Nuke plants to keep PDP-11 UNTIL 2050!
|
|
|
Comments (17)
How many of you remember what it meant to "fat finger" a bootstrap loader?
Remember "core memory" ?
Ahh, the nostalgia of it...
Remember "core memory" ?
Ahh, the nostalgia of it...
My first college computer was a PDP 11/35 running RSTS/E
Quoting: How many of you remember what it meant to "fat finger" a bootstrap loader?
Ever talk someone through the process over the phone? That was fun.
Quoting:How many of you remember what it meant to "fat finger" a bootstrap loader?
It used to take me about 5 minutes to cold boot a PDP-8. That is, if I got it right the first time around. If I got it right, I was promptly greeted by a single asterisk printed to the teletype.
To answer the original question:
The PDP/VAX space is a much smaller attack surface, so to speak. It would take an evil "old-school" crack to get into a PDP-{8,10,11}. A "ping of death" might knock it around a bit, but that's just a Denial of Service attack. Smashing the stack, or a "return to address" exploit, means one would actually need to know how to make the machine carry out some arbitrary action. Something besides raw calculations. I mean, what are you going to do, send it into an infinite Collatz-sequence calculation loop?
OTOH, a firewall to protect all the machines is still common sense.
The PDP/VAX space is a much smaller attack surface, so to speak. It would take an evil "old-school" crack to get into a PDP-{8,10,11}. A "ping of death" might knock it around a bit, but that's just a Denial of Service attack. Smashing the stack, or a "return to address" exploit, means one would actually need to know how to make the machine carry out some arbitrary action. Something besides raw calculations. I mean, what are you going to do, send it into an infinite Collatz-sequence calculation loop?
OTOH, a firewall to protect all the machines is still common sense.
Caught up in the nostalgia, I was. I doubt that the machines are running COBOL. Accounting functions would be run on commodity hardware. The ad in the article says they are running robotics control applications. For that, you'd be coding in assembler.
I suspect the PDP-11s are running RT-11 OS, used for real time systems and process control. As for networking, there are a few comments on that. In particular, one commenter says:
I suspect the PDP-11s are running RT-11 OS, used for real time systems and process control. As for networking, there are a few comments on that. In particular, one commenter says:
Quoting:PDP-endian... One "endian" nobody checks for any more and which will break nearly any network to host/host to network (including telemetry networks) conversion. I love the smell of meltdown early in the morning, it smells like radioactivity...
PDP 11s are the workhorses of all computers. They are built on solid engineering, which DEC was famous for, and with robust hardware making them highly reliable. No worry about intrusion since they have no wireless to speak off and are on their own isolated hard-wired network when used for instrument monitoring and process control. I worked in a refinery and I am assuming nuclear power plants are even more restricted. If they are not, Systems Managers ought to be fired.
Edited: I think the main reason is cost. Replacing old instrumentations, software, and other equipment is extremely cost prohibitive especially when considering the strict verification and certification process for a nuclear plant.
Edited: I think the main reason is cost. Replacing old instrumentations, software, and other equipment is extremely cost prohibitive especially when considering the strict verification and certification process for a nuclear plant.
@penguinist........You just brought a long forgotten memory back. I used to be an avid reader of Scientific American and circa 1965 I can still remember an issue found at a high school on the New England Tablelands of NSW where I was teaching at the time, and in which a picture of a memory "card" was displayed.
This consisted of a frame with large numbers of fine, horizontal and vertical wires so they formed a regular grid. At the junctions of the wires, tiny ferrite toroids were held in place by the physical structure of the array itself so that there was a single toroid at each of the intersection points of the vertical and horizontal wires. To read the toroidal array there were diagonally running wires through each toroid intersection. It really stuck in my mind as I found it amazing that this "thing" could remember too. Sigh........and everybody, but everybody, fought to use the school sole electronic calculator with its glow tube numerical readouts because it made adding up exam marks so much easier. The cost for a single unit in 1966 must have been enormous.......Now they are throw away items.
