Have you checked that node DUSTY is sending its router hello packets on the Ethernet circuit?
Does node LULU receive the router hellos from node DUSTY?
I do have "Hello timer = 15" on DUSTY. I do have connectivity between LULU and DUSTY, but now I wonder whether it's full connectivity. Perhaps there's a problem with Personal Alpha's network emulation. It may not be receiving all of the packets.
What does node LULU indicate as the adjacency? It should be node DUSTY unless you have another Level 1 (with higher priority) or Level 2 router on the same circuit.
When I do a "SHOW CIRC SVA-0 STAT" on LULU, I don't see any adjacent node.
--Marc
LULU, on the other hand, can only connect to DUSTY. It can't connect to
LEGATO.
Hmm. What makes you think this? I ask because LEGATO can connect to LULU
(I just tried it a minute ago), so it is working at least one way.
That is strange... I saw your connection attempt, but my attempts to contact LEGATO don't work as well:
$ set host legato
%SYSTEM-F-UNREACHABLE, remote node is not currently reachable
When I look in NCP, I see something that doesn't look right.
NCP>show node legato
Node Volatile Summary as of 6-APR-2009 12:06:19
Node State Active Delay Circuit Next node
Links
2.1 (LEGATO) SVA-0 0
I would expect that the next node should be 2.210 (DUSTY), but it doesn't know how to get there.
--Marc
Marc Chametzky wrote:
I have my new system, LULU, up and it's connected via Ethernet on the same network segment as DUSTY. LULU is a non-routing end node.
DUSTY is "TYPE ROUTING IV" and has two circuits, TCP-0-0 to LEGATO and QNA-0 for the local network. DUSTY has connectivity to both LEGATO (and the rest of HECnet) and also to LULU.
LULU, on the other hand, can only connect to DUSTY. It can't connect to LEGATO.
I thought that DUSTY, being a routing node, would route packets between its two circuits, but that appears not to be the case. I tried setting DUSTY up as an area router, thinking that making it a level 2 router would permit LULU's packets to go through DUSTY, but that didn't seem to work either.
What should I do in order to get DUSTY to route the packets from LULU to the rest of HECnet?
--Marc
.
Have you checked that node DUSTY is sending its router hello packets on the Ethernet circuit?
Does node LULU receive the router hellos from node DUSTY?
What does node LULU indicate as the adjacency? It should be node DUSTY unless you have another Level 1 (with higher priority) or Level 2 router on the same circuit.
How are you trying to connect? Do you have an up-to-date node list on LULU?
SET HOST LEGATO
or
SET HOST 2.1
Zane
On Mon, 6 Apr 2009, Marc Chametzky wrote:
LULU, on the other hand, can only connect to DUSTY. It can't connect to
LEGATO.
Hmm. What makes you think this? I ask because LEGATO can connect to LULU
(I just tried it a minute ago), so it is working at least one way.
That is strange... I saw your connection attempt, but my attempts to contact LEGATO don't work as well:
$ set host legato
%SYSTEM-F-UNREACHABLE, remote node is not currently reachable
When I look in NCP, I see something that doesn't look right.
NCP>show node legato
Node Volatile Summary as of 6-APR-2009 12:06:19
Node State Active Delay Circuit Next node
Links
2.1 (LEGATO) SVA-0 0
I would expect that the next node should be 2.210 (DUSTY), but it doesn't know how to get there.
--Marc
Marc Chametzky wrote
DUSTY is "TYPE ROUTING IV" and has two circuits, TCP-0-0 to LEGATO and
QNA-0 for the local network. DUSTY has connectivity to both LEGATO (and
the rest of HECnet) and also to LULU.
This is the way it should be.
I tried setting
DUSTY up as an area router, thinking that making it a level 2 router
would permit LULU's packets to go through DUSTY, but that didn't seem to
work either.
An area router isn't necessary since all your nodes (and LEGATO too) are
in the same area.
