Slashdot Mirror
DECnet Isn't Dead
Ronald Dumsfeld writes "The odds of folks under the age of 25 on Slashdot having heard of DECnet are pretty slim. This article over at Datamation gives some insight into people who've not given up on it. Poke around and find the documentation for the OSI-compliant version, or download the Linux version of the older DECnet IV and bask in the Security Through Obscurity."
It's been ages since I've heard of that. I thought it was confirmed to be dead, unlike our BSD friends whom netcraft seems to confirm to some /.ers... :D
IP, though, is the industry standard protocol. These days, everybody knows how to use TCP/IP. That means anyone also deploying DECnet has to license both protocols. The good news is that the DECnet fees are a bit less than those for TCP/IP.
Did I miss something? So far as I know, the specifications for TCP, IP, and (most) assorted support protocols are openly avaliable, free of charge to implement, screw up, use and abuse. Is this suggesting DECnet fees involve someone paying you to use it? If that's the case, sign me up!
This flies in the face of science.
It is dead.
:-/
And has been for a very very very very long time.
Just like IRC
-Bill
DECnet is often used these days for very mission critical applications. The firm I work for uses DECnet because it is the easiest and most reliable way for us to maintain our VAX and Alpha clusters. Indeed, it is by far the most superior form of networking out there for applications where the uptime must be literally 100%.
We have had sales reps from various vendors come and suggest moving to a Windows 2003/PC setup (HAHA!) or towards a more UNIX/Linux-based setup. But we will stick with our DECnet-based VAX and Alpha clusters because they are known to work, and they work pretty damn well! But that's because it is amongst the finest of DEC engineering. That's the sort of engineering you just don't find these days.
Cyric Zndovzny at your service.
...richie - It is a good day to code.
I'm 35 and I never heard of this. Go figure.
Just because my Windows 3.11 box still can run doesn't mean I am willing to trust it with data.
Maybe to license a third-party's TCP/IP stack implementation?
"God fights on the side with the best artillery." - Napoleon, Marshal of France - speaking truth to power
The OpenVMS implementation of TCP/IP and DECnet must be licensed seperately from the operating system. That is what they mean. The OpenVMS TCP/IP implementation costs less than the OpenVMS DECnet implementation.
Cyric Zndovzny at your service.
The coming of OSI and it's asinine 7-layer model stiltified DECnet in the 90's. I'm sure that being OSI-compliant was a big deal at the time, but nobody cares anymore. And other than crossing the t's and dotting the i's to meet some government spec at the time, nobody really wanted it.
Before OSI, DECnet was sleek, widespread, easy, and portable across many platforms.
After OSI compliance, it was sluggish, cantankerous, difficult, and verbose.
Haven't you been contacted by Nigeria-Soft yet? They kindly sent me an e-mail informing me that I was using TCP/IP unlawfully! I was shocked! Thankfully I was checking my spam folder because this was one message that I definitely didn't to miss because Yahoo was stupid and flagged it as spam. I'm sure you want to avoid using their protocol illegally just as much as I do. You can send your liscencing fee of $1000 to them at 84 Goat Herd Rd, Ngeria. Sure it's a little expensive, but you don't want to be stealing their protocol from them, do you?
The laws of probability forbid it!
I guess if you wanted to keep really sensitive stuff from being seen, you could PIX your application server and behind that, everything runs on DECnet behind the application server.
But, if you're going to do that anyway... you could just as well use Appletalk, VINES or NetBIOS (w/o TCPIP) instead of DECnet... neither of which would be visible to outsiders.
None of this will save you from VBS attacking desktops. Email is email, whether it came in over IP, a floppy disk, or DECnet.
Sorry, dec weenies but ... DEC was dead 10 years ago. There will be no "big comeback".
The DEC way was not "the right way" (TM). Unix was the right way, that's why DEC died.
George Romero will not be making a sequel on this.
I am 27, and nothing about me is slim, however I do have some familiarity with this. Number one, and I am serious, I read a lot (don't get out much- not since my girlfriend started losing air...), and computer history is interesting to me, and number 2, I actually went to a school district where as wee ones in the gifted class (I was in there due to a clerical/administrative error) were heavily exposed tothe computer tech of the day. I actually have spent plenty of time using an acoustic modem, the type where the old style phone compnay receiver fits in the the two holes on the modem.... Now those were cool. Even ascii images took 30 minutes to download with .05 BAUD....
And All I Ask is a Tall Ship And a Star to Steer Her By
Stumbled onto a VMS/DECNet machine and want to explore a little? First try "show known nodes", and then... our friends at Phrack have a HOWTO guide, including a copy of the all-important "TELL.COM".
-Mark: (remembers VMS) && (age > 25)
Go figure.
Back in the day, companies charged you for a working tcp/ip stack to go with your spiffy OS that didnt have it. There were several for the Vax and even the Mac you had to have an add-on tcp/ip stack to use because they were all appletalk based. Of course, that was when there were only appletalk networks going to the desktop. Most of the other networks were only in cold storage (data centers).
-----
wish i had some mod points ;)
thanks for your insight
Some little startup company called Microsoft says it's going to be out next year.