@Fettoosh.....you may be right on cost, but I also take the points you and gus3 make in that the PDP-11 would be a difficult device to attack. Both make a lot of sense.
@djohnston.....yes, COBOL is a business language.....I remember that from days spent programming a miniTran computer at the University of New England using punched cards. I also doubted the truth of COBOL, but an associated article I found on this matter of the PDP-11's seemed to suggest it.....
and last but not least @penguinist again......"fatfinger a bootstrap loader".......the last time I used that term of "bootstrap" was with triode valve amplifiers. I have NO idea of what your reference means.
This consisted of a frame with large numbers of fine, horizontal and vertical wires so they formed a regular grid. At the junctions of the wires, tiny ferrite toroids were held in place by the physical structure of the array itself so that there was a single toroid at each of the intersection points of the vertical and horizontal wires. To read the toroidal array there were diagonally running wires through each toroid intersection. It really stuck in my mind as I found it amazing that this "thing" could remember too. Sigh........and everybody, but everybody, fought to use the school sole electronic calculator with its glow tube numerical readouts because it made adding up exam marks so much easier. The cost for a single unit in 1966 must have been enormous.......Now they are throw away items.
@Fettoosh.....you may be right on cost, but I also take the points you and gus3 make in that the PDP-11 would be a difficult device to attack. Both make a lot of sense.
@djohnston.....yes, COBOL is a business language.....I remember that from days spent programming a miniTran computer at the University of New England using punched cards. I also doubted the truth of COBOL, but an associated article I found on this matter of the PDP-11's seemed to suggest it.....
and last but not least @penguinist again......"fatfinger a bootstrap loader".......the last time I used that term of "bootstrap" was with triode valve amplifiers. I have NO idea of what your reference means.
Quoting:"fatfinger a bootstrap loader".......the last time I used that term of "bootstrap" was with triode valve amplifiers. I have NO idea of what your reference means.
The PDP-8 was a 12bit machine. On the console of the machine I ran, there were four sets of switches in groups of three each for directly entering machine words instead of using the paper tape reader on the teletype console. It was easier to read the bootstrap instructions in octal, so each group of three switches represented an "octal" number. The groups of four "octal" numbers were posted on the machine's console. The toggle switches were set on or off to represent 0 through 7.
The first set of 4 "octal" numbers, (or 12bit word) were entered by positioning the toggle switches, then flipping an incrementer switch. If I remember correctly, there were 20 words to enter into the incrementer to boot the machine. After entering the 20 words, you got a boot prompt.
"Fat fingering" meant that you set one or more switches to on, when they should have been off. Or vice-versa. Or a switch wasn't fully in the up or down position. Or you flipped the incrementer switch twice, instead of once, for the same 12bit word.
The experiences and knowledge that so often comes out on this site is staggering. Okay, okay, I know it's bleedin' obvious, but somebody had to say it. The information I am learning is amazing. The last time I saw computers vaguely like you are describing djohnston was at HMAS Nirimba, where I was an officer teaching basic electronics at that apprentice training establishment of the RAN (now decommissioned and replaced by blocks of flats I think) but in the computer training labs, programming by switches was the norm.........and they were mostly dealing with analogue computers into the bargain.....I believe that is because an analogue computer used to find it much easier to deal with real time events ? And that would account for analogue computers being used for gunnery control, etc. at that time.....Mind you, this is going back to 1974.
@Ridcully
Here's an even earlier analogue computer -- probably the oldest analogue computer that I've heard about (a little before your time, I suspect) ;)
http://en.wikipedia.org/wiki/Antikythera_mechanism
http://www.math.sunysb.edu/~tony/whatsnew/column/antikythera...
http://www.youtube.com/watch?v=4eUibFQKJqI
Here's an even earlier analogue computer -- probably the oldest analogue computer that I've heard about (a little before your time, I suspect) ;)
http://en.wikipedia.org/wiki/Antikythera_mechanism
http://www.math.sunysb.edu/~tony/whatsnew/column/antikythera...