LULU, on the other hand, can only connect to DUSTY. It can't connect to
LEGATO.
Hmm. What makes you think this? I ask because LEGATO can connect to LULU
(I just tried it a minute ago), so it is working at least one way.
Bob
I have my new system, LULU, up and it's connected via Ethernet on the same network segment as DUSTY. LULU is a non-routing end node.
DUSTY is "TYPE ROUTING IV" and has two circuits, TCP-0-0 to LEGATO and QNA-0 for the local network. DUSTY has connectivity to both LEGATO (and the rest of HECnet) and also to LULU.
LULU, on the other hand, can only connect to DUSTY. It can't connect to LEGATO.
I thought that DUSTY, being a routing node, would route packets between its two circuits, but that appears not to be the case. I tried setting DUSTY up as an area router, thinking that making it a level 2 router would permit LULU's packets to go through DUSTY, but that didn't seem to work either.
What should I do in order to get DUSTY to route the packets from LULU to the rest of HECnet?
--Marc
Guys, just in case any of you fancy a play, I've installed MP/M II V2.1 on a SIMH Altair Z80 emulator.
It's accessible through CHIMPY by logging in as ALTAIR. There are a number of floppy drives (write locked) and a hard drive (writeable) as follows:
A: System Disk
B: Multiplan
C: VEDIT
D: WordStar
E: Games
I: Writeable hard drive.
Let me know how it's working!.
Sampsa
Marc Chametzky wrote:
http://www.sparetimegizmos.com/Downloads/DCN%20Node%20List.pdf
DUSTY and CALHAN appear to have the wrong addresses ...
I'd also like to add 2.212 as LULU, a Personal Alpha running OpenVMS ...
Ok, I've updated these things on the list.
Bob
Marc Chametzky wrote:
http://www.sparetimegizmos.com/Downloads/DCN%20Node%20List.pdf
Looking at this list, last updated in August 2008, I notice that my systems, DUSTY and CALHAN appear to have the wrong addresses. They are 2.210 and 2.211, respectively. The NCP database on LEGATO has them correct.
I'd also like to add 2.212 as LULU, a Personal Alpha running OpenVMS V8.3. I haven't yet decided where I'm going to run LULU, either in my VMware Fusion system on occasion, or as part of my "trailing-zone" environment in ESXi.
I've added LULU on MIM now.
I thought I'd take this chance to make two comments.
1. Nodenames in DECnet phase IV are actually local to each machine, so there is no strict requirement that we all have the same view on the name to address mappings. That said, it sure helps if we do, so I try to keep MIM up to date, and others can copy the nodename database from there (in VMS, you just do a "NCP COPY KNOWN NODES FROM MIM TO BOTH", if someone wants to know how you do in RSX, write me a mail. :-) )
2. In addition to the nodename database, I also keep a plaintext file with both nodenames people use, and nodenames people have asked to be reserved. You can find that at MIM::US:[DECNET]NODENAMES.ALL
In the same directory I also have a DECNET.TXT file, which have a list of all areas allocated, and by whom.
(And the file from sparetimesgizmos is not my creation or responsibility, so I can't make any changes in that.)
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
Marc Chametzky wrote:
http://www.sparetimegizmos.com/Downloads/DCN%20Node%20List.pdf
Looking at this list, last updated in August 2008, I notice that my systems, DUSTY and CALHAN appear to have the wrong addresses. They are 2.210 and 2.211, respectively. The NCP database on LEGATO has them correct.
I'd also like to add 2.212 as LULU, a Personal Alpha running OpenVMS V8.3. I haven't yet decided where I'm going to run LULU, either in my VMware Fusion system on occasion, or as part of my "trailing-zone" environment in ESXi.
Ok. I've added LULU at my "master" database on MIM.
As for that list, I'm not the one compiling it, so I can't make any changes, or comments on it.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
http://www.sparetimegizmos.com/Downloads/DCN%20Node%20List.pdf
Looking at this list, last updated in August 2008, I notice that my systems, DUSTY and CALHAN appear to have the wrong addresses. They are 2.210 and 2.211, respectively. The NCP database on LEGATO has them correct.
