20 Dec
2019
20 Dec
'19
7:12 p.m.
I have been working with a kind volunteer who is using OpenVMS Alpha.?
It's been quite productive.
While I haven't run into any security exposures, yet, the experience has
enabled me to find a few things to tweak up.? For example, the VMS
logging and reporting was clearly superior to what was being put in the
FAL log (which wasn't even timestamped and didn't have the requested
file name).
We ran into some interesting behavior.? When ANONYMOUS access is
restricted to the stand-in user and its subdirectories, wildcard listing
requests that are outside of that lexical scope do not report any errors
but rather return zero files (none).? That was puzzling because the part
of the code that did the access check was clearly doing the reject with
appropriate return values.
Further analysis showed that in the main loop for stepping wildcards, a
reject causes the file in question to be skipped; nothing is actually
done with the return values.? I believe this may be intended behavior
for two reasons.? The first is that if the wildcard includes files to
which the ANONYMOUS user does legitimately have access, it's friendlier
to report those as opposed to stopping everything.
The other reason I believe has to do with DAP itself; I don't think you
can't return more than one return value per ACCESS request, even with a
wildcarded file specification.? So perhaps this isn't a bug so much as
an unintended consequence?? Does anybody know what the behavior should be?
At any rate, I don't believe the current behavior is 'informative'.?
What I'm thinking of doing is saving the last error code and returning
that if the count of files returned in list is zero.? Otherwise, the
user had something to look at.
Do I have any volunteers who operate something besides VMS?? I'd be
interested in RSX(+), IAS, RSTS and OS-8.? They never did DECnet on
RT-11, did they?
20 Dec
20 Dec
7:33 p.m.
On Dec 20, 2019, at 1:12 PM, Thomas DeBellis
<tommytimesharing at gmail.com> wrote:
I have been working with a kind volunteer who is using OpenVMS Alpha. It's been
quite productive.
While I haven't run into any security exposures, yet, the experience has enabled me
to find a few things to tweak up. For example, the VMS logging and reporting was clearly
superior to what was being put in the FAL log (which wasn't even timestamped and
didn't have the requested file name).
We ran into some interesting behavior. When ANONYMOUS access is restricted to the
stand-in user and its subdirectories, wildcard listing requests that are outside of that
lexical scope do not report any errors but rather return zero files (none). That was
puzzling because the part of the code that did the access check was clearly doing the
reject with appropriate return values.
Not clear if that's an error. In RSTS at least, if you asked for a list of files,
you'd be given the subset of files that you have read access to. Files for which you
have no permission would appear nonexistent. This is rather different from the behavior
in Unix, which shows you all files in a brief listing, and a pile of error messages in a
long listing.
Whether VMS follows the same model I'm not sure. It's a bit more work in systems
like RSX or VMS, which like Unix separate directory names from the underlying file object
where the permissions live.
Further analysis showed that in the main loop for
stepping wildcards, a reject causes the file in question to be skipped; nothing is
actually done with the return values. I believe this may be intended behavior for two
reasons. The first is that if the wildcard includes files to which the ANONYMOUS user
does legitimately have access, it's friendlier to report those as opposed to stopping
everything.
The other reason I believe has to do with DAP itself; I don't think you can't
return more than one return value per ACCESS request, even with a wildcarded file
specification. So perhaps this isn't a bug so much as an unintended consequence?
Does anybody know what the behavior should be?
At any rate, I don't believe the current behavior is 'informative'. What
I'm thinking of doing is saving the last error code and returning that if the count of
files returned in list is zero. Otherwise, the user had something to look at.
Do I have any volunteers who operate something besides VMS? I'd be interested in
RSX(+), IAS, RSTS and OS-8. They never did DECnet on RT-11, did they?
I can offer help with RSTS. Yes, there is DECnet/RT11, but I don't know if we have
any on HECnet or if the code has even been preserved.
paul
8:28 p.m.
On Dec 20, 2019, at 10:33 AM, Paul Koning <paulkoning at comcast.net> wrote:
I can offer help with RSTS. Yes, there is DECnet/RT11, but I don't know if we have
any on HECnet or if the code has even been preserved.
paul
I?ve been trying to find a copy for well over 20 years. There may be a copy in a storage
container back east, but I?ve never heard further from the person.
Zane
21 Dec
21 Dec
3:06 p.m.
Zane,
DECnet/RT11 is one of the most significant pieces of lost PDP-11 software. I haven?t
been searching for it as long as you, but it is one of the few remaining things I?d like
to see the DEC software archeologists find. My most promising lead was that I heard of a
professor who taught computer networking ?back in the day? and he used two PDT 11/150s
connected with a serial port to demonstrate computer networking. Thus DECnet/RT11 was
small enough to boot on a dual RX02 system under RT-11 in 32KW Ram.
Phase III DECnet can be very small if DDCMP is the protocol. We used to run 11/23s
(128KW) with RSX11S that were downline loaded (no disk) their OS and they served 20 users
(1 DUV11 and 5 DZV11 or DLV11-J serial cards) with a commodity trading information system.
Of course that was re-entrant Macro-11code we wrote but still amazing in the day.
Best,
Mark
On Dec 20, 2019, at 1:28 PM, Zane Healy <healyzh at
avanthar.com> wrote:
On Dec 20, 2019, at 10:33 AM, Paul Koning <paulkoning at comcast.net> wrote:
I can offer help with RSTS. Yes, there is DECnet/RT11, but I don't know if we have
any on HECnet or if the code has even been preserved.
paul
I?ve been trying to find a copy for well over 20 years. There may be a copy in a storage
container back east, but I?ve never heard further from the person.
Zane
4:59 p.m.
New subject: Tops-20 ANONYMOUS FAL Preliminary Testing Results/DAP Query
I?m the one who has the RX50?s that contain DECnet/RT-11. The problem is I only can find
floppies 1 & 3. Floppy #2 has been missing in action for many years. I have started to
go through boxes that were in storage. Hopefully I will get lucky but I?m not holding my
breath...
-Steve Davidson
SF:iP1
On Dec 21, 2019, at 09:07, Mark Matlock <mark at
matlockfamily.com> wrote:
?Zane,
DECnet/RT11 is one of the most significant pieces of lost PDP-11 software. I haven?t
been searching for it as long as you, but it is one of the few remaining things I?d like
to see the DEC software archeologists find. My most promising lead was that I heard of a
professor who taught computer networking ?back in the day? and he used two PDT 11/150s
connected with a serial port to demonstrate computer networking. Thus DECnet/RT11 was
small enough to boot on a dual RX02 system under RT-11 in 32KW Ram.
Phase III DECnet can be very small if DDCMP is the protocol. We used to run 11/23s
(128KW) with RSX11S that were downline loaded (no disk) their OS and they served 20 users
(1 DUV11 and 5 DZV11 or DLV11-J serial cards) with a commodity trading information system.
Of course that was re-entrant Macro-11code we wrote but still amazing in the day.
Best,
Mark
On Dec 20, 2019, at 1:28 PM, Zane Healy
<healyzh at avanthar.com> wrote:
On Dec 20, 2019, at 10:33 AM, Paul Koning
<paulkoning at comcast.net> wrote:
I can offer help with RSTS. Yes, there is DECnet/RT11, but I don't know if we have
any on HECnet or if the code has even been preserved.
paul
I?ve been trying to find a copy for well over 20 years. There may be a copy in a storage
container back east, but I?ve never heard further from the person.
Zane
6:22 p.m.
Steve,
Thanks so much for your post! Even getting part of the DECnet/RT-11 distribution is
helpful. Do keep us updated on your status with this. This is the closest to having it,
I?ve ever heard about!! Thanks and Good Luck finding RX50 #2 !!
Best,
Mark
On Dec 21, 2019, at 9:59 AM, Steve Davidson <steve
at davidson.net> wrote:
I?m the one who has the RX50?s that contain DECnet/RT-11. The problem is I only can find
floppies 1 & 3. Floppy #2 has been missing in action for many years. I have started to
go through boxes that were in storage. Hopefully I will get lucky but I?m not holding my
breath...
