DECserver 300 questions
Michael Kukat
Talking 'bout terminals... I use a DECserver 300 to control most of my
VAX consoles via telnet, and i have a little problem, which i solved a
slightly uncommon way recently.
The DECserver doesn't seem to know about a way to send a break from my
telnet session to the serial port to get to the console prompt or so. I
checked nearly all configuration parameters, a BREAK just isn't generated.
Does someone know a solution for this problem?
If not, here is my solution: decrease the speed of the port to 75 Bps with
CHANGE PORT 2 SPEED 75
or so, press space several times, and re-change the speed to 9600 Bps
again. You have to disconnect the session after the speed changes, but
this is a usable, but complicated way to generate a BREAK my VAXen accept.
So long... Michael
J. Buck Caldwell
Have you tried ^P? Isn't that supposed to be interperteted by the VAX as a
break?
Just a thought - I don't have my DECServer 300 running yet, due to a
detestable lack of MMJ cables.
Michael Kukat wrote:
> Hi all!
>
> Talking 'bout terminals... I use a DECserver 300 to control most of my
> VAX consoles via telnet, and i have a little problem, which i solved a
> slightly uncommon way recently.
> The DECserver doesn't seem to know about a way to send a break from my
> telnet session to the serial port to get to the console prompt or so. I
> checked nearly all configuration parameters, a BREAK just isn't generated.
>
> Does someone know a solution for this problem?
--
- J. Buck Caldwell
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GAT d- s:+ a C++$ UB++$ P>+++ L- !E W++ N+ o K? w+++$ O- M+$ V(+) PS@ !PE
Y+ PGP t++(-) 5+++ X- R tv+ b+++ DI+++ D(++) G e(+++) h r+++ y++++>$
------END GEEK CODE BLOCK------
Michael Kukat
"J. Buck Caldwell" wrote:
> Have you tried ^P? Isn't that supposed to be interperteted by the VAX as a
> break?
I know this from my 4000/300 where i can do "SET CONTROLP 1" to enable
this feature. But this is the only VAX reacting on ^P. Of course i tried
it, and i even wired this key to the BREAK function in the terminal
server, but with no success. It really seems to be thought as a
_TERMINAL_ server, not as a _CONSOLE_ server, as BREAK from the terminal
can be used as disconnect for telnet sessions :)
> Just a thought - I don't have my DECServer 300 running yet, due to a
> detestable lack of MMJ cables.
Just use those modem connectors, RJ11, i got 100m of the cable, 50
connectors and the tool to make my cables for under $50 in germany. And
with a side cutter, just cut away half of the pin and the leading block,
if you work exactly enough, you get a very well MMJ cable which even
locks the connector and doesn't fall out when touching it. Works fine here.
And the DECserver 200/MC has usual 25pin DSub connectors, but only talks
LAT. Nice in combination with a VXT2000 :)
...Michael
der Mouse
>> VAX as a break?
> I know this from my 4000/300 where i can do "SET CONTROLP 1" to
> enable this feature. But this is the only VAX reacting on ^P.
I'm fairly sure that's not true. Back when I used VAXen at work, one
of them used ^P to break to the console, and I doubt the 4000/300 even
existed back then. I think it was the 780, though it might have been
the 750.
der Mouse
mo...@rodents.montreal.qc.ca
7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
Carlini, Antonio
VAX as a
>> break?
The early VAXen interpreted ^P in hardware.
The later MicroVAXen mostly expected a <BREAK>.
> server, but with no success. It really seems to be thought as a
> _TERMINAL_ server, not as a _CONSOLE_ server, as BREAK from the
terminal
> can be used as disconnect for telnet sessions :)
If you get to the terminal server via LAT (i.e. you
have the server offering a LAT service for
the port which has the console attached) then
you can do this (at least it worked for me!).
Via telnet, maybe you can do this. Certainly the
OpenVMS and Solaris telnets allow you to do
"send brk". Whether this gets gobbled up by the
terminal server or actually passed on I have no
way of testing right now.
Antonio
Michael Kukat
der Mouse wrote:
> > I know this from my 4000/300 where i can do "SET CONTROLP 1" to
> > enable this feature. But this is the only VAX reacting on ^P.
>
> I'm fairly sure that's not true. Back when I used VAXen at work, one
> of them used ^P to break to the console, and I doubt the 4000/300 even
> existed back then. I think it was the 780, though it might have been
> the 750.
I don't have a 750 or a 780, maybe it is true on these machines, but at
least the 4000 series doesn't seem to listen for ^P by default. Ok, it
even might be that the DECserver300 filters this, but i don't think so,
because ^P is echoed. Have to look into the firmware, if i can enable it
on the 4000/105A, as i can on the 4000/300, but by default, none of the
VAXstation 4000 or VAX 4000 machines here knows about ^P coming from the
DECserver 300.
