Slashdot Mirror
Software Update Shuts Down Nuclear Power Plant
Garabito writes "Hatch Nuclear Power Plant near Baxley, Georgia was forced into a 48-hour emergency shutdown when a computer on the plant's business network was rebooted after an engineer installed a software update. The Washington Post reports, 'The computer in question was used to monitor chemical and diagnostic data from one of the facility's primary control systems, and the software update was designed to synchronize data on both systems. According to a report filed with the Nuclear Regulatory Commission, when the updated computer rebooted, it reset the data on the control system, causing safety systems to errantly interpret the lack of data as a drop in water reservoirs that cool the plant's radioactive nuclear fuel rods. As a result, automated safety systems at the plant triggered a shutdown.' Personally, I don't think letting devices on a critical control system accept data values from the business network is a good idea."
Must restart reactor to complete software installation.
[Yes] [No] [OMFG!]
I'd rather it shut itself down then suffer major failure.
Adds a whole new meaning to "Critical Update".
http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
When updating the computer that controls the entire facility, HAVE AN UNDO PLAN!
Personally, I am reassured that these reactors are designed to shut down at the drop of a hat. This is not a situation were fuck-ups should be masked, any discontinuity, however minor, really needs to be highlighted and dealt with immediately.
"I bless every day that I continue to live, for every day is pure profit."
Surely this computer thingy must be the same as my home computer thingy....it always works when I turn it off and on again.
Sure glad the safety systems kicked in as per normal.
As a regulatory wouldn't there be some check and balances to keep critical systems being on their separate ring and not on directly interdependent?
This is beyond incompetence... it is gross negligence.
"Vent radioactive gas? Venting gas prevents explosion. [Yes / No]"
Took this comment seriously, did you?
To me it sounds much more like they have a bad system design if it's impossible to reboot one of the machines / it can't run with one of them offline. Not something which are to blame on the software update (shouldn't such things be expected anyway?)
I guess "software update" can have been used to bash Microsoft a little or something, not that it say windows update, or maybe the poster hates all kinds of software updates?
I wonder if they were using something like EPICS. I worked on a large experiment which used EPICS to control the system. Rebooting a machine would sometimes expose a problem with resources not being freed, eventually leading to a situation where data channels would read the 'INVALID/MISSING' value. The solution, as anyone who has worked on this sort of experiment will know, was to reboot more machines until the thing worked.
(I don't mean to complain about EPICS. It is very powerful and flexible... it's just that the version we used had these occasional hiccups.)
Reminds me of Terminal Error.
did it run Windows?
The only possible interpretation of any research whatever in the 'social sciences' is: some do, some don't
Loose lips lose spit.
It says right in the EULA that it's not to be used in a nuclear power plant!
jX [ Make everything as simple as possible, but no simpler. - Einstein ]
While perhaps the system should be designed to behave differently, what happened here was a good thing. When things went wrong, rather than the reactor systems freaking out and doing random crap, they were properly designed to shift to a known safe state (i.e. Shut the hell down).
I write this type of software for a living so I know that having a computer on the business network connected to the control computers is a risk, bur that risk can be managed. The problem here is that the software update wiped out the nuclear control system data. This exposes two bad problems. First customers are always asking why they can't update their system while it is still running. We liken that to changing your tire while driving down the road. Secondly the software update did not respect the data in the nuclear control system and synchronized it to new initial data in the update on the other system! Not a good idea. In critical safety systems, you always practice an update before actually doing one.
Think about the cost associated with having and maintaining a completely hot-pluggable second control system. How much do you want your power bills to go up to pay for that? And what would be the point?
They have a perfectly adequate safety system that did exactly what it's supposed to do. It read confusing data and decided to shut the reactor down until a human came along and explained things satisfactorily. What's wrong with that? Aside from having the reactor offline for 48 hours, there was no other cost.
"... The move to SCADA systems boosts efficiency at utilities because it allows workers to operate equipment remotely."
Another proof that Homer Simpson was truly ahead of his time.
Are you mad, woman? You never know when an old calendar might come in handy. Sure, it's not 1985 now, but who knows what tomorrow will bring? -Homer
"Don't let fools fool you. They are the clever ones."