-Steve Davidson
SF:iP1
On Dec 21, 2019, at 09:07, Mark Matlock <mark
at matlockfamily.com> wrote:
?Zane,
DECnet/RT11 is one of the most significant pieces of lost PDP-11 software. I haven?t
been searching for it as long as you, but it is one of the few remaining things I?d like
to see the DEC software archeologists find. My most promising lead was that I heard of a
professor who taught computer networking ?back in the day? and he used two PDT 11/150s
connected with a serial port to demonstrate computer networking. Thus DECnet/RT11 was
small enough to boot on a dual RX02 system under RT-11 in 32KW Ram.
Phase III DECnet can be very small if DDCMP is the protocol. We used to run 11/23s
(128KW) with RSX11S that were downline loaded (no disk) their OS and they served 20 users
(1 DUV11 and 5 DZV11 or DLV11-J serial cards) with a commodity trading information system.
Of course that was re-entrant Macro-11code we wrote but still amazing in the day.
Best,
Mark
On Dec 20, 2019, at 1:28 PM, Zane Healy
<healyzh at avanthar.com> wrote:
On Dec 20, 2019, at 10:33 AM, Paul Koning
<paulkoning at comcast.net> wrote:
I can offer help with RSTS. Yes, there is DECnet/RT11, but I don't know if we have
any on HECnet or if the code has even been preserved.
paul
I?ve been trying to find a copy for well over 20 years. There may be a copy in a storage
container back east, but I?ve never heard further from the person.
Zane
7:26 p.m.
New subject: Tops-20 ANONYMOUS FAL Preliminary Testing Results/DAP Query
DECnet/RT-11 was phase III only. It required the XM monitor and, was a bit of a memory
hog for sure. It only supported serial line connections from what little I remember. In
1983 the RT-11 group acquired the sources from the DECnet group to update it to phase IV
but the project was cancelled almost immediately. But that's another story for
another time...
The kit I have is the RX50 binary distribution from the SDC. I got that kit before I
joined the RT-11 group, (I was working in Software Services at the time).
-Steve
On Dec 21, 2019, at 12:22, Mark Matlock <mark at
matlockfamily.com> wrote:
Steve,
Thanks so much for your post! Even getting part of the DECnet/RT-11 distribution is
helpful. Do keep us updated on your status with this. This is the closest to having it,
I?ve ever heard about!! Thanks and Good Luck finding RX50 #2 !!
Best,
Mark
On Dec 21, 2019, at 9:59 AM, Steve Davidson
<steve at davidson.net> wrote:
I?m the one who has the RX50?s that contain DECnet/RT-11. The problem is I only can find
floppies 1 & 3. Floppy #2 has been missing in action for many years. I have started to
go through boxes that were in storage. Hopefully I will get lucky but I?m not holding my
breath...
-Steve Davidson
SF:iP1
On Dec 21, 2019, at 09:07, Mark Matlock <mark
at matlockfamily.com> wrote:
?Zane,
DECnet/RT11 is one of the most significant pieces of lost PDP-11 software. I haven?t been
searching for it as long as you, but it is one of the few remaining things I?d like to see
the DEC software archeologists find. My most promising lead was that I heard of a
professor who taught computer networking ?back in the day? and he used two PDT 11/150s
connected with a serial port to demonstrate computer networking. Thus DECnet/RT11 was
small enough to boot on a dual RX02 system under RT-11 in 32KW Ram.
Phase III DECnet can be very small if DDCMP is the protocol. We used to run 11/23s
(128KW) with RSX11S that were downline loaded (no disk) their OS and they served 20 users
(1 DUV11 and 5 DZV11 or DLV11-J serial cards) with a commodity trading information system.
Of course that was re-entrant Macro-11code we wrote but still amazing in the day.
Best,
Mark
> On Dec 20, 2019, at 1:28 PM, Zane Healy
<healyzh at avanthar.com> wrote:
>
> On Dec 20, 2019, at 10:33 AM, Paul Koning <paulkoning at comcast.net> wrote:
>
> I can offer help with RSTS. Yes, there is DECnet/RT11, but I don't know if we
have any on HECnet or if the code has even been preserved.
>
> paul
I?ve been trying to find a copy for well over 20 years. There may be a copy in a storage
container back east, but I?ve never heard further from the person.
Zane
1:07 a.m.
Yes, Unix' ls does have its unique ideas...? I was forced to implement
some of them as certain brain-damaged FTP clients think that when are
you speaking to an FTP server which claims TVS compliance (I.E., Tops-20
Extended Mode FTP), that you must be Unix and be able to shell out.?
They literally issue the ls directly over the control channel instead of
the RFC959 compliant STAT or LIST verbs!? No where is this documented in
any RFC; you've got to snoop the line.? Grrr...
But we may be talking apples and oranges; there is a difference between
VMS requesting a listing of a locally attached file system (or even in a
cluster) and using DAP.? My guess is that DAP only allows one response
per command; that quite different from the previous case where you are
free to whine about each file.
Tops-20 directories are quite different animals than either Unix or
Windows, both in implementation and semantics.
I will follow up with you off list concerning tests with RSTS, perhaps
over the weekend when I'm finished figuring out what to do.
I have some very fond memories of the RSTS system at Columbia. It was
our first experiment in time sharing technology before we acquired 6
DECSYSTEM-20's.? The RSTS system was an 11/50 running RSTS/E and it was
a fabulous base for introductory programming. A number of us were sad to
see it go as opposed to being replaced with an 11/70, but we had to put
a 20 there.
------------------------------------------------------------------------
On 12/20/19 1:33 PM, Paul Koning wrote:
On Dec 20, 2019, at 1:12 PM, Thomas DeBellis
<tommytimesharing at gmail.com> wrote:
I have been working with a kind volunteer who is using OpenVMS Alpha. It's been
quite productive.
While I haven't run into any security exposures, yet, the experience has enabled me
to find a few things to tweak up. For example, the VMS logging and reporting was clearly
superior to what was being put in the FAL log (which wasn't even timestamped and
didn't have the requested file name).
We ran into some interesting behavior. When ANONYMOUS access is restricted to the
stand-in user and its subdirectories, wildcard listing requests that are outside of that
lexical scope do not report any errors but rather return zero files (none). That was
puzzling because the part of the code that did the access check was clearly doing the
reject with appropriate return values.
Not clear if that's an error. In RSTS
at least, if you asked for a list of files, you'd be given the subset of files that
you have read access to. Files for which you have no permission would appear nonexistent.
This is rather different from the behavior in Unix, which shows you all files in a brief
listing, and a pile of error messages in a long listing.
Whether VMS follows the same model I'm not sure. It's a bit more work in systems
like RSX or VMS, which like Unix separate directory names from the underlying file object
where the permissions live.
Further analysis showed that in the main loop for
stepping wildcards, a reject causes the file in question to be skipped; nothing is
actually done with the return values. I believe this may be intended behavior for two
reasons. The first is that if the wildcard includes files to which the ANONYMOUS user
does legitimately have access, it's friendlier to report those as opposed to stopping
everything.
The other reason I believe has to do with DAP itself; I don't think you can't
return more than one return value per ACCESS request, even with a wildcarded file
specification. So perhaps this isn't a bug so much as an unintended consequence?
Does anybody know what the behavior should be?
At any rate, I don't believe the current behavior is 'informative'. What
I'm thinking of doing is saving the last error code and returning that if the count of
files returned in list is zero. Otherwise, the user had something to look at.
Do I have any volunteers who operate something besides VMS? I'd be interested in
RSX(+), IAS, RSTS and OS-8. They never did DECnet on RT-11, did they?
I can offer
help with RSTS. Yes, there is DECnet/RT11, but I don't know if we have any on HECnet
or if the code has even been preserved.
paul 
1:27 a.m.
On 2019-12-21 01:07, Thomas DeBellis wrote:
Yes, Unix' ls does have its unique ideas...? I was
forced to implement
some of them as certain brain-damaged FTP clients think that when are
you speaking to an FTP server which claims TVS compliance (I.E., Tops-20
Extended Mode FTP), that you must be Unix and be able to shell out.
They literally issue the ls directly over the control channel instead of
the RFC959 compliant STAT or LIST verbs!? No where is this documented in
any RFC; you've got to snoop the line.? Grrr...
I've never seen an ftp client that actually gives an "ls" command. That
would be interesting. Could you point me to one?
Otherwise, the more annoying thing in my opinion is that some tools and
web browsers tries to parse the output from LIST, and really expects it
to be in a totally Unix-like format.
There is some extension to the ftp protocol that is about a common file
system representation (can't remember the exact terminology right now).