I'd also like to add 2.212 as LULU, a Personal Alpha running OpenVMS V8.3. I haven't yet decided where I'm going to run LULU, either in my VMware Fusion system on occasion, or as part of my "trailing-zone" environment in ESXi.
--Marc
Just a quick update:
I've registered the domain CHIMPYMAIL.COM for use as a hecnet mail gateway.
The format of the messages is the same, i.e. incoming mail should be addressed to:
<HOST>::<USER>@chimpymail.com
Outgoing mail remains the same, i.e.:
CHIMPY::smtp%"user at example.com"
Sampsa
I want to say that there was some freeware utility, perhaps on the DECUS tapes throughout the years, that offered very basic UPS monitoring. If I remember correctly, a UPS that's having a power-fault would drop DTE or something like that, allowing a program to monitor the serial port for a hangup, and then do whatever (presumably start a countdown before taking action to see if DTE is raised again).
I don't know whether the current models do anything that basic any longer.
--Marc
Fred wrote:
Good afternoon ...
Would anyone have a piece of software that would run on OpenVMS Alpha that could monitor an APC Smart-UPS via a serial port? I moved some equipment around and now have one UPS that ended up not being monitored, and I'd like to have the Alpha monitor it (and run shutdown.com on low battery). I did a little googling around but couldn't find anything. I've seen Powerchute for VAX but it had a rather obscene price tag attached to it.
Cheers,
MISER::FRED :)
.
The only UPS monitoring software for OpenVMS Alpha I know of is for Exide (aka Powerware) UPS's. Don't know if it would work with other UPS's as well.
Regards,
Kari
Good afternoon ...
Would anyone have a piece of software that would run on OpenVMS Alpha that could monitor an APC Smart-UPS via a serial port? I moved some equipment around and now have one UPS that ended up not being monitored, and I'd like to have the Alpha monitor it (and run shutdown.com on low battery). I did a little googling around but couldn't find anything. I've seen Powerchute for VAX but it had a rather obscene price tag attached to it.
Cheers,
MISER::FRED :)
Begin forwarded message:
From: Johnny Billquist <bqt at softjar.se>
Date: 9 March 2009 22:01:54 GMT
To: hecnet at Update.UU.SE
Subject: Re: [HECnet] DEC fan meeting in Lake District, anyone from HECnet want to speak?
Reply-To: hecnet at Update.UU.SE
majordomo at update.uu.se
Just put a "subscribe hecnet" in the body.
Johnny
Sampsa Laine wrote:
Oh ok, I just forgot the subscription address, someone remind me? :)
Sampsa
On 9 Mar 2009, at 21:42, Johnny Billquist wrote:
Sampsa Laine wrote:
Ladies, Gentlemen,
I've been talking to a Mr Mark Williams today about some VMS related stuff, wonderfully helpful chap etc etc.
Anyway, he asked me to relay this message to our mailing list in case we were interested in attending or even speaking at this "old DEC kit" gathering.
If it had been any closer... ;-)
Might I also suggest that we add Mr Williams to this mailing list as he's likely to be a great asset to HECnet.
In general I dislike adding someone else on their behalf. I prefer that they request to subscribe themself.
Johnny
Sampsa
Begin forwarded message:
*From: *Mark Wickens <mark at wickensonline.co.uk <mailto:mark at wickensonline.co.uk> <mailto:mark at wickensonline.co.uk>>
While I've got your attention - I'm organising a meeting later in the
year that will focus on old DEC kit - HECNET would definitely be of
interest - do you know of anyone in the HECNET community who I might
approach about giving a short talk? It'll be autumn this year (when I
get my finger out and book a venue!) up here in the Lake District.
Sounds like you might be interested in this sort of thing - do you want
me to add you to the distribution list?