The chemical diagnostic data is damn important because it may determine things like corrosion rates and the amount of impurities circulating in the water, potentials for clogs etc... As with all other software, occasionally errors occur, and the appropriate way to respond when it does is to shutdown and blow some whistles as to ensure that the reactor is brought into a safe state before something else goes wrong. This is one of those cases where "Better safe than sorry" is a really rather good motto.
Don't forget about the now mutated sharks living in the coolant water growing frickin' laser beams on their heads.
As our way of thanking you for your positive contributions to Slashdot, you are eligible to disable Slashdot 2.0.
I'm gonna have to agree with that last statement in the summary. Basically under these circumstances, you take out the switch and you take out the plant and I doubt they guard the network closet as well as the reactor core. Plus the whole hacking thing. You really don't need to watch youtube videos and check your e-mail from a control computer and you can bring any actually needed updates and files to it manually via USB drive.
Google's Super Secret Search Algorithm: SELECT @search_results FROM internet WHERE @search_results = 'good'
The summary said: when a computer on the plant's business network was rebooted after an engineer installed a software update
We all know what really happened. Dude rebooted the computer so that Windows automatic update reminder to reboot wouldn't interrupt his Solitaire game every 10 minutes.
You are in a maze of twisty little passages, all alike.
"Don't forget about the now mutated sharks living in the coolant water growing frickin' laser beams on their heads."
Wow! Imagine a beowulf cluster of these!
What exactly do you find frightening about an automatic safety system doing exactly what it's supposed to in response to unusual input?
Correct. It is not the better choice. In the foreseeable future, it is the only choice.
Wow, way to parrot the summary.
"GROSS NEGLIGENCE - Failure to use even the slightest amount of care in a way that shows Recklessness or willful disregard for the safety of others." - 'Lectric Law Library.
Yeah, those bastards, the way they used THE SLIGHTEST AMOUNT OF CARE in designing a system that shuts down in response to unexpected data so as to avoid RECKLESSNESS with the SAFETY OF OTHERS.
Secondly the software update did not respect the data in the nuclear control system and synchronized it to new initial data in the update on the other system! Not a good idea. In critical safety systems, you always practice an update before actually doing one.
I have no problem with a computer on the process control subnet reporting information to a computer on the business subnet.
I have a BIG problem with a computer on the business subnet being able to modify and corrupt data in a computer on the process control subnet.
"I can't dump data to the business side" is a reason to make a log entry and maybe sound a minor alarm. It's not a reason to shut down the reactor (unless the data is needed for regulatory compliance and the process control side isn't able to buffer it until the business side is working correctly.)
But if a business subnet computer can tamper with something as critical as a process control machine's idea of the level of coolant in a reservoir, it rings my "design flaw" alarms.
Is it ONLY able to reset it to "empty" as poorly-designed part of a communication restart sequence? Or could it also make the process control machine think the level was nominal when it WAS empty?
IMHO this should be examined more closely. It may have exposed a dangerous flaw in the software design.
Security flaws don't care if they're exercised by mischance or malice. If nothing else, this is a way to Dos a nuclear plant through a breakin on the business side of the net.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
This is why you keep the IT nerds away from the process network.
I've had a whole plant lose view of it's system because some well meaning retard in IT decided to push updates onto a SCADA system without qualifying the updates....... never had it KILL the control side of things though....well done whoever you were, you've done well.
Burma?
The problem here is that the system didn't shut down because it detected an error in the data collection system, instead it incorrectly detected a problem that did not in fact exist and then proceeded to take action. While the engineer in me is fairly certain that the system is designed to always fail to a safe state (as in, any automatic emergency response couldn't accidentally make things worse - at least not without raising all sorts of alarms), it is still concerning that internal control systems can be so effected by external servers.
In the article they mention that the system wasn't designed for security (since it was meant to be internal) - but this isn't a security issue at all! Any sort of system that relies upon other systems should be designed to assume failure can and will occur in other systems - that is not to say that it needs to verify/evaluate incoming data to make sure it is "good", but rather that it can tell the difference between receiving data (such as current water levels) and receiving no data at all (system failure). Once it has that it can ideally automatically switch to a backup system, or do what it did here and enter a fail-safe state (the difference being that it does so while pointing out the actual problem and not a incorrectly perceived problem in a different part of the system).
Huh? I've read the NTSB report on that accident - and nowhere in it (IIRC) are computers implicated. The accident occurred due to damage to the pipes from construction equipment.