http://www.youtube.com/watch?v=4eUibFQKJqI
@BernardSwiss.........Oh yes......I have known about the Antikythera orrery for a long time, but I'd like something a little more modern and one you can find in a number of places around the world.....Here it is:
http://en.wikipedia.org/wiki/MONIAC_Computer
The Philips Hydraulic Computer is a fascinating analogue machine and the things still work. IF you are a Terry Pratchett fan (yes, I am), you will know that the Philips computer has been immortalised in his book "Making Money" where it is called "The Glooper"........The problem becomes that the makers of The Glooper make it "too accurate" so that instead of it simply modelling what the city economics do, it can control what the city economics do.
As it's creator says: "You mean Igor, that if I smash this computer, the city will collapse into financial chaos ?"
which then prompts Igor's reply:
"Yeth Thir, thall I fetch a hammer ?"
(Do remember that Igors ALWAYS lisp. LOL)
PS.......I am now deeply hurt that you have implied that I just might have been around 2000 years ago.....Antique I certainly am, but Ancient........I shall go and have another glass of wine as a "restoratif". :-)
http://en.wikipedia.org/wiki/MONIAC_Computer
The Philips Hydraulic Computer is a fascinating analogue machine and the things still work. IF you are a Terry Pratchett fan (yes, I am), you will know that the Philips computer has been immortalised in his book "Making Money" where it is called "The Glooper"........The problem becomes that the makers of The Glooper make it "too accurate" so that instead of it simply modelling what the city economics do, it can control what the city economics do.
As it's creator says: "You mean Igor, that if I smash this computer, the city will collapse into financial chaos ?"
which then prompts Igor's reply:
"Yeth Thir, thall I fetch a hammer ?"
(Do remember that Igors ALWAYS lisp. LOL)
PS.......I am now deeply hurt that you have implied that I just might have been around 2000 years ago.....Antique I certainly am, but Ancient........I shall go and have another glass of wine as a "restoratif". :-)
Ridcully, you speak of Core memory.
It was used on satellites long after it stopped being used on Earth due to its costs, because it is nearly completely immune to radiation. Unlike semiconductors.
http://en.wikipedia.org/wiki/Magnetic-core_memory
It was used on satellites long after it stopped being used on Earth due to its costs, because it is nearly completely immune to radiation. Unlike semiconductors.
http://en.wikipedia.org/wiki/Magnetic-core_memory
@Riducully Nice -- I'd forgotten about that one. (Saw it demo]d on TV once).
@Bob_Robertson -- I had no idea.
Apparently magnetic-core memory was still used at least as recently as the early Space Shuttles (including the Challenger). http://en.wikipedia.org/wiki/Magnetic-core_memory#Physical_c...
Do you know whether it's still being used in practice (outside of Arduino projects http://www.corememoryshield.com/report.html)?
@Bob_Robertson -- I had no idea.
Apparently magnetic-core memory was still used at least as recently as the early Space Shuttles (including the Challenger). http://en.wikipedia.org/wiki/Magnetic-core_memory#Physical_c...
Do you know whether it's still being used in practice (outside of Arduino projects http://www.corememoryshield.com/report.html)?
Mr. Swiss, I must say that I have no idea if core memory is still being used.
If the use of cell phones in micro satellites is any indication, it could be that shielding has improved a bit.
If the use of cell phones in micro satellites is any indication, it could be that shielding has improved a bit.
The shielding has improved, but the best thing going is a combination of ECC checksums and redundant RAM.
Post a comment
A free account is needed to post. Write your comment here first: your text is kept in this browser while you sign up or log in, and is waiting for you when you come back.
Your message:Markdown: **bold**, *italic*, [text](https://link), - lists, > quotes, $x^2$ for math. Formatting help
Sign up and post (it is free) or Log in and post

I took a look around the web and wikipedia has a very nice article on them plus a photo showing tape deck feeds. I am rather impressed; obviously these computers are extremely reliable and well programmed - especially if they are monitoring nuclear power facilities. http://en.wikipedia.org/wiki/PDP-11