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se <mailto:bqt at softjar.se> || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
majordomo at update.uu.se
Just put a "subscribe hecnet" in the body.
Johnny
Sampsa Laine wrote:
Oh ok, I just forgot the subscription address, someone remind me? :)
Sampsa
On 9 Mar 2009, at 21:42, Johnny Billquist wrote:
Sampsa Laine wrote:
Ladies, Gentlemen,
I've been talking to a Mr Mark Williams today about some VMS related stuff, wonderfully helpful chap etc etc.
Anyway, he asked me to relay this message to our mailing list in case we were interested in attending or even speaking at this "old DEC kit" gathering.
If it had been any closer... ;-)
Might I also suggest that we add Mr Williams to this mailing list as he's likely to be a great asset to HECnet.
In general I dislike adding someone else on their behalf. I prefer that they request to subscribe themself.
Johnny
Sampsa
Begin forwarded message:
*From: *Mark Wickens <mark at wickensonline.co.uk <mailto:mark at wickensonline.co.uk> <mailto:mark at wickensonline.co.uk>>
While I've got your attention - I'm organising a meeting later in the
year that will focus on old DEC kit - HECNET would definitely be of
interest - do you know of anyone in the HECNET community who I might
approach about giving a short talk? It'll be autumn this year (when I
get my finger out and book a venue!) up here in the Lake District.
Sounds like you might be interested in this sort of thing - do you want
me to add you to the distribution list?
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se <mailto:bqt at softjar.se> || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
Oh ok, I just forgot the subscription address, someone remind me? :)
Sampsa
On 9 Mar 2009, at 21:42, Johnny Billquist wrote:
Sampsa Laine wrote:
Ladies, Gentlemen,
I've been talking to a Mr Mark Williams today about some VMS related stuff, wonderfully helpful chap etc etc.
Anyway, he asked me to relay this message to our mailing list in case we were interested in attending or even speaking at this "old DEC kit" gathering.
If it had been any closer... ;-)
Might I also suggest that we add Mr Williams to this mailing list as he's likely to be a great asset to HECnet.
In general I dislike adding someone else on their behalf. I prefer that they request to subscribe themself.
Johnny
Sampsa
Begin forwarded message:
*From: *Mark Wickens <mark at wickensonline.co.uk <mailto:mark at wickensonline.co.uk>>
While I've got your attention - I'm organising a meeting later in the
year that will focus on old DEC kit - HECNET would definitely be of
interest - do you know of anyone in the HECNET community who I might
approach about giving a short talk? It'll be autumn this year (when I
get my finger out and book a venue!) up here in the Lake District.
Sounds like you might be interested in this sort of thing - do you want
me to add you to the distribution list?
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
Sampsa Laine wrote:
Ladies, Gentlemen,
I've been talking to a Mr Mark Williams today about some VMS related stuff, wonderfully helpful chap etc etc.
Anyway, he asked me to relay this message to our mailing list in case we were interested in attending or even speaking at this "old DEC kit" gathering.
If it had been any closer... ;-)
Might I also suggest that we add Mr Williams to this mailing list as he's likely to be a great asset to HECnet.
In general I dislike adding someone else on their behalf. I prefer that they request to subscribe themself.
Johnny
Sampsa
Begin forwarded message:
*From: *Mark Wickens <mark at wickensonline.co.uk <mailto:mark at wickensonline.co.uk>>
While I've got your attention - I'm organising a meeting later in the
year that will focus on old DEC kit - HECNET would definitely be of
interest - do you know of anyone in the HECNET community who I might
approach about giving a short talk? It'll be autumn this year (when I
get my finger out and book a venue!) up here in the Lake District.
Sounds like you might be interested in this sort of thing - do you want
me to add you to the distribution list?
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
Ladies, Gentlemen,
I've been talking to a Mr Mark Williams today about some VMS related stuff, wonderfully helpful chap etc etc.