...Michael
Michael Kukat
"Carlini, Antonio" wrote:
> > server, but with no success. It really seems to be thought as a
> > _TERMINAL_ server, not as a _CONSOLE_ server, as BREAK from the
> terminal
> > can be used as disconnect for telnet sessions :)
>
> If you get to the terminal server via LAT (i.e. you
> have the server offering a LAT service for
> the port which has the console attached) then
> you can do this (at least it worked for me!).
I connected internally with LAT, so did a login to the server ans issued
my "CONNECT PORT02". But as during the telnet connection, the BREAK will
be discarded, it never reaches the destination service, it is never sent
to PORT02.
> Via telnet, maybe you can do this. Certainly the
> OpenVMS and Solaris telnets allow you to do
> "send brk". Whether this gets gobbled up by the
> terminal server or actually passed on I have no
> way of testing right now.
send brk exists in telnet, works with other terminal servers, and is
silently ignored or translated to a character by the DS300. Too bad,
there exists no character "BRK" or so, as in the input side somewhere.
...Michael
Peter Svensson
> I connected internally with LAT, so did a login to the server ans issued
> my "CONNECT PORT02". But as during the telnet connection, the BREAK will
> be discarded, it never reaches the destination service, it is never sent
> to PORT02.
Connecting through LAT on a DECserver200 sending break works. I don't
remember any more whether some settings had to be made to allow break to
be sent.
Peter
--
Peter Svensson ! Pgp key available by finger, fingerprint:
<pet...@psv.nu> ! 8A E9 20 98 C1 FF 43 E3 07 FD B9 0A 80 72 70 AF
<pet...@df.lth.se> !
------------------------------------------------------------------------
Remember, Luke, your source will be with you... always...
Peter Svensson
> I don't have a 750 or a 780, maybe it is true on these machines, but at
> least the 4000 series doesn't seem to listen for ^P by default. Ok, it
> even might be that the DECserver300 filters this, but i don't think so,
> because ^P is echoed. Have to look into the firmware, if i can enable it
> on the 4000/105A, as i can on the 4000/300, but by default, none of the
> VAXstation 4000 or VAX 4000 machines here knows about ^P coming from the
> DECserver 300.
Most of the "bigger" vaxen I have worked with use ctrl-p to get to the
chevron prompt. This includes the 8350 eg. I think nautilus series used
ctrl-p as well.
Carlini, Antonio
> Michael Kukat wrote:
>
>I connected internally with LAT, so did a login to the server ans
issued
>my "CONNECT PORT02". But as during the telnet connection, the BREAK
will
>be discarded, it never reaches the destination service, it is never
sent
>to PORT02.
I used terminal servers that had been set up to advertise
a LAT service on a given port (via DEFINE SERVICE iirc)
and then used SET HOST/LAT from an OpenVMS
workstation.
In addition, the BREAK stuff only works on the
alternate console (again, that's from memory).
Antonio
Paul Thompson
perhaps there are port settings that differ between the two that could be
set on the 300. Let's see a SHOW PORT from the 200 that works, and a
SHOW PORT from the 300 and determine what's different. There are a lot of
customizable settings hidden in there.
--
Michael Kukat
Matthew Fredette wrote:
> There shouldn't be any reason why I can't just put up a complete
> source distribution plus my changes, right? I haven't talked with
> latd's author about any of this, so I'm a little hesitant, but I don't
> really know why :)
Hm. Is it GPL or just free? Could be interesting, and much easier to
write it from scratch :)
...Michael
Michael Kukat
Paul Thompson wrote:
> If someone says they are able to get break to work on a 200 and not a 300
> perhaps there are port settings that differ between the two that could be
> set on the 300. Let's see a SHOW PORT from the 200 that works, and a
> SHOW PORT from the 300 and determine what's different. There are a lot of
> customizable settings hidden in there.
I don't think the behaviour is different in 200 and 300 models, i just
said, a _TELNET_ session discards any tries to send a break to the LAT
service. At home, i could just use my 200 as a real _TERMINAL_ server to
connect the 300 in the rack, but i need my break via telnet on the
terminal server ports, and exactly here is my only problem. Never said,
it doesn't work via LAT, the setting is something like "CHANGE PORT <n>
BREAK REMOTE" or something like that. But the "send brk" in a telnet
session is ignored. This is the only problem, LAT works.
Hm... are there any tries to get some LAT stuff running under NetBSD? :)
Would be cool to have some nodes in /dev talking to the DECserver ports
via LAT. I saw some sources for this for Lin** recently, maybe i should
have a closer look at this, i just need the protocol specs.