That was "fun" to implement in RSX...
I will follow up with you off list concerning tests
with RSTS, perhaps
over the weekend when I'm finished figuring out what to do.
You can ping me more abour RSX...
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
2:58 a.m.
1. The client that does the ls is gnuemacs ange-ftp mode; which was
quite shocking.
2. Trivial Virtual File System (TVFS) is the RFC that specifies a
common syntax, essentially codifying Unix syntax.? That's
unfortunate because when you blow away most of the punctuation, you
limit your search abilities.
"fun"?? Dear me...? Actually, what really drove me crazy was TENEX paged
file structures; I was months getting that right.
------------------------------------------------------------------------
On 12/20/19 7:27 PM, Johnny Billquist wrote:
------------------------------------------------------------------------
On 2019-12-21 01:07, Thomas DeBellis wrote:
Yes, Unix' ls does have its unique ideas...? I was forced to
implement some of them as certain brain-damaged FTP clients think
that when are you speaking to an FTP server which claims TVS
compliance (I.E., Tops-20 Extended Mode FTP), that you must be Unix
and be able to shell out.? They literally issue the ls directly over
the control channel instead of the RFC959 compliant STAT or LIST
verbs!? No where is this documented in any RFC; you've got to snoop
the line.? Grrr...
I've never seen an ftp client that actually gives an "ls" command.
That would be interesting. Could you point me to one?
Otherwise, the more annoying thing in my opinion is that some tools
and web browsers tries to parse the output from LIST, and really
expects it to be in a totally Unix-like format.
There is some extension to the ftp protocol that is about a common
file system representation (can't remember the exact terminology right
now).
That was "fun" to implement in RSX...
I will follow up with you off list concerning
tests with RSTS,
perhaps over the weekend when I'm finished figuring out what to do.
You can ping me more abour RSX...
? Johnny
11:02 a.m.
On 2019-12-21 02:58, Thomas DeBellis wrote:
1. The client that does the ls is gnuemacs ange-ftp
mode; which was
quite shocking.
Aha. Thanks. I have never tested that one. I always find even the mode
extremely annoying...
2. Trivial Virtual File System (TVFS) is the RFC that
specifies a
common syntax, essentially codifying Unix syntax.? That's
unfortunate because when you blow away most of the punctuation, you
limit your search abilities.
Thanks. Yes, TVFS was what I was trying to remember. Yes, it is
annoying, and some clients goes and dies immediately if they don't get
information in that silly format. I implemented it for RSX, and trigger
it when a client does a "cwd /", since most such clients also does that
first of all.
"fun"?? Dear me...? Actually, what really
drove me crazy was TENEX paged
file structures; I was months getting that right.
That sounds way too complicated. I don't think any specific feature or
detail in my RSX servers ever have taken more than a day or two to
implement. But I do feel that some things are silly.
But I should also admit that I've only implemented things I felt I
needed, so I've skipped some stuff which supposedly is mandatory. Noone
complained (yet).
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
22 Dec
22 Dec
7:53 p.m.
Yes, Ange-ftp is mode is a bit annoying; particularly since each new
version of Emacs seems to manage to break it in new and interesting
ways.? I'm still trying to fix an infinite recursion bug in the version
I have on the Mac.? Why use it?? Why *colors* (in other words, font-lock
mode) and memory.
As you may be aware, the EMACS under Tops-20 is TECO based, the TECO
being written in an assembler dialect called MIDAS which will emit code
for ITS, TENEX and Tops-20.? It is a pre-extended model, which means you
are 'limited' to a single section of 256K, which is something like 1.25
megabytes, depending on how you slice it.
If you are programming assembler, that is an awful lot of space before
you have to resort to overlays.? It's a decent amount of data.? The text
of whole novels will easily fit as (7 bit) ASCII; I have several from
Project Gutenberg.? Where you start to run out of memory is commonly
with multi-media and if you're me.?? It's not unusual for me to have 16
files open, flipping between them. On TECO based Emacs, you can run out
of memory.? Not so 64 bit gnuemacs; it just keeps going and going.? This
was a major driver for the FTP server rewrite.
The PDP-10 memory architecture had an enhancement done in the early 70's
to support extended memory, which provided an additional 4,095 sections
of virtual memory or about 5 GB total. In the 80's, that was known as
incomprehensible, 20 years ago, perfectly adequate and today, well...?
Unless you are talking HD video, 5 GB is reasonable for many
applications, including editing NTSC and PAL video.
At the time, a colleague and I were considering rewriting TECO in MACRO
to gain the advantage of that memory paradigm. Unfortunately, the
university switched off of the (cancelled) 36 bit line to Unix.? First
Digital Unix on a VAX 8650, the 8700 and then Sun SPARC machines.? They
never trusted DEC again.
gnuemacs has been ported to Tops-20, so there is even less reason to do
it now, but I didn't know that ten years ago when I started as a hobbyist.
Another reason was that I wasn't getting the performance I wanted out of
the TENEX based FTP server.? I worked with it for about 6 months until I
was advised that it would be better to just pitch it.? The new server
easily gets over 6.5 Mbps, whereas the old one could barely get into the
high KB/s.? I eventually figured out why, but I don't think the old one
could be made this fast.
The final reason was that I just kept getting annoyed that 'modern'
graphical FTP clients had no idea of what the 20 was returning.? And if
you know me, when it comes to computers, it doesn't take much to get me
really annoyed.? I couldn't believe the junk I was getting over the
control channel.? Lots of commands that are not in any RFC.? Looking at
other FTP servers didn't help either; they were hopelessly broken.
But now I can fool all but the very most brain damaged graphical FTP
clients (unfortunately I happen to have one on my iPhone). Character
mode FTP clients were easy.
I'm actually thinking of implementing FTP over DECnet; the performance
(in terms of processor utilization) would easily exceed DAP.? And it
would be my kind of fun.
------------------------------------------------------------------------
On 12/21/19 5:02 AM, Johnny Billquist wrote:
Aha. Thanks. I have never tested that one. I always find even the mode
extremely annoying...
Thanks. Yes, TVFS was what I was trying to remember. Yes, it is
annoying, and some clients goes and dies immediately if they don't get
information in that silly format. I implemented it for RSX, and
trigger it when a client does a "cwd /", since most such clients also
does that first of all.
That sounds way too complicated. I don't think any specific feature or
detail in my RSX servers ever have taken more than a day or two to
implement. But I do feel that some things are silly.
But I should also admit that I've only implemented things I felt I
needed, so I've skipped some stuff which supposedly is mandatory.
Noone complained (yet).
? Johnny
> ------------------------------------------------------------------------
> On 2019-12-21 02:58, Thomas DeBellis wrote:
>
> 1. ?The client that does the ls is gnuemacs ange-ftp mode; which was
> quite shocking.
> 2. ?Trivial Virtual File System (TVFS) is the RFC that specifies a
> common syntax, essentially codifying Unix syntax.? That's
> unfortunate because when you blow away most of the punctuation,
> you limit your search abilities.
>
> "fun"?? Dear me...? Actually, what really drove me crazy was TENEX
> paged file structures; I was months getting that right.
20 Dec
20 Dec
9:46 p.m.
New subject: Tops-20 ANONYMOUS FAL Preliminary Testing Results/DAP Query
-----Original Message-----
Do I have any volunteers who operate something besides VMS? I'd be
interested in RSX(+), IAS, RSTS and OS-8. They never did DECnet on
RT-11, did they?
If you run across anyone with DECET-IAS, please let me know...
--
Lee K. Gleason N5ZMR
Control-G Consultants
lee.gleason at comcast.net
11 p.m.
On Dec 20, 2019, at 3:46 PM, Lee Gleason
<lee.gleason at comcast.net> wrote:
Did that exist? I'm fairly sure it did. What about DECnet for RSX-11/D? Not so sure
about that one.
paul
-----Original Message----- Do I have any
volunteers who operate something besides VMS? I'd be interested in RSX(+), IAS, RSTS
and OS-8. They never did DECnet on RT-11, did they?
If you run across anyone with DECET-IAS, please let me know...
11:44 p.m.