Rereading the report[PDF file] pretty much confirms my recollection, the SCADA system was not implicated as a primary or contributory cause of the accident. The SCADA system was malfunctioning at the time of the accident, but did not cause the overpressure, and 'may' have allowed the operators to relieve pressure had it been functioning and had they observed the pressure spike. The rupture was caused by construction damage to the pipeline and a faulty relief valve.
It's since been disconnected.
Nerd rage is the funniest rage.
Remember when the Slammer worm hit the net a few years ago? There was an article in some defense newspaper I saw that mentioned that they were concerned about power generation and management facilities being hit by the worm. So, from that, I would say it's a reasonable assumption that the facility was running some version of Microsoft Windows (probably NT4 or 2000).
Anthony Papillion
Advanced Data Concepts, Inc.
"Quality Custom Software and IT Services"
And a shutdown, while incovenient, is not a catastrophe. In fact, it speaks well for the plant's safety that it did automatically shut down when faced with bad data.
We liken that to changing your tire while driving down the road.
;}
Oh sure, NOW you think of a debian slogan
Good thing it wasn't written in Smalltalk. The slogan there is building the rest of the boat while underway.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
The thing I'm a bit puzzled about. . . if this system has data which is so important that the whole plant must be SHUT DOWN for two days if it fails, then why aren't there *at least* TWO of them (I'd say there's a good argument for 3 or 4, but. . .)? That way, you can take one out of the loop for updates, verify the update didn't hose your data, sync the data from the 'live' system, then put it online, take the other one offline, and complete the update on it.
If I were the power co owning this plant, I'd be ticked if the plant was dark for 2 days. With the price of energy these days, and the amount of energy a single Nuclear plant can generate, you're talking some real serious cash when the thing is down for 2 days. Especially if I have to look forward to the same thing happening again, potentially every time our systems need updating (not that it necessarily would happen every time, I would sure hope it wouldn't, but with only one system, every update is a potential for the whole plant to go down for some period of time).
No, but I have heard that frogs occasionally live there...
Contrary to the popular belief, there indeed is no God.
Before there are too many retarded "OMG why was it on the business network!!!?LOL!??!" comments, I'll cover that right here:
It says the software is supposed to sync data between the control system and the business network. Obviously it has to be connected to both sides somehow. I'm not a power plant designer, but there's probably a good reason why people might need access to that data from the control system, and thus some kind of system acting as a safe bridge between the two rather than allowing unrestricted access from the business network.
The update f'd up and the control network went "Holy crap where did the cooling water go? Abort!" Everything worked like it was supposed to. The failure was caused by not testing the update in a lab environment before applying it to a live system.
At least it did not turn it into a meltdown, so at least the safety features worked in the software.
That is definitely a glass half full, as opposed to empty.
He's trying to find an opportunity to bash Microsoft!
most freakouts surrounding nuclear power are based on 1960s technology. modern reactor designs, such as pebble bed reactors, are designed to be passively safe. that is, you can just walk away from them, doing nothing, and they will not release gas, go china syndrome, or anything else unsafe. older nuke tech requires active safety management: someone must always be on the job, making sure nothing f***s up. designing safety into nuclear reactor design from the philosophical ground up is the way of the future
intellectual property law is philosophically incoherent. it is your moral duty to ignore it or sabotage it
"In the game of life, someone always has to lose. To me, if life were fair, that someone would always be Oklahoma." -DKR
Agreed. That was good software design to assume a worst-case scenario when the sensors stopped sending in data. The alternative (sending pager alerts or something) would be far worse.
The chemical company I work for has VAX/Unix systems that haven't been rebooted in over four years... and only then because of power outages.
Good enough evidence for me! Microsoft caused a nuclear meltdown! Quickly, to the Blogo-Sphere!
Nuclear power plants run Windows NT?
No wonder we have such a N.I.M.B.Y. problem with them.
At the nuc site I worked at, there were two networks. The business network and the ops network. Data flowed from the ops network to the business network for statistics gathering only. The single thing that the business network did that affected operations and safety (regardless of my boss' attempt to justify budget) was the generation of work-orders. A total failure of the buisiness network would - at worst - result ina routine observation job to be missed which would cause the systems on the ops network to detect a 'fault' and bring the reactor away from criticality.