Sorry about the writing. Robot fingers, you know? Cliff Steele in DOOM PATROL #23
Like ASK JEEVES for instance. I'm going more and more to ask jeeves which is powered by the teoma search engine when google gives me results that aren't too relevant. Normally ask jeeves gives me more relevant results. So don't be too quick to pronounce things as dead.
/. because you never get to hear about other search engines , but when google does something minor it's front page news, so to speak.
I've often wondered if google indirectly pays for editorials on
I think we should see more stories about other search engines if only to prove that google isn't paying for editorial.
So while DECnet and things from the past hang around the reason is because people like them.
I'm 17 and have heard of DECnet. Considering that I'm the grandchild a former Digital employee and I have a DCU check card in my wallet, I think I have a reason to know a fair bit about them.
If you're similarly interested in computer history, segue into The Soul Of A New Machine by Tracy Kidder. It talks about Data General building a 32-bit minicomputer.
1)Switch reading material (I suggest "Home Depots' Guide to Plastic Hole Repair" /. that actually makes sense...
2)Dump that b*tch! How dare she "deflate" your ego to the point where you won't leave the house
3)Short yellow buses don't mean "gifted" (call your mom, she might be straight with you now)
4)Acoustic modems were not cool. Ever.
5)Lose some weight and wtf, post something to
Why, I use DECnet all the time. I am a purist, and refuse to work with upstart protocols like TCP/IP. I believe the height of technology in its purist form came in 1985. So why upgrade?
Why, my gopher web server works just fine, and I run it from a floppy disk on my 8088 XT. In fact, I do everything on my 8088 XT. It can even play Midi files in mono! Sure, it's not as pretty as some of the fancy shmacy new wave "windows" systems, but show me something you can't do from the command line in DOS, and I'll show you something I refuse to learn how to do.
Why, when I was your age, we had to walk ten miles to school, program in BASIC, and the games we played were based on revolutionary 8 bit technology! We didn't have an "internet." Internet schminternet, give me a text based BBS for my Hayes 3 Baud modem, and I can download over pixilated porn till the cows come home. And we liked it!
This signature has Super Cow Powers
And here I was just sitting here thinking I needed an application to do some DNA session control.
I'm ready to get my RFC 1859 on.
To an operating system with TCP/IP, DECNET, IPX and SNA support -
OS/2
In the early 90's, if you wanted, you could get OS/2 to load a whole pile of transport protocols - which was pretty much necessary for the alphabet soup that ran client-server apps back then. In fact, Doom ran on IPX/SPX before it ran in TCP/IP.
That is quality comedy!
Not trolling, but genuinely curious as to what requires 100% up time?
The system that is responsible for stroking the CEO's ego.
That said, there are many communications systems that are *supposed* to work 100% of the time by law. As an example: EMS call routing is outsourced in many localities and there are stiff fines for downtime. Then there are military mission critical systems which are spec'd to have 100% uptime (though they do generally have failover or some other type of redundancy). The monitoring system for a power plant is generally both spec'd for 100% uptime and covered by redundancy. However, I would guess that most businesses fall into the "ego stroking" category. Downtime of a single system should not impact operations--there should be a redundant system, a queuing service, or some other kind of failover.
BITNet had a chat facility that is similar to IRC, but in BITNet chatland, network operators would roam the rooms. They would force everyone to use their real names and would ban people for the flimsiest of reasons: e.g. discussing sex-oriented topics.
Those who wished to mock VMS would say "VMS Only has two commands, SET and LOGOUT"
Sadly, SET was terribly overloaded ... SET DEFAULT was how you changed (among other things) your current working directory; logging into another host across the network was SET HOST; disabling traps in a .com file was SET NOON;
I loved VMS, not because it was a speedy lightweight OS (it was absolutely the opposite in every way) ; but it was the friendliest OS out there for the hard-core assembly language programmer, and the VAX has an architecture that makes programming in assembly a joy.
How does the Slashdot Effect happen given that no slashdotters ever RTFA?
Yeah, I remember DECnet. The coolest thing about it is that it required you to have a special DECnet MAC address for every Ethernet port on each host. The good news is that this led to widespread Ethernet MAC reprogrammability...
I'm 25, the IT Admin of a company which still uses DecNET! Yes, Imagine! A company using ancient Technology, we may soon change our name to Microsoft even!.
DecNet was never a bad technology, it worked well for what it needed to do. And that's that.
Very entertaining book, if a bit dramatized. It's the story of the invention of the then-modern VAX system in the late 1970s. Guess what, they used emulation technology, just like VMWare et al.
Required reading in my college 15 years ago.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
What about gambling sites? Especially around the superbowl. We have seen reports the cyberblackmail of these types of sites.
Fight Spammers!
As a user of the linux decnet stack, I would say the Linux decnet stack works pretty well for talking to old VAXen. There are still places with old VAX computer embedded in equipment that would take millions to replace. The Navy is using Charon VAX http://www.softresint.com/charon-vax/ in some places to keep from having to replace the attached hardware. SIMH http://simh.trailing-edge.com/ works very well for emulating a vax, but is software only. A vax emulation running on SIMH on linux can talk decnet, and so can the linux machine it runs on. However, because DECNET sets the mac address as the decnet address, the Linux's decnet can't talk to the SIMH running on it. So, I had to put tcp/ip on the simulator to get them to talk. It would be nice if Linux's emulator could set it's mac address at runtime, and have several, so it could to the routing, and talk to the SIMH emulator, but it isn't possible now.