On 2019-12-20 23:00, Paul Koning wrote:
Both existed. See the spd at http://www.bitsavers.org/pdf/dec/spd/
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
On Dec 20, 2019, at 3:46 PM, Lee Gleason
<lee.gleason at comcast.net> wrote:
Did that exist? I'm fairly sure it did. What about DECnet for RSX-11/D? Not so
sure about that one. -----Original Message----- Do I have any
volunteers who operate something besides VMS? I'd be interested in RSX(+), IAS, RSTS
and OS-8. They never did DECnet on RT-11, did they?
If you run across anyone with DECET-IAS, please let me know... 
21 Dec
21 Dec
1:07 a.m.
On 12/20/2019 5:44 PM, Johnny Billquist wrote:
On 2019-12-20 23:00, Paul Koning wrote:
Both existed. See the spd at http://www.bitsavers.org/pdf/dec/spd/
? Johnny
Nice link! Tell me if one OS was going to be learned, I would like
anyone out there who knows, what would it be? RSX RT DECnet VMS or what?
Taking into consideration, easiness to learn because it would take some
time. And Availability or modernized tools like VMS has of example, TCP/IP.
Bill
On Dec 20, 2019, at 3:46 PM, Lee Gleason
<lee.gleason at comcast.net>
wrote:
Did that exist?? I'm fairly sure it did.? What about DECnet for
RSX-11/D?? Not so sure about that one. -----Original Message----- Do I have any
volunteers who operate
something besides VMS?? I'd be interested in RSX(+), IAS, RSTS and
OS-8.? They never did DECnet on RT-11, did they?
If you run across anyone with DECET-IAS, please let me know...
1:18 a.m.
On 2019-12-21 01:07, Bill Cunningham wrote:
On 12/20/2019 5:44 PM, Johnny Billquist wrote:
DECnet is not an operating system. And VMS is obviously not for the PDP-11.
If you want both DECnet and TCP/IP, then the choice is essentially only
RSX-11M-PLUS.
But apart from that, it depends on what you prefer. RSTE/E is more pure
timesharing, RT-11 is more pure realtime and minimalistic, RSX-11D and
IAS are sortof timesharing with realtime with more of a focus on
timesharing and large systems. RSX-11M is reatime with timesharing, and
a little primitive in some corners, while RSX-11M-PLUS is essentially
RSX-11M with lots of extra bells and whistles added, so still realtime
with timesharing, but friendlier, more efficient, and more capable.
In the end I would probably boil it down to either RSTS/E or
RSX-11M-PLUS, depending on preferences and what tools you want.
For most parts, there are more and more capable tools for RSX-11M-PLUS,
but RSTS/E is probably still a bit friendlier.
If you know VMS, then in many ways RSTS/E will be rather familiar at a
user level, but RSX will be more familiar if you start digging under the
hood.
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
On 2019-12-20 23:00, Paul Koning wrote:
Both existed. See the spd at http://www.bitsavers.org/pdf/dec/spd/
? Johnny
Nice link! Tell me if one OS was going to be learned, I would like
anyone out there who knows, what would it be? RSX RT DECnet VMS or what?
Taking into consideration, easiness to learn because it would take some
time. And Availability or modernized tools like VMS has of example, TCP/IP. On Dec 20, 2019, at 3:46 PM, Lee Gleason
<lee.gleason at comcast.net>
wrote:
> -----Original Message----- Do I have any volunteers who operate
> something besides VMS?? I'd be interested in RSX(+), IAS, RSTS and
> OS-8.? They never did DECnet on RT-11, did they?
If you run across anyone with DECET-IAS, please let me know...
Did that exist?? I'm fairly sure it did.? What about DECnet for
RSX-11/D?? Not so sure about that one.
1:26 a.m.
On 12/20/2019 7:18 PM, Johnny Billquist wrote:
DECnet is not an operating system. And VMS is obviously not for the
PDP-11.
If you want both DECnet and TCP/IP, then the choice is essentially
only RSX-11M-PLUS.
But apart from that, it depends on what you prefer. RSTE/E is more
pure timesharing, RT-11 is more pure realtime and minimalistic,
RSX-11D and IAS are sortof timesharing with realtime with more of a
focus on timesharing and large systems. RSX-11M is reatime with
timesharing, and a little primitive in some corners, while
RSX-11M-PLUS is essentially RSX-11M with lots of extra bells and
whistles added, so still realtime with timesharing, but friendlier,
more efficient, and more capable.
In the end I would probably boil it down to either RSTS/E or
RSX-11M-PLUS, depending on preferences and what tools you want.
For most parts, there are more and more capable tools for
RSX-11M-PLUS, but RSTS/E is probably still a bit friendlier.
If you know VMS, then in many ways RSTS/E will be rather familiar at a
user level, but RSX will be more familiar if you start digging under
the hood.
? Johnny
Thank you very much for your knowledgeable opinion Johnny.
Bill
2:24 a.m.
On Dec 20, 2019, at 7:18 PM, Johnny Billquist <bqt
at softjar.se> wrote:
...
For most parts, there are more and more capable tools for RSX-11M-PLUS, but RSTS/E is
probably still a bit friendlier.
If you know VMS, then in many ways RSTS/E will be rather familiar at a user level, but
RSX will be more familiar if you start digging under the hood.
Yes, RSTS/E will look familiar if you're using V9 or later. That's when DCL
became a serious capability. It started in V8.0, but then it was still quite limited.
Before then it looks quite different from VMS.
paul
2:51 a.m.
That is an extremely difficult question to answer; they all had their
quirks and advantages, which is not surprising given that they targeted
very different markets.? As a whole, I would say that the quality of
DEC's documentation was the same on all the platforms; I found it easier
to wade through than IBM's VM/CMS docs.
In terms of raw capacity, probably the later mainframe operating systems
would be the best choice as these most approximate what is available
today on commercial platforms (with the clear exception of the graphics
interface).? What matters is the address space and functionality:
1. VMS started out on a 32 bit platform as a high super-mini computer.?
DEC soon grew this into low mainframe class with the VAX 8650
series, the Alpha being a completely modern platform. VMS is an
acquired taste; some people think DCL is great and others...
2. Tops-10 started as a mainframe platform and was eventually rewritten
to have symmetric multi-processing.? The command interface is simple
enough.? The process paradigm is limited in that there is a
one-to-one mapping of one process per job.? The address space had
been limited to 256K, but I think they may have done Extended
Addressing.? I believe TCP/IP may have been available from a 3rd
party vendor.
3. Tops-20 is more modern, having a multi-process per job architecture
and a 30 bit address space.? Many Unix programs are easily ported.?
It ran the most networking protocols, including TCP/IP.? It is quite
easy to learn due to the COMND% interface, which provides
completions, guiding, prompting and recognition. There was a time
when there were more of these systems on the ARPAnet then Unix.? I
have been looking to write an IP6 interface after I complete some
more items on my To Do list.
But the mini-computer operating systems are just plain cool.? It is
amazing what they squeezed into the PDP-8's 12 bit address space and
PDP-11's run some of the most interesting collection of OS's that I've
ever seen.
It really all depends on what you feel like playing with; it's a
profitable learning experience all around.
------------------------------------------------------------------------
On 12/20/19 7:07 PM, Bill Cunningham wrote:
Nice link! Tell me if one OS was going to be learned, I would like
anyone out there who knows, what would it be? RSX RT DECnet VMS or
what? Taking into consideration, easiness to learn because it would
take some time. And Availability or modernized tools like VMS has of
example, TCP/IP.
Bill
------------------------------------------------------------------------
On 12/20/2019 5:44 PM, Johnny Billquist wrote:
Both existed. See the spd at
http://www.bitsavers.org/pdf/dec/spd/
? Johnny
------------------------------------------------------------------------
On 2019-12-20 23:00, Paul Koning wrote:
------------------------------------------------------------------------
On Dec 20, 2019, at 3:46 PM, Lee Gleason <lee.gleason at comcast.net>
wrote:
> -----Original Message----- Do I have any volunteers who operate
> something besides VMS?? I'd be interested in RSX(+), IAS, RSTS and
> OS-8.? They never did DECnet on RT-11, did they?
If you run across anyone with DECET-IAS, please let me know...
Did that exist??
I'm fairly sure it did.? What about DECnet for
RSX-11/D?? Not so sure about that one.
3:08 a.m.
On Dec 20, 2019, at 8:51 PM, Thomas DeBellis
<tommytimesharing at gmail.com> wrote:
...
But the mini-computer operating systems are just plain cool. It is amazing what they
squeezed into the PDP-8's 12 bit address space and PDP-11's run some of the most
interesting collection of OS's that I've ever seen.