Yes, a simple software fault can 'shut down' a nuclear plant. These things are designed to 'trip' and shut-down automatically at the slightest thing going wrong. The most advanced and safest Nuc plant in the UK (SXB) does - or at least did - trip once a month or more.
Get a volt-meter that is sensitive to a thousandth of a volt, and allow it to shut down your house when it's input is not ideal. Give yourself three thousands of a volt either way off 'normal' and you are maybe experiencing the ridiculous measures a modern nuc plant puts itself under.
From the summary: If it's monitoring the primary control system then it seems to me like the machine would have to be on the control network. The real issue is why did the primary control system accept a reset from a monitoring system. It sounds like there's more than one bug to track down.
When our name is on the back of your car, we're behind you all the way!
^c^c^c
I have quite a few Windows 2003 servers that haven't been rebooted since August 2006 when we upgraded our computer room to a small datacenter (we went from a single busline and a constantly breaking AC unit to dual UPS's powered by separate generators and dual chillers with separate condensers.) It's not like it's impossible to get good uptimes on Windows, the only servers we reboot on a regular basis are our Citrix servers due to some bad code on Citrix's part that leaks memory over time and our Oracle server due to a bug where 10gR2 pulls time from the deprecate ticks counter (the same one that used to crash Windows9x) which rolls over after ~42 days. Both of those are the result of poor third party coding, not bugs in Windows.
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
I'd go just a bit further and say that it speaks well for the software coders. There are at least three ways to treat any 'out of bounds' condition. They chose to make sure that the safe action was chosen.
An area where that loosely controlled type of team work gets into trouble unless all coders treat data passed to their code, and from their code in the same uniform functional ways.
It also makes me wonder how the code will react to certain malicious software, should it get loose in the facility. If I were writing code to destroy a nuclear facility, it is how data is passed from one process to another that I would definitely attack as well as other vulnerable places.
It is sort of reassuring to have seen a failure result in a controlled shutdown rather than some other, more undesirable action.
Support NYCountryLawyer RIAA vs People
Scenario : System comes up. Things don't work quite right. Some configurations are tweaked and system is now working fine.
Reboot. The tweaked configurations happen to go away. No one remembers which ones they were. The system is b0rked for a while.
I would hope that isn't the case for that system, but I have seen it happen before.
....you can just imagine that like most companies, their business network is all MS windows boxes that also have internet access, so is completely vulnerable to outside hacking too.
If this hadn't have happened it would have probably only been a matter of time before some hacker chanced upon the fact that they could actually control the nuclear facility from some compromised windows box.
Its amazing that these days some sys-admins/network admins still don't get it.
Lets just hope that this incident is enough to get them fired and for the comapny to hire people that know enough to make the system properly secure.
... enter 4, 8, 15, 16, 23, 42.
Or else all hell breaks loose.
Have gnu, will travel.
Yeah, so when a sensor breaks and stops sending in data, it'll keep running like usual, with maybe a small error code in the background. Cause, you know, that's how we want nuclear fucking powerplants to work.
... because then the computer would enter an infinite loop of reboots after the update.
Dohh! You beat me to this!! Well, I'll have a donut and console myself.
http://slashdot.org/comments.pl?sid=573665&cid=23655635
It sounds like the computer responsible for sending data about the water level in one of the cooling tanks had to be rebooted. When the safety systems noticed the lost connection, they assumed a problem with a critical part of the reactor and preformed a controlled shutdown. This is really exactly what should have happened.
Perhaps in the future the safetys could be designed to accept a 5 min. operator initiated override on systems that don't require immediate feedback on failure. Maybe they do already and someone forgot to set it.
If I had a dime for every function, that says: "There is 0 foo", when it really means: "I don't know, how much foo there is", I'd be millionaire...
In Soviet Washington the swamp drains you.
Subnet? It should not even been on the same physical network!
could have been WinME
As they say in reactor operations, "When in doubt, SCRAM it out."
Nuclear engineers build weapons. Civil engineers build targets.
Yeah, sure. When somebody screws up an expression in a way that makes no sense, we should just accept it. In addition, since people on Slashdot constantly misuse pairs of homonyms like then/than, effect/affect, their/they're, we should just ignore historical usage differences and use them interchangeably. We should just accept sloppiness and mediocrity because that's how Western civilization was built.