Ol' Rick Dawson had a farm EIEIO
Unimpressed by a lack of slashdot effect?
Disappointed that you didn't make an impression on the Deathrow Cluster?
Slashdot them for real!!! by telnetting to dahmer.vistech.net, manson.vistech.net, or gein.vistech.net.
Monitor your attempts at destruction here.
Alpha is dead, but DECnet lives on.
*sigh*
500GB of disk, 5TB of transfer, $5.95/mo
Yes, TCP/IP has not evolved these layers at all. In stead, TCP/IP users are forced to accept that TCP/IP only provides reliable socket-based bi-directional streams of data routed world wide. Meanwhile in stead of being comfortably locked into a proprietary system, they face the challenge of choosing which applications to use to manage their sessions, presentation and file transfer.
HTTP(S), WebDAV, (S)FTP, SSH, SOAP, JMS, BXXP, XMPP, RTSP, SIP, NFS, SMB, NNTP, IMAP, etc. etc. And all of these protocols come with their own strengths and weaknesses! Worse, you could even swap TCP/IP out from underneath some of these protocols in favor of, for example, IPv6 or in some cases even an old dinosaur like NetBIOS.
To make matters worse, all these protocols come with easy-to use APIs, libraries, executable tools and even multi-vendor support, so far as to even be integrated into development environments such as "Java" or "Perl"..
The obvious drawback of this is of course that relying on these, for the most part, "open standards" makes it easier for your software to interoperate and be compatible across platforms and networks.
Next article written by Captain Obvious; "Many enterprises using Windows file sharing to replicate mission-critical information across Windows systems."
(Not that there's anything wrong with being lazy and using the OS' default transparant network thingamajig.. But that's not exactly winning on merits)
Now, if any one has any information as why DECnet is (supposedly) so much more robust and dependable than TCP/IP (especially DECnet-over-IP), I'd like to hear it. Does it use error correcting codes? Does it have some sort of secDNS equivalent (or even an analog to secure BGP? that would be kinda neat).
SCO employee? Check out the bounty
Then that puts me back in the Devonian era (1985), when I did the same thing in Vax Fortran and assembler!
an ill wind that blows no good
DecNet sounds wonderful and all; without doing a lot of hunting, can someone summarize the technical aspects of the protocol/implementation that makes it so?
Love many, trust a few, do harm to none.
Imagine a non-Beowulf cluster of MicroVAXes, out of the box seamlessly providing clustered batch processing for rooms full of kerosene-fueled electrostatic plotters . . . . dagnabbit, you kids get the hell off my lawn!
Give a man a fish and you have fed him for today. Teach a man to fish, and he'll say "WHERE'S MY FISH, YOU IDIOT?"
John Sauter (J_Sauter@Empire.Net), a former DEC software engineer who worked on DECnet
Vax's (and PDPs) were awesome in their day. But that is over. Now, they are expensive, and a Linux cluster can do the job for a fraction of the price.
But with that said, they were good boxes in the days. I have a number of years on Mumps on PDPs and Vaxs.
DECNET? Bah, too much work! :)
...hey wait a sec, NetBEUI sucks!! UGH! Darn broadcast protocols. What was I smoking back then?
I ran NetBEUI on a small company network for years. Want to know why? It wasn't routable.
The logic was you can't get compromised from the Internet if you cant route off network. Of course that's not true, but it really does make it harder for anyone to break in. Of course, that was before they discovered VPN's and Terminal Services, so NetBEUI went away and the network went all TCP/IP.
NetBEUI...good times...good times...
MicroVAX? Newbie. I was imagining a DECluster of 785s and a refrigerator-size disk subsystem, used for development work in the 1980s and 1990s (and, for all I know, today). Four packs, 220 megabytes each. Fun, fun, fun... :)
The clearance system sounds logical. It is not. It is completely arbitrary. -- John Bolton
Dang and I thought I was l33t 'cause of my old tech but you are 100x as l33t.
No, BITNET was based on one of the technologies IBM mainframes (especially VM systems, IIRC) used quite often. We had several VAXes, and had to buy third-party software (JNet) to connect them to BITNET.
We also bought third-party software to do TCP/IP (MultiNet). In the latter days I was there, we ran our BITNET links over TCP connections to other universities, instead of having dedicated leased lines or whatever.
Well not really but we (two of us) do happen to be working after hours on a decnet problem as this article appeared.
It may not be dead but it's future life is pretty limited despite it's strengths.
I am almost 30! :(
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
Thats 3c per packet for every data packet you've ever sent.
To avoid unpleasant visits from the RIAA, FAST, NAACP, CIA, MIB, etc, email me for details of where to the outstanding money to:
Sent from my ASR33 using ASCII
File under 'M' for 'Manic ranting'
You are correct. The DECnet fees often are less than those for TCP/IP. My mistake.
Cyric Zndovzny at your service.
Back in the late 90's I did some work on the VMS TCPip stack. It was handy being able to uninstall the IP stack remotely over a DECnet connection.
Where law ends, tyranny begins -- William Pitt
Subject says it all (but the posting UI insists I type something here).
MTS anyday
According to the wikipedia link:
"It evolved into one of the first peer-to-peer network architectures"
This thing could be used for sharing music! The RIAA will be after them!