If you go far enough back, the space efficiency gets pretty amazing. There's RSTS-11,
which ran 16 timesharing users on a 28kW PDP-11/20. (Not well, but it ran.) Or RT-11,
quite comfortable in 8 kW and a 256 kbyte system disk. Or DOS-11, which would even run, I
think, in 4 kW.
Somewhat earlier still, in 1961 two people implemented the first ever ALGOL compiler in 6
months, and it ran on a 4 kW machine (27 bit). (It's known to have two bugs.)
paul
5:21 a.m.
Indeed; but there was a time when the PDP-11's 16 bit (64K) address
space and eight register file seemed positively generous. That's when
you compare it to the PDP-8's /single/ accumulator and 12 bit (4K)
address space.
I continue to be astounded what they managed to do with that.? In
addition to a nice package of languages, with a memory management unit
(essentially a bank switcher), they got the thing to timeshare.? That's
right; TSS-8.? There's one still running at the Computer History Museum
on an 8/I.
So they got DECnet running on the PDP-8?? Wow.? I wonder how they did
that; whether they re-targeted a BLISS compiler to emit PAL.? I remember
looking at the source to PDP-8 VT (video) TECO.? Many awe most inspiring
kludges.? What a tour de force. Very humbling.
The Algol compiler on the 20 has more than 2 bugs...? Sigh...
------------------------------------------------------------------------
On 12/20/19 9:08 PM, Paul Koning wrote
If you go far enough back, the space efficiency gets pretty amazing. There's
RSTS-11, which ran 16 timesharing users on a 28kW PDP-11/20. (Not well, but it ran.) Or
RT-11, quite comfortable in 8 kW and a 256 kbyte system disk. Or DOS-11, which would even
run, I think, in 4 kW.
Somewhat earlier still, in 1961 two people implemented the first ever ALGOL compiler in 6
months, and it ran on a 4 kW machine (27 bit). (It's known to have two bugs.)
paul
> ------------------------------------------------------------------------
> On Dec 20, 2019, at 8:51 PM, Thomas DeBellis <tommytimesharing at gmail.com>
wrote:
> ...
> But the mini-computer operating systems are just plain cool. It is amazing what they
squeezed into the PDP-8's 12 bit address space and PDP-11's run some of the most
interesting collection of OS's that I've ever seen.
11:22 a.m.
On 2019-12-21 05:21, Thomas DeBellis wrote:
Indeed; but there was a time when the PDP-11's 16
bit (64K) address
space and eight register file seemed positively generous. That's when
you compare it to the PDP-8's /single/ accumulator and 12 bit (4K)
address space.
Yeah. Coming from a PDP-8, the PDP-11 is like a whole different world.
I continue to be astounded what they managed to do
with that.? In
addition to a nice package of languages, with a memory management unit
(essentially a bank switcher), they got the thing to timeshare.? That's
right; TSS-8.? There's one still running at the Computer History Museum
on an 8/I.
There are actually multiple timesharing systems for the PDP-8. I think
Multos is way cooler than TSS-8. With Multos you get a bunch of users,
each running their own instance of OS/8, everyone with 32K of virtual
memory, and then a bunch of additional functions provided by the OS for
things like communication between users and programs, getting various
system information, spooling and what not. All with a standard omnibus
PDP-8. No extra hardware required. And it then obviously goes beyond
bank-switching. You actually have virtual memory, memory protection, it
pages memory in and out as needed, and actually reduce the memory needed
in individual users memory space, since all device drivers are already
resident, but not in your address space. Not to mention that it can even
make some operations be more efficient, like I/O actually being properly
interrupt driven.
And of course, then you have things like ETOS-8, which requires special
hardware, but also gives timesharing.
The amount of stuff possible on the PDP-8 is nothing short of amazing.
So they got DECnet running on the PDP-8?? Wow.? I
wonder how they did
that; whether they re-targeted a BLISS compiler to emit PAL.? I remember
looking at the source to PDP-8 VT (video) TECO.? Many awe most inspiring
kludges.? What a tour de force. Very humbling.
No BLISS for the PDP-8. DECnet-8 is written in PAL-8. And it's a set of
services and a task or two running under RTS-8. RTS-8 is a realtime OS.
Not too different in many concepts to RT-11 or RSX. But it's statically
built. So you have all the tasks you might want to run compiled into one
big binary along with the RTS-8 kernel (and DECnet-8 if you want it),
and then you run that binary from OS/8.
OS/8 can also exist as a background task of RTS-8, so you can sortof
continue doing development work while RTS-8 is running.
TECO-8 is very impressive. I wrote an Emacs clone in it which was rather
decent and very usable. I believe there is a comment in there that if
someone can just shake out a couple of more words, reverse searching
could be implemented. That tells you how tight everything is. :-)
The Algol compiler on the 20 has more than 2 bugs...?
Sigh...
I know I've seen some Algol thingy for the PDP-8 as well, but I've never
tested 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
22 Dec
22 Dec
7:01 p.m.
/A bit of things snipped./
I don't believe I've ever heard of Multos; I was using OS/8 and TSS-8 in
the early 70's before I got access to a PDP-10, which was felt to have a
more powerful and interesting instruction set.? It certainly floored
some of my PDP-8 friends.
Actually, now that I'm thinking about it, the total number of operating
systems across the board was amazing.? For the PDP-10, there was
Tops-10, ITS, Tops-20 (TENEX) and WAITS (off the top of my head).? But
the minicomputers were something else altogether. Wow.? I help managed
the CS lab at my undergraduate college, where we had a PDP-11/05 and an
11/45 (with a GT40).? It seemed to me that every time I turned around, I
was tripping over another operating system for one of them.
We had an RK05 with an early version of Unix (circa 1977).? It was
looked at and summarily dismissed as "Interesting; needs to get
finished".? We stuck with RT11 and RSX because they had more
functionality, were faster and used less resources.? Little did we
know...? Now, unless you are on an IBM mainframe, it seems you pretty
much get your choice of Windows or Unix.? I've programmed iOS (under
Swift) and deep down, well, you know.?? Eye popping graphics, though.?
We never had that.? And pretty much infinite memory.
None the less, there can be a lot of great ideas that come up when
you're doing something something from scratch and this is particularly
evident in the realm of file system syntax.? I continue to believe that
DEC:'s grammar and built-in parsing is superior to anything you can get
on Unix, where all of that is tossed out the window and you get "/".?
Windows, which will allow you to specify a device (but not wildcard it)
is only marginally better.
I'm reminded of a talk that Rob Pike gave at the turn of the millennium,
"Systems Software Research is Irrelevant".? I don't agree with
everything he says because I do follow some of the research, but still,
it's grim reading.? What I've found is that you can get some good ideas
by seeing what other people where doing whose backs were up against the
wall and what they thought they'd never be able to do (based on the
hardware restrictions).
That's surprising about no BLISS on the PDP-8.? On my first day at DEC
as a Tops-20 programmer, we got stuck into a BLISS COMMON class and
received the corporate indoctrination.?? Then we went on to design the
File Finder project in MACRO.? BLISS did get used, particularly when
something got tossed at us from another platform.?? But I can't remember
what major package was written in it.? Algol wasn't.? There is a new
part of Galaxy that I have yet to fathom called "FTS" that is written in
BLISS.? I don't understand the EXEC interface to it; perhaps there is a
monitor interface.
Has anybody heard of it?
------------------------------------------------------------------------
On 12/21/19 5:22 AM, Johnny Billquist wrote:
There are actually multiple timesharing systems for the PDP-8. I think
Multos is way cooler than TSS-8.
The amount of stuff possible on the PDP-8 is nothing short of amazing.
No BLISS for the PDP-8. DECnet-8 is written in PAL-8.
> ------------------------------------------------------------------------
> On 2019-12-21 05:21, Thomas DeBellis wrote:
>
> So they got DECnet running on the PDP-8?? Wow.
23 Dec
23 Dec
8:24 p.m.
On Dec 20, 2019, at 11:21 PM, Thomas DeBellis
<tommytimesharing at gmail.com> wrote:
Indeed; but there was a time when the PDP-11's 16 bit (64K) address space and eight
register file seemed positively generous. That's when you compare it to the
PDP-8's single accumulator and 12 bit (4K) address space.