...Michael
Matthew Fredette
> Hm... are there any tries to get some LAT stuff running under NetBSD? :)
>
> Would be cool to have some nodes in /dev talking to the DECserver ports
> via LAT. I saw some sources for this for Lin** recently, maybe i should
> have a closer look at this, i just need the protocol specs.
I was fooling around with that lin** latd a few months ago, because I
have a DECserver 200/MC. I wrapped an autoconf/automake framework
around latd to try to make it more multiplatform, and added some bits
to have it support NetBSD.
I had it compiling on NetBSD/i386 and NetBSD/vax (both 1.5) and doing
some very preliminary things, like picking up the services announced
by the DECserver but I eventually got distracted from working on it.
There shouldn't be any reason why I can't just put up a complete
source distribution plus my changes, right? I haven't talked with
latd's author about any of this, so I'm a little hesitant, but I don't
really know why :)
Matt
--
Matt Fredette
fred...@aquery.com, fred...@mit.edu, fred...@theory.lcs.mit.edu
http://mit.edu/fredette/www
"If you understood everything I said, you'd be me." - Miles Davis
Matthew Fredette
> Hm. Is it GPL or just free? Could be interesting, and much easier to
> write it from scratch :)
It's GPL'ed. I agree that a new BSD-licensed version would be better.
I started working with the lin** version for two reasons: faster path
to getting something useable, and less exposure to hassle over the
patent.
Michael Kukat
Matthew Fredette wrote:
> > Hm. Is it GPL or just free? Could be interesting, and much easier to
> > write it from scratch :)
>
> It's GPL'ed. I agree that a new BSD-licensed version would be better.
> I started working with the lin** version for two reasons: faster path
> to getting something useable, and less exposure to hassle over the
> patent.
BSD licensing would be much better. Ok, we have enough GPL-packages in
the pkgsrc tree, so this shouldn't make a problem.
In my first experiments some years ago, i wondered if tcpdumping to find
out the protocol is some kind of reverse engineering? I don't think so.
LAT is patented i think, but does Compaq care too much about this?
Hmm... i'll look a bit deeper in this when i find the time. For now the
baudrate trick works, i don't need to break my VAXen that often :)
...Michael
Kenn Humborg
> out the protocol is some kind of reverse engineering? I don't think so.
> LAT is patented i think, but does Compaq care too much about this?
The author is based in the UK. I _think_ reverse engineering for
the purposes for interoperability is legal in the UK. I'm fairly
sure it's legal here in Ireland, so it could be an EU-wide thing.
Later,
Kenn
Johnny Billquist
> Hm... are there any tries to get some LAT stuff running under NetBSD? :)
Should be. But LAT is actually a COMPAQ licensed technology...
> Would be cool to have some nodes in /dev talking to the DECserver ports
> via LAT. I saw some sources for this for Lin** recently, maybe i should
> have a closer look at this, i just need the protocol specs.
http://linux-decnet.sourceforge.net/lat.html
ftp://gatekeeper.dec.com/pub/DEC/DECnet/PhaseIV/
http://www.sucs.swan.ac.uk/~rohan/DECnet/index.html
that's all I have found on the subject. DECnet are perhaps not interesting
from a LAT point of view, but it would also be very nice to have...
Johnny
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: b...@update.uu.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
Anders Magnusson
reverse engineering and reimplementation of the IBM PC BIOS.
-- Ragge
Kenn Humborg
But, I have a feeling that, in the US, you have to do a
clean-room implementation (person A reverse engineers and
writes a spec, gives the spec to person B and person B
re-implements it).
IANAL, but I don't think you have to do this here in Europe.
Perhaps someone can clarify?
Later,
Kenn
Kenn Humborg
> >Yes, it is (reverse engineering). It's also so in the US; the
> most famous
> >case is Compaq's
> >reverse engineering and reimplementation of the IBM PC BIOS.
>
> the BIOS situation. This is the law under which the DeCSS programmers are
> being prosecuted I believe.
This is getting scarily off topic, but...
It worrying that US laws are being applied to a Norwegian
national. Or are they going to extradite him?
Later,
Kenn
Chuck McManis
>reverse engineering and reimplementation of the IBM PC BIOS.
Except that the "Millennium Copyright Act" made it illegal and post dates
the BIOS situation. This is the law under which the DeCSS programmers are
being prosecuted I believe.
--Chuck
Matt Thomas
That's only for trade-secrets. LAT is patented and reverse engineering
a patent still infringes on the patent.
BTW, when I worked for Digital, I owned the implementation of LAT for
ULTRIX and (for awhile) Digital UNIX.
I don't know whether Compaq retained the patent rights for LAT or
whether it transferred to Cabletron when Digital sold its network
division to them. And if it did, did it transfer to Digital Network
Products Group (www.dnpg.com) when they split off from Cabletron.