Coder's Stone: The programming language quick ref for iPad
"How true. What other architecture has a single instruction for factorising polynomials."
o _024.html#4515ch9_134
I wasn't entirely sure you were joking.. so I did a quicky search on the web for this instruction. I didn't find one that factored polynomials, but there _is_ one which evaluates polynomials - quite silly.
see http://h71000.www7.hp.com/doc/72final/4515/4515pr
Curiously, there is also an instruction to compute CRCs, in addition to instructions for working on queue data structures (apparently internally implemented as a circular doubly linked list).
CISC, indeed.
I'm 24.5 (and for the ./ers that are too long in front of the screen, that's under 25) and have used it in practice.
Some organizations just don't let go.
When I tried to look at the network rack, I was amazed of the lengths that people use to keep these things alive.
I also encountered with SNA (heavy use), I had to support 486's and 386's in the days of P4's in order to use the ISA SNA card. That was already (mostly) converted to some cheaper TCP/IP based technology.
If any administative person is reading this:
It is cheaper (TCOwise) to redevelop a software that uses an architecture that is supposed to be dead than keeping that technology alive.
The best (old) tech I have used are dumb terminals (coax). They just work(tm). There is no reason new techonlogy will even work as good as those green terminals. And they force you to use a text mode and no mouse (a good thing). I simply don't know why this particular tech was replaced on my organization. Wierd priorities.
I'm not going to downplay the stability of the old mainframes, but what did you actually do with them?
Console apps, simple OS design, hardware from the same vendor..
If you set up a single new big Sun box with X terminals, you'd get similar availability and a lot more functionality then the old PDP 11.
But when you try to use hardware that's mass-produced and multi-vendor, based on some good and some loose standards, with software from all over the place and an operating system from Microsoft, you'd bound to run into problems and it's really no surprise.
- It's not the Macs I hate. It's Digg users. -
We even had a testing MicroVAX/1 at the FDA that survived a momentary power outage that downed the cluster (your tax dollars were too meager to afford UPSes in those days!). The standing joke was that the uVAX was so damn slow it didn't notice that the power went out.
I also remember being called a clueless kid, because I'd never flipped the switches on the front of a PDP to bootstrap it. Good times indeed.
You can only drink 30 or 40 glasses of beer a day, no matter how rich you are.
-- Colonel Adolphus Busch
Let's not forget about the old X.25 Telenet network which is still publicly accessible. A few years back myself and a friend of mine wrote an NUA scanner to poke around Telenet and see what's out there. We were awfully surprised to see that there were many systems still openly accessible. Not only that but we also found that there were a plethora of freely available PADs in most major cities. At one point we had compiled a list of several hundred of these numbers and methodically began mapping out large portions of the network. I would fully recomend those who are interested in arcane forms of networking to dig into this a little further. You may be pleasantly surprised.
All this talk about DECNet and no mention of twinax?
While Dave Cutler (perpitrator of a great many OS atrocities) once remarked that "UNIX is a junk OS designed by a committee of PH.D.s," his operating systems have some profound problems.
Can anyone argue that VMS DCL has evolved as much as the Bourne environment? I believe it was Dennis Ritchie who severely criticised VMS for integrating most of the command interpreter into the kernel (which Cutler again did by moving many drivers from Ring 3 to Ring 0 in NT - same mistake?).
Yes, VMS has awesome capability in clustering and security, far outpacing most past and future implementations, but much of it has evolved about as far as JCL in becoming a modern system.
Lots of things just simply can't be done on VMS. UNIX is much more of a "happy medium" and has proven to be highly adaptable.
The Deathrow OpenVMS Cluster [http://deathrow.vistech.net/ uses DECNet between Alphas (SET HOST GEIN/SET HOST DAHMER), VAXen (SET HOST MANSON) and Linux (SET HOST JACK). If you're really wanting to play with DECNet, you can setup an account there and "hack" away.
The projects I've worked for the last 8 years or so have used VAX and Alpha VMS and I can say that the much-vaunted uptime for VMS tends to be exaggerated. Yes VMS is generally solid, that I won't argue. However, it is very vulnerable to HW failure, just like anything else, and maybe more so than anything else we have around. We have had many many instances of a rogue VAXStation or microVAX taking out an entire cluster, redundancy and all. I see that as unacceptable.
... bleh.
You might say it must have been a admin/config problem. Weeeellll maybe (those guys seem to really know their sh-t cold, but one never knows) but then if it's that easy to misconfigure, how reliable is it really? And have you ever tried FINDING people that can maintain this stuff?
Lately we've been migrating off to the wintel world (and to some SGI as well) and the uptime numbers really have not changed that much. Some windows services tend to go down more often than their VMS equivalent, but things are mostly the same. The only reason we have to keep VAXen around is legacy applications that would be very very hard to port off of VMS. Anyone who has ever had to convert a G-Float to an IEEE double in order to use old VAX centric data sets know what I'm talking about here
"There's no secret. You just press the accelerator to the floor and keep turning left." -- Bill Vukovich
I like playing "spot the userID numbers" when a story like this about some old tech comes along ... The contrast of "Holy crap! I'd forgotten all about that stuff!" from the low-number userIDs versus "WTF? O1d junx 1s teh sux0r!!" from the high-number userIDs cracks me up every time :-)
Me? Debunk an American myth? And take my life in my hands?