I continue to be astounded what they managed to do with that. In addition to a nice
package of languages, with a memory management unit (essentially a bank switcher), they
got the thing to timeshare. That's right; TSS-8. There's one still running at
the Computer History Museum on an 8/I.
So they got DECnet running on the PDP-8? Wow. I wonder how they did that; whether they
re-targeted a BLISS compiler to emit PAL. I remember looking at the source to PDP-8 VT
(video) TECO. Many awe most inspiring kludges. What a tour de force. Very humbling.
That was implemented on RTS-8 and looks like a Phase I implementation - all
hand-crafted PAL code. The floppies are available on the net and includes full source
code. When I joined the DECnet development group in early 1977, there were a couple of
PDP-8 developers as part of the group. I don?t know if they were developing a Phase II
implementation but they disappeared after about 6 months - not surprising given the
difficulties we were having getting it to fit in a 28KW PDP-11.
John.
The Algol compiler on the 20 has more than 2 bugs...
Sigh...
>
> On 12/20/19 9:08 PM, Paul Koning wrote
> If you go far enough back, the space efficiency gets pretty amazing. There's
RSTS-11, which ran 16 timesharing users on a 28kW PDP-11/20. (Not well, but it ran.) Or
RT-11, quite comfortable in 8 kW and a 256 kbyte system disk. Or DOS-11, which would even
run, I think, in 4 kW.
>
> Somewhat earlier still, in 1961 two people implemented the first ever ALGOL compiler
in 6 months, and it ran on a 4 kW machine (27 bit). (It's known to have two bugs.)
>
> paul
>>
>> On Dec 20, 2019, at 8:51 PM, Thomas DeBellis <tommytimesharing at
gmail.com> <mailto:tommytimesharing at gmail.com> wrote:
>> ...
>> But the mini-computer operating systems are just plain cool. It is amazing what
they squeezed into the PDP-8's 12 bit address space and PDP-11's run some of the
most interesting collection of OS's that I've ever seen.
11:51 p.m.
On 2019-12-23 20:24, John Forecast wrote:
Poul took a look at the DECNET-8 sources, and figured it was actually
not phase I (I thought it was phase I as well). Poul thinks it's close
to, if not actually phase II. There are things in there that apparently
did not exist in phase I.
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
On Dec 20, 2019, at 11:21 PM, Thomas DeBellis
<tommytimesharing at gmail.com <mailto:tommytimesharing at gmail.com>>
wrote:
Indeed; but there was a time when the PDP-11's 16 bit (64K) address
space and eight register file seemed positively generous. That's when
you compare it to the PDP-8's /single/ accumulator and 12 bit (4K)
address space.
I continue to be astounded what they managed to do with that.? In
addition to a nice package of languages, with a memory management unit
(essentially a bank switcher), they got the thing to timeshare.
That's right; TSS-8.? There's one still running at the Computer
History Museum on an 8/I.
So they got DECnet running on the PDP-8?? Wow.? I wonder how they did
that; whether they re-targeted a BLISS compiler to emit PAL.? I
remember looking at the source to PDP-8 VT (video) TECO.? Many awe
most inspiring kludges.? What a tour de force. Very humbling.
That was implemented on RTS-8 and looks like a Phase I implementation -
all hand-crafted PAL code. The floppies are available on the net and
includes full source code. When I joined the DECnet development group in
early 1977, there were a couple of PDP-8 developers as part of the
group. I don?t know if they were developing a Phase II implementation
but they disappeared after about 6 months - not surprising given the
difficulties we were having getting it to fit in a 28KW PDP-11.
24 Dec
24 Dec
5:59 p.m.
On Dec 23, 2019, at 5:51 PM, Johnny Billquist <bqt
at softjar.se> wrote:
On 2019-12-23 20:24, John Forecast wrote:
Poul took a look at the DECNET-8 sources, and figured it was actually not phase I (I
thought it was phase I as well). Poul thinks it's close to, if not actually phase II.
There are things in there that apparently did not exist in phase I.
Yes, I agree. I just took a look at the code and the early dates match when Phase
I was available (I was using it around the middle of 1976) and the later dates (mid-1977)
would have been when we were finalizing the DECnet-RSX Phase II system architecture. The
code claims to support NSP version 2.2 but I have no way to map that to a particular
phase, only that version 1.0 was dated July 1975 and version 3.1 (Phase III) was dated
March 1978.
John.
On Dec
20, 2019, at 11:21 PM, Thomas DeBellis <tommytimesharing at gmail.com
<mailto:tommytimesharing at gmail.com>> wrote:
Indeed; but there was a time when the PDP-11's 16 bit (64K) address space and eight
register file seemed positively generous. That's when you compare it to the
PDP-8's /single/ accumulator and 12 bit (4K) address space.
I continue to be astounded what they managed to do with that. In addition to a nice
package of languages, with a memory management unit (essentially a bank switcher), they
got the thing to timeshare. That's right; TSS-8. There's one still running at
the Computer History Museum on an 8/I.
So they got DECnet running on the PDP-8? Wow. I wonder how they did that; whether they
re-targeted a BLISS compiler to emit PAL. I remember looking at the source to PDP-8 VT
(video) TECO. Many awe most inspiring kludges. What a tour de force. Very humbling.
That was implemented on RTS-8 and looks like a Phase I implementation - all
hand-crafted PAL code. The floppies are available on the net and includes full source
code. When I joined the DECnet development group in early 1977, there were a couple of
PDP-8 developers as part of the group. I don?t know if they were developing a Phase II
implementation but they disappeared after about 6 months - not surprising given the
difficulties we were having getting it to fit in a 28KW PDP-11. 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
25 Dec
25 Dec
2:56 a.m.
On Dec 24, 2019, at 11:59 AM, John Forecast <john
at forecast.name> wrote:
A while ago I found a Phase I RTS-8 document that describes the protocol in reasonable
detail. It's clear that Phase I has only a vague resemblance to the later protocols.
For example, the NSP protocol is there, but it's seriously different. For example, it
offers both a connection service and a datagram service. And various other things work
entirely differently than the later versions. So while Phase III and IV can talk to Phase
II NSP with no real trouble, there is no way for any of them to make sense out of what a
Phase I node is saying.
If you have source code, it should be pretty obvious if it's Phase I or Phase II from
looking at the NSP packet formats.
paul
On Dec 23, 2019, at 5:51 PM, Johnny Billquist
<bqt at softjar.se> wrote:
On 2019-12-23 20:24, John Forecast wrote:
Poul took a look at the DECNET-8 sources, and figured it was actually not phase I (I
thought it was phase I as well). Poul thinks it's close to, if not actually phase II.
There are things in there that apparently did not exist in phase I.
Yes, I agree. I just took a look at the code and the early dates match when Phase
I was available (I was using it around the middle of 1976) and the later dates (mid-1977)
would have been when we were finalizing the DECnet-RSX Phase II system architecture. The
code claims to support NSP version 2.2 but I have no way to map that to a particular
phase, only that version 1.0 was dated July 1975 and version 3.1 (Phase III) was dated
March 1978. ...
That was implemented on RTS-8 and looks like a Phase I implementation - all
hand-crafted PAL code. The floppies are available on the net and includes full source
code. When I joined the DECnet development group in early 1977, there were a couple of
PDP-8 developers as part of the group. I don?t know if they were developing a Phase II
implementation but they disappeared after about 6 months - not surprising given the
difficulties we were having getting it to fit in a 28KW PDP-11.
4:11 a.m.
Paul, sorry for mis-spelling your name.
You looked at the DECNET-8 code before. I sent you either links, files,
or some reference to a machine where you could check it. I can't
remember now. But anyway, you looked, and figured it was beyond phase I.
Johnny
On 2019-12-25 02:56, Paul Koning wrote:
--
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
On Dec 24, 2019, at 11:59 AM, John Forecast
<john at forecast.name
<mailto:john at forecast.name>> wrote:
A while ago I found a Phase I RTS-8 document that describes the protocol
in reasonable detail. ?It's clear that Phase I has only a vague
resemblance to the later protocols. ?For example, the NSP protocol is
there, but it's seriously different. ?For example, it offers both a
connection service and a datagram service. ?And various other things
work entirely differently than the later versions. ?So while Phase III
and IV can talk to Phase II NSP with no real trouble, there is no way
for any of them to make sense out of what a Phase I node is saying.