Anyway, he asked me to relay this message to our mailing list in case we were interested in attending or even speaking at this "old DEC kit" gathering.
Might I also suggest that we add Mr Williams to this mailing list as he's likely to be a great asset to HECnet.
Sampsa
Begin forwarded message:
From: Mark Wickens <mark at wickensonline.co.uk>
While I've got your attention - I'm organising a meeting later in the
year that will focus on old DEC kit - HECNET would definitely be of
interest - do you know of anyone in the HECNET community who I might
approach about giving a short talk? It'll be autumn this year (when I
get my finger out and book a venue!) up here in the Lake District.
Sounds like you might be interested in this sort of thing - do you want
me to add you to the distribution list?
Hello all,
does anyone here has any experience running a TOPS-10 host with DECnet-10?
More than a year ago, we connected to our DECnet a KLH10 emulator running
TOPS-10 and its DECnet-10 software, but we had to shut it down just after some
days because it was causing us many troubles and had serious usability issues.
Now we would like to try again, so I'm writing here looking for some ideas.
We would prefer to use KLH10 instead of SIMH because of its quite interesting
multiprocess implementation which scales way better on the hosting system.
Moreover, the KLH10 emulator uses a very effective idling mechanism, by means
of a special device which has to be accessed by the guest OS in its idle loop.
Activation of both this idle mechanism (patching CLOCK1.MAC) and DECnet-10
requires a monitor rebuild, which should be almost easy because we have all
the sources and instructions, but which is actually not so straight-forward.
By reading manuals and Usenet posts, we found that we'd have to use the MONGEN
procedure to rebuild the monitor. So we patched CLOCK1.MAC and answered all
the relevant MONGEN questions about DECnet, but the final result was not good.
(Probably we ran NETGEN too, but at the moment we are not sure about this).
DECnet did start and we were able to make connections to and from the TOPS-10
host, but other parts of the system (e.g. BATCON) stopped working and, most of
all, there was absolutely no idling of the KLH10 host CPU.
We discovered that probably MONGEN only compiles some macro configuration
files and then relinks the monitor using a prebuilt TOPS10.REL, made up from
many macro files, including CLOCK1.MAC. So we need to entirely rebuild
TOPS10.REL from sources, but we do not know ho to do that.
The Software Installation Guide directs to Appendix B where it is suggested
that, for more information about building the monitor library file (i.e.
TOPS10.REL), we should read the batch control file COMPIL.CTL. Unfortunately
COMPIL.CTL does NOT contain any comment or explanation. We think it is
intended for somebody with a previous somewhat deep knowledge about such
tasks, which we do not actually have. we are using a quite standard
installation so common defaults should be fine, but there are anyway lots of
totally obscure files to customize (BUILDS.MIC, BUILDS.PTH, GEN.MIC, ...).
Any help, suggestion or pointer would be much appreciated!
G. (http://decnet.ipv7.net)
For anyone who might be interested:
I've done a tiny modification to the TETRIS game on CHIMPY, the keys are now more sensible if you don't like using a numeric keypad (I'm mostly on my Macbook Pro so I don't).
So If you're bored and fancy a game of tetris, log in to CHIMPY as TETRIS and have at it. I might even send out some type of award to the high scorer of the year should there be significant take up :)
Sampsa
On 2 Mar 2009, at 08:12, Sampsa Laine wrote:
Done.
Sampsa
On 2 Mar 2009, at 01:17, Steve Davidson wrote:
Sampsa,
Could you update the DECnet database using MIM:: please. Area 19
appears to be unknown to your system.
Thanks.
-Steve
-----Original Message-----
From: owner-hecnet at Update.UU.SE [mailto:owner-hecnet at Update.UU.SE] On
Behalf Of Sampsa Laine
Sent: Sunday, March 01, 2009 19:09
To: hecnet at Update.UU.SE
Subject: Re: [HECnet] What use are *you* making of HECnet?