I remember well having to type in those damn long IPX addresses to connect to a network multiplayer game.
...the bad old days, I say...
Ah, the days of suffering under this dreadful bolt-on to Win3.11...
I still wince whenever I see an old Gateway 486 sitting in the back of some closet somewhere.
I have something in common with Stephen Hawking...
The Dead Collector: Bring out yer dead.
[a man puts a body on the cart]
Large Man with Dead Body: Here's one.
The Dead Collector: That'll be ninepence.
The Dead Body That Claims It Isn't: I'm not dead.
The Dead Collector: What?
Large Man with Dead Body: Nothing. There's your ninepence.
The Dead Body That Claims It Isn't: I'm not dead.
The Dead Collector: 'Ere, he says he's not dead.
Large Man with Dead Body: Yes he is.
The Dead Body That Claims It Isn't: I'm not.
The Dead Collector: He isn't.
Large Man with Dead Body: Well, he will be soon, he's very ill.
The Dead Body That Claims It Isn't: I'm getting better.
Large Man with Dead Body: No you're not, you'll be stone dead in a moment.
The Dead Collector: Well, I can't take him like that. It's against regulations.
The Dead Body That Claims It Isn't: I don't want to go on the cart.
Large Man with Dead Body: Oh, don't be such a baby.
The Dead Collector: I can't take him.
The Dead Body That Claims It Isn't: I feel fine.
Large Man with Dead Body: Oh, do me a favor.
The Dead Collector: I can't.
Large Man with Dead Body: Well, can you hang around for a couple of minutes? He won't be long.
The Dead Collector: I promised I'd be at the Robinsons'. They've lost nine today.
Large Man with Dead Body: Well, when's your next round?
The Dead Collector: Thursday.
The Dead Body That Claims It Isn't: I think I'll go for a walk.
Large Man with Dead Body: You're not fooling anyone, you know. Isn't there anything you could do?
The Dead Body That Claims It Isn't: I feel happy. I feel happy.
[the Dead Collector glances up and down the street furtively, then silences the Body with his a whack of his club]
Large Man with Dead Body: Ah, thank you very much.
The Dead Collector: Not at all. See you on Thursday.
Large Man with Dead Body: Right.
Well, there's spam egg sausage and spam, that's not got much spam in it.
Netcraft didn't confirm it.
File under 'M' for 'Manic ranting'
Somehow using "y'all" and "nappies" together seems really odd. Are you an English Redneck???
TGV, makers of the de facto TCP/IP stack for VMS, were as much to blame for the failure of DECnet V. DEC's own pathetic UCX (know as sUCX) TCP/IP stack could not compare.
Where are they now?
The dual-processor 780 was the 782. Asymmetric multi-processing (only one CPU could do I/O, I believe). ISTR that the 785 was a 780 implemented entirely with CMOS parts (instead of the TTL used in the 780 and 750). I'm pretty fuzzy on the details of both of these, I haven't worked on a VAX since '94 or thereabouts. Still have my old architecture handbook and green card around somewhere, and my VMS 5.5 internals book.
Just junk food for thought...
I remember me using pathworks on DEC pc's to connect an Alpha 150 server loaded with OSF/1... that old days when gopher and ftp where the kings of the net surfing.
Somehow, I figure the number of people still using DECnet exactly matches the number of people still reading Datamation.
Help us build a better map!
VAX/VMS had 64-bit computing, seamless virtual memory management, no root superuser, granular permissions, clustering, and all the other stuff *nix is just getting now (thanks to Hans Reiser, Ted T'so, Linus Torvalds and friends) decades ago. VMS was also the first POSIX-compliant system, didja realize that?
The problems with VMS were that it was expensive and closed source, and it was unfriendly to people whose native language was not English (which is why Torvalds hated it, incidentally).
I left unix for VMS because the unix geeks were condescending and unable to admit their OS had flaws, which made it impossible to fix them. I left VMS for linux because the linux geeks were actually addressing the fundamental flaws of unix.
I remember when it came out, we installed it on our test MicroVax, and the machine basically fell over. After some modparams tuning got it up and it ran really slowly. Tried it on our live system one weekend and got the same results, so went back to Phase IV. Been running Phase IV on our VaxCluster ever since and that was nearly 10 years ago.
Jonathan
Need a car checked out? - http://www.goodbuy-carchecks.co.uk/
In an attempt to satisfy the federal government, DEC actually implemented the OSI reference model. The whole bloody thing, as documented by the model itself, which is how the world found out it is a bad idea.
I've installed it. I've used it. I remember the whole GOSIP debacle. I remember ripping it out by the roots and reinstalling DECnet Phase IV - which was excellent, although a bit bursty on low bandwidth links.
DECnet addresses map into MAC addresses. For instance, the DECnet address 3.100 maps to AA-00-04-00-0C-64
I have enough trouble with FC3 boxen that complain tht their physical address has changed (due to the luser concerned having applied 240v instead of the generally accepted low lan voltage to the RJ45 connector, etc), plus the overall headache that is DHCP based on MAC addresssses, that the last thing I need is a bunch of machines changing their apparent MAC addresses to fit with some crufty network architecture that I expunged from my network over 10 years ago.