If you have source code, it should be pretty obvious if it's Phase I or
Phase II from looking at the NSP packet formats.
paul
On Dec 23, 2019, at 5:51 PM, Johnny Billquist
<bqt at softjar.se
<mailto:bqt at softjar.se>> wrote:
On 2019-12-23 20:24, John Forecast wrote:
Yes, I agree. I just took a look at the code and the early dates match
when Phase I was available (I was using it around the middle of 1976)
and the later dates (mid-1977) would have been when we were finalizing
the DECnet-RSX Phase II system architecture. The code claims to
support NSP version 2.2 but I have no way to map that to a particular
phase, only that version 1.0 was dated July 1975 and version 3.1
(Phase III) was dated March 1978. > ...
That was implemented on RTS-8 and looks like a Phase I
implementation - all hand-crafted PAL code. The floppies are
available on the net and includes full source code. When I joined
the DECnet development group in early 1977, there were a couple of
PDP-8 developers as part of the group. I don?t know if they were
developing a Phase II implementation but they disappeared after
about 6 months - not surprising given the difficulties we were
having getting it to fit in a 28KW PDP-11.
Poul took a look at the DECNET-8 sources, and figured it was actually
not phase I (I thought it was phase I as well). Poul thinks it's
close to, if not actually phase II. There are things in there that
apparently did not exist in phase I.
6:43 p.m.
On Dec 24, 2019, at 8:56 PM, Paul Koning
<paulkoning at comcast.net> wrote:
I?ve never seen a Phase I protocol spec but DECnet/8 was probably closer to Phase
I than Phase II although the dates in the source code appear to cover both phases. Even
after Phase II development started the protocols were still being updated - datagram
service was still in the NSP spec in early 1977 but was dropped about half way through the
development cycle. One advantage we had for DECnet-RSX development (11M/11S/11D and IAS)
was that they all shared common protocol processing code, which had been written for an
earlier advanced development project, so protocol changes were relatively easy to make.
John.
On Dec 24, 2019, at 11:59 AM, John Forecast
<john at forecast.name <mailto:john at forecast.name>> wrote:
A while ago I found a Phase I RTS-8 document that describes the protocol in reasonable
detail. It's clear that Phase I has only a vague resemblance to the later protocols.
For example, the NSP protocol is there, but it's seriously different. For example, it
offers both a connection service and a datagram service. And various other things work
entirely differently than the later versions. So while Phase III and IV can talk to Phase
II NSP with no real trouble, there is no way for any of them to make sense out of what a
Phase I node is saying.
If you have source code, it should be pretty obvious if it's Phase I or Phase II from
looking at the NSP packet formats.
On Dec 23, 2019, at 5:51 PM, Johnny Billquist
<bqt at softjar.se <mailto:bqt at softjar.se>> wrote:
On 2019-12-23 20:24, John Forecast wrote:
Yes, I agree. I just took a look at the code and the early dates match when Phase
I was available (I was using it around the middle of 1976) and the later dates (mid-1977)
would have been when we were finalizing the DECnet-RSX Phase II system architecture. The
code claims to support NSP version 2.2 but I have no way to map that to a particular
phase, only that version 1.0 was dated July 1975 and version 3.1 (Phase III) was dated
March 1978. > ...
That was implemented on RTS-8 and looks like a Phase I implementation - all hand-crafted
PAL code. The floppies are available on the net and includes full source code. When I
joined the DECnet development group in early 1977, there were a couple of PDP-8 developers
as part of the group. I don?t know if they were developing a Phase II implementation but
they disappeared after about 6 months - not surprising given the difficulties we were
having getting it to fit in a 28KW PDP-11.
Poul took a look at the DECNET-8 sources, and figured it was actually not phase I (I
thought it was phase I as well). Poul thinks it's close to, if not actually phase II.
There are things in there that apparently did not exist in phase I.
11:44 p.m.
What language did you implement in?? I've got a number of sources in
BLISS, some of which don't appear to have originated on the 36 bit
platform; perhaps some version of DAP for FTS.
How did it work between you RSX folks and the people doing whatever was
in the DN20's?? The 20 host got up to being a Phase IV router, but I
don't believe the DN20's went past Phase III.
------------------------------------------------------------------------
On 12/25/2019 12:43 PM, John Forecast wrote:
I?ve never seen a Phase I protocol spec but DECnet/8 was probably
closer to Phase I than Phase II although the dates in the source code
appear to cover both phases. Even after Phase II development started
the protocols were still being updated - datagram service was still in
the NSP spec in early 1977 but was dropped about half way through the
development cycle. One advantage we had for DECnet-RSX development
(11M/11S/11D and IAS) was that they all shared common protocol
processing code, which had been written for an earlier advanced
development project, so protocol changes were relatively easy to make.
? ?John.
26 Dec
26 Dec
12:11 a.m.
I can at least say that most of DECnet in RSX is written in Macro-11.
The only things written in Bliss, that I have found is PHONE, and the
P/OS network management code.
PHONE is shared with RSX-11M-PLUS, but the P/OS Network Management is
P/OS only. RSX have a different implementation.
Johnny
On 2019-12-25 23:44, Thomas DeBellis wrote:
What language did you implement in?? I've got a
number of sources in
BLISS, some of which don't appear to have originated on the 36 bit
platform; perhaps some version of DAP for FTS.
How did it work between you RSX folks and the people doing whatever was
in the DN20's?? The 20 host got up to being a Phase IV router, but I
don't believe the DN20's went past Phase III.
> ------------------------------------------------------------------------
> On 12/25/2019 12:43 PM, John Forecast wrote:
>
> I?ve never seen a Phase I protocol spec but DECnet/8 was probably
> closer to Phase I than Phase II although the dates in the source code
> appear to cover both phases. Even after Phase II development started
> the protocols were still being updated - datagram service was still in
> the NSP spec in early 1977 but was dropped about half way through the
> development cycle. One advantage we had for DECnet-RSX development
> (11M/11S/11D and IAS) was that they all shared common protocol
> processing code, which had been written for an earlier advanced
> development project, so protocol changes were relatively easy to make.
>
> ? ?John.
--
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
5:16 p.m.
On Dec 25, 2019, at 5:44 PM, Thomas DeBellis
<tommytimesharing at gmail.com> wrote:
What language did you implement in? I've got a number of sources in BLISS, some of
which don't appear to have originated on the 36 bit platform; perhaps some version of
DAP for FTS.
As Johnny pointed out, pretty much everything was written in Macro-11. PHONE was
not midnight project during Phase III IIRC.
How did it work between you RSX folks and the people
doing whatever was in the DN20's? The 20 host got up to being a Phase IV router, but
I don't believe the DN20's went past Phase III.
Are you talking about MCB? Sometime during Phase III development, we were asked to produce
a source kit for LCG. That was pretty much the only interaction we had other than doing
certification testing before a release.
John.
> On 12/25/2019 12:43 PM, John Forecast wrote:
>
> I?ve never seen a Phase I protocol spec but DECnet/8 was probably closer to Phase I
than Phase II although the dates in the source code appear to cover both phases. Even
after Phase II development started the protocols were still being updated - datagram
service was still in the NSP spec in early 1977 but was dropped about half way through the
development cycle. One advantage we had for DECnet-RSX development (11M/11S/11D and IAS)
was that they all shared common protocol processing code, which had been written for an
earlier advanced development project, so protocol changes were relatively easy to make.
>
> John.
5:47 p.m.
New subject: Tops-20 ANONYMOUS FAL Preliminary Testing Results/DAP Query
VMS phone was a hack by Paul Agnastopolous (sp?) from the VMS mail sources. The other
implementations came later. I remember talking to Paul about it. He was very insistent
that it was truly a HACK!
-Steve Davidson
SF:iP1
On Dec 26, 2019, at 11:18, John Forecast <john at
forecast.name> wrote:
?
Are you talking about MCB? Sometime during Phase III development, we were asked to
produce a source kit for LCG. That was pretty much the only interaction we had other than
doing certification testing before a release.
John.
> On 12/25/2019 12:43 PM, John Forecast wrote:
>
> I?ve never seen a Phase I protocol spec but DECnet/8 was probably closer to Phase I
than Phase II although the dates in the source code appear to cover both phases. Even
after Phase II development started the protocols were still being updated - datagram
service was still in the NSP spec in early 1977 but was dropped about half way through the
development cycle. One advantage we had for DECnet-RSX development (11M/11S/11D and IAS)
was that they all shared common protocol processing code, which had been written for an
earlier advanced development project, so protocol changes were relatively easy to make.