Ok, so I just realised CHIMPY is in fact acting as a gateway into and
out of HECnet already :)
To send mail in:
Address the message to "<HECNET NODE>::
<USER>"@chimpymail.sampsa.com
To send mail out
Address the message to CHIMPY::SMTP%"<Internet email address>"
Sampsa
On 1 Mar 2009, at 23:53, Peter Lothberg wrote:
If I remember right, it was/is Robert Armstrong who had that up and
running.
LEGATO used to relay email between HECnet (using MAIL11) and the
Internet/SMTP, but I quit doing that last fall when I switched
registrars.
There's no reason I couldn't - I just never bothered to set up the
necessary
domain name and mail routing again.
It was pretty much never used, even by me - after all, everyone
with a
HECnet connection already has an Internet connection.
Bob
We could talk "SMTP" over DECnet.....
-P
My system (DUSTY) is a SIMH VAX running on Linux (ubuntu JeOS) under an
ESXi hypervisor. I've given the Linux system two NICs (that's easy in
ESXi), one for Linux, and one dedicated to the VAX. For my purposes, I
also have the VAX's interface set up on a VLAN to move it onto a
separate virtual network for my "trailing edge" systems.
The VAX speaks DECnet-over-MultiNet to LEGATO.
Yes, the Multinet link provides a second connection into HECnet,
Christine and I have been the sites routing between the two types of
connections. There were also one or two people using Cisco router
links as well, but I don't know if anyone still has those links up.
Do we need any additional such routing points? Now that I have an ESXi system, it's trivial for me to add another small virtual machine that could act as such a bridge.
--Marc
My NetBSD machine runs the two NIC's so that I *can* keep DECnet apart from IP. I have dedicated LAN's for each protocol here. Overkill??? Absolutely! It allows me to duplicate the multiple environments that I may encounter during the course of things. It sure does make it easier when debugging a network problem/outage.
The VAXen have only one NIC and as such run only DECnet (and of course SCS traffic). IP is relegated to other machines/hardware/paths. It does take more hardware but in the end it is VERY clean. The Alpha's, on the other hand, have multiple NIC's - one for IP, one for DECnet, and one other.
-Steve
From: owner-hecnet at Update.UU.SE [mailto:owner-hecnet at Update.UU.SE] On Behalf Of Sampsa Laine
Sent: Monday, March 02, 2009 09:16
To: hecnet at Update.UU.SE
Subject: Re: [HECnet] How do you connect to HECnet?
Weirdly enough, I have no problem with running the bridge and SIMH on the same machine - caveat: I am running on OS X.
At the moment, I have the following:
- Gorilla (OS X 10.4.something) running the bridge on interface 0
- GORVAX (SIMH, OpenVMS 7.3, MULTINET) running on interface 0, acting as a router between the bridge and a MULTINET tunnel to the outside
- CHIMPY (running VMS 8.3, TCP/IP services), acting as an end node on the bridged network
So on the main host, there's only one interface and it all works (all = DECNET, I can't get IP traffic from Gorilla to GORVAX).
Sampsa
On 2 Mar 2009, at 09:50, Angela Kahealani wrote:
On Sunday, 2009-03-01 21:50:24 Johnny Billquist wrote:
The one limitation is that it is usually problematic to have the same
machine that is running the bridge also participate as a DECnet node
on the same interface. Because then you need to see all your own
outgoing packets. Not all hardware will do that for you.
So, if you try to run in one machine,
with hardware that can't see its own packets,
then you need 3 interfaces,
1: To talk TCP/IP / UDP for main interface plus UDP side of bridge
2: DECnet side of bridge
3: DECnet speaking software
or
a machine with only one interface running the bridge program, and
a machine with two interfaces, one for TCP/IP/UDP and one DECnet.
Or have I still misunderstood?
Aloha, Angela
--
"(I'll) Be Seeing You..." All information and transactions are
private between the parties, and are non negotiable. All rights
reserve without prejudice, Angela Kahealani. http://kahealani.com