LEGATO Line # 0
AUTHORIZED USERS ONLY!
Please LOGIN
.log 13,730
Job 14 SIMH KS10 VENTI 7.04 NET TTY0
Password:
[LGNLAS Last access to [13,730] succeeded on 6-Sep-122:07:27:46]
12:38 6-Sep-122 Tuesday
.set host mim
[Connected to RSX-11M+ system MIM::, using CTERM protocol]
[Type ^\,<RET> to return]
?NRTRMT Required message types not all supported
.
It even understands how to talk to other areas, although MIM doesn’t
seem very friendly.
The 2020 machines never had Ethernet, at least not in any official
capacity. Perhaps that’s what you are confused about? A 2020 only
supported synchronous links and could only be an end node.
Bob
_______________________________________________
HECnet mailing list --hecnet(a)lists.dfupdate.se
To unsubscribe send an email tohecnet-leave(a)lists.dfupdate.se 
6 Sep
6 Sep
8:10 p.m.
On Sep 5, 2022, at 13:50, Robert Armstrong
<bob(a)jfcl.com> wrote:
I think I already sent this to all the people who are directly connected to A2RTR, but
just in case I missed anybody, here it is for general consumption –
I don't recall getting an off-list email about it (which does not necessarily mean
that I did not get one), but in any case my home router was already connecting via the
FQDN instead of the IP address, and it appears to be happily connected. So, I assume that
it just seamlessly followed the virtual A2RTR move. Yay! I took down the machine at work
that made my second connection, and I haven't gotten around to plugging it in anywhere
else yet. I expect it'll just work when I plug it in and turn it on again.
My always-on machines are SIMH instances, but I also have real hardware which I bring up
on HECnet from time to time. I'm looking forward to bringing my VAX-11/730 up in fall
or winter of this year, but I'm not turning on that space heater during the heat wave
that is cooking me right now.
Just out of curiosity, did anything in particular prompt you to move A2RTR into the cloud?
It seems to me like a logical way to deploy pyDECnet based HECnet backbone routers. I had
briefly toyed with Amazon cloud services a few years ago, with the thought of moving my
Wordpress site to one, and maybe trying to implement other silly retrocomputing-related
services. But I decided it was too far outside my comfort zone and dropped the project.
I'm pleased to see DECnet running on the cloud in 2022. I wonder if I might look into
deploying retro fun on the cloud again sometime in the future?
--
Mark J. Blair <nf6x(a)nf6x.net>
Blog: https://www.nf6x.net
Git: https://gitlab.com/users/NF6X/groups
HECnet: DOGPAK::MBLAIR
837
days inactive
839
days old
28 comments
11 participants
participants (11)
-
Dave McGuire -
John H. Reinhardt -
Johnny Billquist -
Mark J. Blair -
Paul Koning -
Peter Lothberg -
Robert Armstrong -
Steve Davidson -
Supratim Sanyal -
Thomas DeBellis -
Zane Healy