>
> John.
On Dec 25, 2019, at 5:44 PM, Thomas DeBellis
<tommytimesharing at gmail.com> wrote:
What language did you implement in? I've got a number of sources in BLISS, some of
which don't appear to have originated on the 36 bit platform; perhaps some version of
DAP for FTS.
As Johnny pointed out, pretty much everything was written in Macro-11. PHONE was
not midnight project during Phase III IIRC.
How did it work between you RSX folks and the people doing whatever was in the
DN20's? The 20 host got up to being a Phase IV router, but I don't believe the
DN20's went past Phase III.
21 Dec
21 Dec
12:12 a.m.
New subject: Tops-20 ANONYMOUS FAL Preliminary Testing Results/DAP Query
-----Original Message-----
From: Paul Koning
Sent: Friday, December 20, 2019 4:00 PM
To: hecnet at update.uu.se
Subject: Re: [HECnet] Tops-20 ANONYMOUS FAL Preliminary Testing Results/DAP
Query
>
?> If you run across anyone with
DECET-IAS, please let me know...
Did that exist? I'm fairly sure it did. What
about DECnet for RSX-11/D?
Not so sure about that one.
DECNET-IAS existed for certain - I worked at a site that had it. I'm still
kicking myself for not preserving a copy.
--
Lee K. Gleason N5ZMR
Control-G Consultants
lee.gleason at comcast.net
12:12 a.m.
On Dec 20, 2019, at 5:00 PM, Paul Koning
<paulkoning at comcast.net> wrote:
Yes, both DECnet-IAS and DECnet-11/D existed, I was the project lead for the initial phase
II releases (I think there was an updated release later on).
DECnet for RSX-11/D was also one of the initial phase I releases, I used it at a customer
site in Sweden.
John.
On Dec 20, 2019, at 3:46 PM, Lee Gleason
<lee.gleason at comcast.net> wrote:
Did that exist? I'm fairly sure it did. What about DECnet for RSX-11/D? Not so
sure about that one.
paul
-----Original Message----- Do I have any
volunteers who operate something besides VMS? I'd be interested in RSX(+), IAS, RSTS
and OS-8. They never did DECnet on RT-11, did they?
If you run across anyone with DECET-IAS, please let me know...
4:55 a.m.
Was 11/D what also ran in DN20's?? The DECSYSTEM-20 was actually a
cluster of computers with the PDP-10 being connected to up to four
11/40's (or 34?) via interfaces called DTE's.? The DN20 was a repackaged
PDP-11 (34?) and they were incredibly useful.
CU20B, our largest system, had three such units:
1. Front-end PDP-11, ran a variant of RSX called 20F and handled
communications lines, unit record equipment and (I believe) the AN20
ARPAnet IMP interface.
2. DECnet DN20; we had lines to other 20's (KMC11's) before we got
Ethernet, plus a line to a VAX and a DN200 RJE.
3. DN65, used for IBM/HASP communications.? I had source to the DN65,
but I can not remember for the life of me whether it was an RSX
variant or custom code.? That's surprising because I had to make
some modifications to have it talk with one of our 4341's.
I have heard a site (I believe it was Case Western) that had a fourth
PDP-11 serving as a communications concentrator for more lines
(DH11's).? One shudders to think of a 20 with that many users.? Things
got pretty slow at 50 and once you got up to 70, it was time to find
something else to do.
------------------------------------------------------------------------
On 12/20/19 6:12 PM, John Forecast wrote:
Yes, both DECnet-IAS and DECnet-11/D existed, I was the project lead for the initial
phase II releases (I think there was an updated release later on).
DECnet for RSX-11/D was also one of the initial phase I releases, I used it at a customer
site in Sweden.
John.
> ------------------------------------------------------------------------
> On Dec 20, 2019, at 5:00 PM, Paul Koning <paulkoning at comcast.net> wrote:
> Did that exist? I'm fairly sure it did. What about DECnet for RSX-11/D? Not so
sure about that one.
>
> paul
>> ------------------------------------------------------------------------
>> On Dec 20, 2019, at 3:46 PM, Lee Gleason <lee.gleason at comcast.net>
wrote:
>> If you run across anyone with DECET-IAS, please let me know...
>>> ------------------------------------------------------------------------
>>> On 2019-12-20 1:12 PM, Thomas DeBellis wrote:
>>>
>>> Do I have any volunteers who operate something besides VMS? I'd be
interested in RSX(+), IAS, RSTS and OS-8. They never did DECnet on RT-11, did they?
6:26 a.m.
On 12/20/19 10:55 PM, Thomas DeBellis wrote:
I have heard a site (I believe it was Case Western) that had a fourth
PDP-11 serving as a communications concentrator for
more lines
(DH11's).? One shudders to think of a 20 with that many users.? Things
got pretty slow at 50 and once you got up to 70, it was time to find
something else to do.
...and my phone, with eight processors running at 2.7GHz and 4GB of
RAM, can barely keep up with just ME, typing with one or two fingers.
In this industry, we call that "progress".
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
11:10 a.m.
I haven't looked, but I thought the DN20 also ran RSX-20F.
I looked over RSX-20F a few years ago, and it's sortof a mix between
RSX-11D and unmapped RSX-11M. Rather weird beast actually.
One of our -2060 at the University had (well, still have) an extra
cabinet with a Unibus expansion and more terminal lines. Extra DH-11s,
but no extra PDP-11s. All still running through the FE PDP-11.
Johnny
On 2019-12-21 04:55, Thomas DeBellis wrote:
Was 11/D what also ran in DN20's?? The
DECSYSTEM-20 was actually a
cluster of computers with the PDP-10 being connected to up to four
11/40's (or 34?) via interfaces called DTE's.? The DN20 was a repackaged
PDP-11 (34?) and they were incredibly useful.
CU20B, our largest system, had three such units:
1. Front-end PDP-11, ran a variant of RSX called 20F and handled
communications lines, unit record equipment and (I believe) the AN20
ARPAnet IMP interface.
2. DECnet DN20; we had lines to other 20's (KMC11's) before we got
Ethernet, plus a line to a VAX and a DN200 RJE.
3. DN65, used for IBM/HASP communications.? I had source to the DN65,
but I can not remember for the life of me whether it was an RSX
variant or custom code.? That's surprising because I had to make
some modifications to have it talk with one of our 4341's.
I have heard a site (I believe it was Case Western) that had a fourth
PDP-11 serving as a communications concentrator for more lines
(DH11's).? One shudders to think of a 20 with that many users.? Things
got pretty slow at 50 and once you got up to 70, it was time to find
something else to do.
> ------------------------------------------------------------------------
> On 12/20/19 6:12 PM, John Forecast wrote:
> Yes, both DECnet-IAS and DECnet-11/D existed, I was the project lead for the initial
phase II releases (I think there was an updated release later on).
> DECnet for RSX-11/D was also one of the initial phase I releases, I used it at a
customer site in Sweden.
>
> John.
>> ------------------------------------------------------------------------
>> On Dec 20, 2019, at 5:00 PM, Paul Koning<paulkoning at comcast.net>
wrote:
>> Did that exist? I'm fairly sure it did. What about DECnet for RSX-11/D?
Not so sure about that one.
>>
>> paul
>>> ------------------------------------------------------------------------
>>> On Dec 20, 2019, at 3:46 PM, Lee Gleason<lee.gleason at comcast.net>
wrote:
>>> If you run across anyone with DECET-IAS, please let me know...
>>>> ------------------------------------------------------------------------
>>>> On 2019-12-20 1:12 PM, Thomas DeBellis wrote:
>>>>
>>>> Do I have any volunteers who operate something besides VMS? I'd be
interested in RSX(+), IAS, RSTS and OS-8. They never did DECnet on RT-11, did they?
--
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
1795
days inactive
1801
days old
38 comments
10 participants
participants (10)
-
bill.cu＠suddenlink.net -
bqt＠softjar.se -
healyzh＠avanthar.com -
john＠forecast.name -
lee.gleason＠comcast.net -
mark＠matlockfamily.com -
mcguire＠neurotica.com -
paulkoning＠comcast.net -
steve＠davidson.net -
tommytimesharing＠gmail.com