Not to mention the general nightmare that is trying to cram all of those DEC network drivers into high memory on Win9x/DOS to save on base memory so you could provide network services via PATHWorks.
No thanks. I do have a VAX, but I'll make do with the wonderfully broken network interface that is TCP/IP Services for VMS, rather than inflict DECnet on my LAN.
From the article: "Since DECnet is a less well-known protocol, nobody is attempting to hack it," says Quayle.
Well, they weren't. Until you said that.
-Graham
Scary devil monastery?
this website surely isn't all that remains of that fine
publication is it?
Do you own/run that ISP Empire.net or do you just work there? Never used DECnet myself. I was only born in 1983. ;)
It's pining for the fjords!
The world's burning. Moped Jesus spotted on I50. Details at 11.
A couple of years ago, I was putting together a web server front-end that was going back to an Oracle database. TCP/IP came in through one NIC, but the connection to the back-end was through another NIC that only had DECNet installed. The architecture guys were pretty sure that, even if the web server were somehow hacked in the DMZ, young 'uns would never figure out how to get back to the database.
I've only heard of it from kernel compiling. I never bothered reading about it to learn what it was. I did read some about ARCnet however, since I saw there was an intriguing comment in the menuconfig--"check out the (arguably) beautiful poetry in file:Documentation/networking/arcnet.txt"
:)
I looked in there and it found the following text:
"This driver's getting fat and beefy,
But my cat is still named Fifi."
how cute
Sigs pose an operational security risk and help the baddies aggregate data. I guess commenting does too, oops.
... or download the Linux version...
... doubt anyone reading slashdot under 45 has heard of that one...
Can I get that off DECUS?
DECNET is like a lot of other things made by the Digital Equipment Corporation: Solid. It's amazing how much of this heavy old equipment and opscure operating systems are still around, remembered only they fail, which is sometimes a decade after they were forgotten about. Unfortunately, With the demise of DEC shortly after it's takeover by Compaq is 1998, parts and skill are getting hard to come by. I have a friend who rebuilds certain tape controllers into disk controlles, and he gets good money for them. DEC made Industrial equipment, so far different than the disposable crap that we have nowadays. But damn, it sure was expensive...hundreds of thousands of dollars for a whole setup.
I miss the old stuff. But I don't have the room to store it or the money to pay for the electricity to power it.
You can add water treatment plants and sewage plants to the list of businesses that require 100% uptime.
Pretty much any business that also makes use of Real Time systems.
Most of us have moved on to other things. Quite a few of the old crowd are still in the Santa Cruz area, though.
Breathe continuously
Aha, sorry, so that's not your current occupation then, hehe. ;)
In the early 90's, OSI was going to take over. Several interoperating implementations of the OSI services (X.509, etc.) were shown at InterOp 1991.
OSI was hot, and everything was going to have to be fully OSI-compliant or die.
It may not have really happened (IP of course still remains dominant), but it was the hot thing in the industry, and DEC had to respond.
I even used the ISO/OSI replacement for FTP once (over IP) back then, I forget the name of it. Anyone remember?
http://lkml.org/lkml/2005/8/20/95
hey, there's a decnet sourceforge project.
...
AT LAST! we can marry freedce, a little used, totally misunderstood and strategically critical project back with DECnet, a little used
DCE 1.1 (from which freedce is derived) shows signs of support for DECnet having been pulled (from the "commercial" version(s)) in rather a hurry.
the neat thing about dce/rpc is that it supports literally _any_ transport and i do mean any.
TGV was acquired by Cisco (I still have the t-shirt with the Cisco logo and the Multinet jaguar that says "Assimilation Complete").
:^)
:^)
Later, Multinet ended-up in the capable hands of Process Software (makers of TCPware).
I've used UCX, TCPware and Multinet in a number of different production environments. Give me Multinet over sUCX any day. (TCPware is a close second to Multinet).
Sadly, most of the Slashdot crowd is too young to be able to tell whether I am telling the truth or if I just threw a bunch of random words together. No mods points for me.
As for DECnet Phase V, or "DECnet Plus", it was a sucky piece of software that was wayyyy too complex for what we (the customers) needed. I played with DECnet Pus for a while and then de-installed it to go back to Phase IV.
If I had a dime for every CSC call or DSNlink case that was caused by DECnet Plus, I'd be one rich dude.
Compared to other mini systems of the era the DEC VAX kicked major butt in the scientific and education arenas before getting sucked into Compaq. Still, we had a major challenge getting DECnet and TCP/IP to coexist at the desktop level. Macintoshes required DAVE, and until Microsoft came out with their TCP/IP stack (anyone remember Wolverine for Win3.11?) we had to struggle with Pathworks. I built three identical PC's running WFW3.11 and installed Pathworks on all three (massaging the loads and so forth carefully). Only one ran! Never found out why as we junked Pathworks.
In another job, I sysadmin'd a VAX network that was used to build complex optics. The company wanted to keep that network separate, so we installed DEMPR's (Digital Equipment MultiPort Repeaters - 8-10Mb coax ports with one uplink) all over the building for the engineers and the fabrication stands. The system ran 7x24 for over four years. I think the IBM mainframe in the same building came down a few times and stayed down longer than anyone cared to admit. Ah, when hardware was really hardware!
I think, therefore I am - Rene Descartes; I yam what I yam, an' that's what I yam - Popeye
Sticking strictly to network protocols, I can think of a few others OS/2 supports: Appletalk, OSI, Banyan Vines, LANtastic (it had a proprietary protocol), and RIPL (which arguably has some attributes of a network protocol) come to mind. I'm probably still forgetting some.
Once upon a time, my company published a version of DECnet for the Amiga.
We'd licensed an Apple Pascal version from Thursby Software and ported it to C. Thursby are the people who later made DAVE for the Mac, the way you'd connect to a Windows SMB network before OS X came along. They'd had a DECnet product before DAVE.
I have a DEC Pro 350 in the basement that we used for debugging and development.
Curator of the Jefferson Computer Museum http://www.threedee.com/jcm
Dios mio! Why don't we bring back ARCNET too while we're at it? There's a reason Ethernet and TCP/IP killed off these old network technologies. If DECnet isn't dead, we all ought to pitch in and stomp it back down into the grave...
--Mythos
The 7 layer model was bad enough (the bottom layers weren't bad, but no one could ever figure out the difference between "Application" and "Presentation as I recall or something like that)but the actual protocols blew donkeys.
For a long time there was no connectionless (UDP equivalent) protocol. We used to have UltraNet as an option on our mini-supers and I was the guy in charge of porting the drivers, etc. UltraNet, in 1990, ran at up to 1Gbps. Very cool stuff. Unfortunately they chose to make their core protocol OSI and it was connection based only (in order to get things running at 1Gbps they offloaded most of the protocol processing into hardware). This was fine in certain supercomputing environments where people wanted to to just blast files around with FTP but as soon as you got out into the working world people wanted to run NFS over it (this was pre-TCP NFS) and the performance just stank.
At another point I want to a Usenix and went to a seminar on the OSI RPC protocol. I had been digging around in the guts of SunRPC and NFS quite a bit at that time and so I went up after the class was over and asked the instructor about how they had made tradeoffs, etc. This guy was one of the people who wrote the spec and he said to me "Well, I've never really looked at SunRPC". And boy, did it show. The protocol was all bit and byte stuffed for maximum bit efficiency just as we were moving to faster networks (so not so much need for bit efficiency) and RISC processors that gave you huge penalties for accessing non-aligned data (so processing the protocol would be damn slown). That convinced me that the whole thing was a crock. When GOSIP (this was the government push to have OSI as their standard networking protocols) died everyone ran away from OSI as fast as they could holding their noses.
...working in a IT department in Orlando in 1992, running thinwire through the walls and ceilings. The damn network crashed on a daily basis. Usually all it took was someone to move their PC and break the connection for the whole hallway. It took a full day of phone support with DEC to configure a laptop for DECNET.
1. It had IS-IS routing with link-state-routing. A very rough analogue is OSPF.
2. It had a end-system autoconfiguration, something that IPV6 is suposed to move us to. End system addresses were 20 byte NSAPs and there was complete support for multi-pathing.
3. It had a complete distributed naming service. Like DNS on acid.
4. It had a distributed time service. Like NTP.
5. It included X.400 mail.
6. It included X.500 directory services.
and so on and so on. And they came pretty close to pulling it all off. 10 years ago. as part of the program, the DECnis faimly of routers were the world's fastest routers while looking up 20 byte addresses. When adapted to run IP late in the development cycle, they were the world's fastest at that too.
I for one was proud to have been a part of that.
P.S. Any mistakes in the above you can put down to my age being >> 25!!!
The lessons of history teach us - if they teach us anything - that nobody learns the lessons that history teaches us.
Finally, the virtues of older products like DECnet are recognized. Solid design and engineering, easy to use and implement plus it keeps going like the Energizer Bunny. Now I understand why there was so much activity when I eBay'd my DECnet documentation and architecture specs for DDCMP, DNA, MAP, etc. Yes, I am an old DECie and former member of the ESG product line DECnet 10 marketing team. I started using DECnet in the 70's. So, when is TECO going to get some respect?
Thank goodness they're talking about phase IV and not V. Anyone ever actually get phase V running. What a nightmare. DECNet.. My first networking job.
I'm going to go lay down now. These old fingers don't type as well as they used to.
Well done Sir!
1 QA-MT1AA-H8 HP OVMS Alpha OS media kit CD-ROM $362.00 $362.00
1 QL-MT3AA-3H HP OpenVMS Concurrent License 64 users $12,132.00 $12,132.00
So thats about 13,000$ (USD) for the operating system for 64 client licenses. The cost of the alphaserver for a decent configuration is about 350,000$.
Where can I get an implementation of DECNet for Linux?!
This is something that I have always just ignored, but occasionally when I am sniffing a network to troubleshoot a problem I will see DECnet protocol come up in the trace. For the most part it is usually coming from the router and in an hour or two of tracing it may only be two or three packets. This is usually at really small offices with zero Digital Equipment machines. Any ideas on why I see these on a suprisingly (1 out 8 maybe) regular basis?
How are some of our other old friends doing.
Modern NOS's have nearly caught up with where Banyan Vines was 15 years ago. Dont flame me I said nearly.
... it only went away...
I'm not even old enough to drive, and I have heard of DECnet. I guess I am more of a geek than I once thought I was.
I have no problem with creimer's view, because he is doing what he wants to do. You, however, have insulted me as well as the intelligence and dedication of many people.
Get you unfounded dogma off of public places. Or, put your money where your mouth is, and prove what you just said.
I'll bet you $100 (no kidding) that you can't defend your position with passages (not just quotes) from the Bible. And, for an even easier bet for you, I'll bet you $1 and let you defend your statement with any Christian references you want to use, even modern works. Bear in mind, I'll use my own references to rebut. You won't even win $1, my Christian friend, because you can't.
And while we're on the way offtopic subject of ignoring rules, at what point did you decide that the rules in the Old Testament didn't apply to Christians? Did you read it between the lines?
Or maybe you should spend your time thinking for yourself, instead of blindly following rules set down by people almost a thousand years ago. You're certainly not following the path of the original Christian church, but instead agreeing with the statements made during the Middle Ages, for two reasons.
First of all, you're not going to find anything about a prophet or Christ demanding virginity. Virginity was not an issue so much until the Middle Ages, for various reasons. Why don't you read some history? It could help your outlook on life.
Second of all, what you will find, if you check the original Hebrew and Greek, is that Jesus, an unmarried man, had sex, as did at least one of his disciples. Was it okay for them, but not you? Were times different then?
Playing pornographics games during the day is evil! Play at night!
The dual-processor 780 was the 782. Asymmetric multi-processing (only one CPU could do I/O, I believe). ISTR that the 785 was a 780 implemented entirely with CMOS parts (instead of the TTL used in the 780 and 750). I'm pretty fuzzy on the details of both of these, I haven't worked on a VAX since '94 or thereabouts. Still have my old architecture handbook and green card around somewhere, and my VMS 5.5 internals book.
You are correct. The782 was only useful in compute-intensive operations as the primary CPU had to handle all interrupts, I/O, and was even responsible for scheduling work on the secondary processor. The 785 was a mid-life kicker for the 780 until the 8000 series of systems arrived.
I'm not yet 25 and I still work with DECnet every day at the office. That's not a good thing mind you....we really need to get off these OpenVMS/Alpha machines!
Curiously, there is also an instruction to compute CRCs, in addition to instructions for working on queue data structures (apparently internally implemented as a circular doubly linked list).
It makes perfect sense. Operating Systems spend a lot of time managing queues. By having a machine instruction that implements those operations, you don't have to worry about syncronization issues, etc.
The VAX hardware engineers worked closely with the VMS software engineers to do just those kinds of things. That's an advantage of having your CPU and OS coming from the same vendor.
Actually, I was using a 780 that was field-upgraded to a 785. The new label on the box said "VAX 11/780-5". Given that almost nothing was left of the 780 except cables and the cabinet, I used to wonder if this was DEC's way of gaming the government regulations about buying computers. "No, sir, this isn't a new computer, it's an upgrade."
My last VAX development work was writing FORTRAN programs to simulate the intermolecular Raman spectra of naphthalene trimer and tetramer in grad school, using what the sysadmin said was the last, fastest workstation model. This was in 1996.
The clearance system sounds logical. It is not. It is completely arbitrary. -- John Bolton
You're right, upgrading just to upgrade isn't such a hot idea. However, upgrading because DEC/Compaq/HP dropped VAX support a couple of years back is a different thing entirely. Several places I've worked kept on going in an unsupported config and patching it together until they were forced to upgrade under extreme pressure, and none of those ever turned out well.
... although actually if you count all of the engineering time being blown when the system is down it does get ugly really fast. I'd never really thought about it in terms of money per time.)
Yeah, you can make the jump from VAX to Alpha to stay on VMS, but I guess where I work we see VMS on Alpha as a ship that's already going down. If we are going to have to migrate, moving to a dying platform just didn't seem like the way to go. Also I think that someone high up in our organization has a lot of microsoft stock, but that's just my cynical side speaking.
(And downtime doesn't quite cost us a grand a minute! Ouch
"There's no secret. You just press the accelerator to the floor and keep turning left." -- Bill Vukovich
As far as I know, only DuPont and DEC ever came close to the DECnet node address limitation (perhaps the other large nets didn't like to brag about their size).
:)
If DEC had sunk the amount of capital they wasted on DECnet-OSI into DECnet phase IV, then opened the specs and bundled DECnet with all their products, we might be using it instead of TCP/IP today! Ah, the glory of hindsight.
At least we didn't end up with SNA.
I never speak lightly. And I'm not a troll. And I never said that I was a Christian.
Obviously, however, I've been instructed incorrectly. Enlighten me. Tell me about how Jesus never had sex and how virginity is a wonderful thing. Tell me how I can save myself. I'm sure my life would be better if someone could just tell me how to live it. Tell me how your life is better, how you know yourself better, from all of your years as a Christian.
You seemed so worked up about all the bad Christians out there. Now that I see that you have a much clearer path than me, why don't you tell me about it? Are you proud? Are you worried for my soul? What about yours? If you could make me a good Christian, would you?
Playing pornographics games during the day is evil! Play at night!
...is the 7-layer theoretical model. It's very easy to talk about network funtionality when you can just say "that's a layer 4 functionality (or protocol)" and it's just understood that it's above IP, but beneath HTTP, and what it does and does not attempt to be.