Then again, maybe intelligent and well-educated people will just ignore people who aren't intelligent enough or who can't be bothered to learn how to properly communicate. The medium is the message, and a badly-formed message says to the recipient either "I don't care enough about talking to you to take the time to say it properly" or "the content of the message can't be that great if I can't be educated enough to learn to express it well enough".
I don't get out of the way for subgeniuses.
Laissez lire, et laissez danser; ces deux amusements ne feront jamais de mal au monde. - Voltaire
You don't think computers connected to a nuclear power plant's control system should be fully patched?
"I've got more toys than Teruhisa Kitahara."
ah, the wonders of try/catch.
"Begs the question" makes perfect sense when used as "raises the question," and it's understood by the majority of those who would hear it. This is my criteria for a functional phrase. Who gives an ass if it also happens to be a name for a logical fallacy? The contextual differences should be enough to immediately determine whether it's meant in the logic sense or as a phrase. Complaining about it is nothing but misguided pedantry.
The problem with your logic here is that using "begs the question" outside of its meaning as a logical fallacy is not actually an error, but a literal interpretation of the phrase. Using it in this sense doesn't actually do any harm to its original sense.
Probably not, they should be airgapped with tight control over access to the network they sit on. I don't like the idea of SCADA systems being on a shared network to begin with. In fact there's speculation that several recent incidents nationwide were due to systems on the shared network being compromised by targeted attacks from China. That may be conspiracy theory speculation but I've seen it discussed enough on serious network security boards that I'm starting to wonder if there isn't some ring of truth to it.
Patching for patching sake is an IT fetish that just as often as not leads to more problems than it solves. In fact the only problem I've had in the last two years that caused any significant client disruption was caused by a bad dat update (patch) to our AV software.
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
Something is not right here...
Yes, the safety system kicking in is "a good thing".
Pulling data from another computer system for a safety related control system is not a bright idea (the weakest link problem).
Historically a safety control system in an Oil & Gas environment, all the inputs to the safety system are either hardwired or pulled from another safety system controller which has the appropriate level of redundancy (CPU boards and communication paths with communication watchdog timers).
Even transmitters in some circumstances can not be trusted hence the 2 out of 3 voting systems (take three transmitters measuring the same value and pick the middle of the three, if one of the transmitters fails high or low your choice will be the safe option).
Someone needs a serious think about where this plant is getting data for its safety shutdown system.
ZombieEngineer
I think you'll find it's civilisation.
It's not a nuclear power plant, but still, my network...
I've set nagios up to monitor my network, and any los of signal is considered CRITICAL, not just a warning, but critical... and I need to know then.
I will not give in to the terrorists. I will not become fearful.
If this was a chemical plant I would be asking to see the HAZOP (http://en.wikipedia.org/wiki/Hazop) reports.
HAZOP studies are serious mind numbing exercises of systematically identifying every possible operational hazard. Should a hazard occur a mitigation action needs to be implemented. The resulting mitigation actions then themselves need to be run through the HAZOP process.
It should be fairly obvious that this is a recursive process and that modern chemical plant designs favor simple, intrinsically safe methods that don't require a complicated control scheme or otherwise the design engineer is condemed till doomsday reviewing the safety of the plant.
ZombieEngineer
I'd guess that there are multiple redundant sensors and that when one goes out of range you'd get an error about bad sensor. In this case where it is actually a software problem all sensors went out of range which points to a real problem being likely so shutting down is the safe move.
https://en.wikipedia.org/wiki/Inverted_totalitarianism
Unfortunately for your argument, there is value in "beg the question" as "raise the question," and language already has moved in that direction.
Just 5 years ago, controls engineers wouldn't breath a word about how vulnerable the world was (and maybe still is). There are special computers called PLCs (Programmable Logic Controllers) that control just about everything in this world from factories to power plants to waste management facilities. They are the brains of all automation. They are also all connected to computers and those computers are all networked on LANS. And in the past, those computers were every bit as likely to get viruses as anyone else's computer. The fortunate thing is that no one who ever wrote a virus ever bothered to write one that would mess with the logic in a PLC. It would have been so easy. In fact, it still is. 99% of all the PLCs in the world are connected to computers in an unsecured fashion. If a virus in a PC were to write randomly into PLC memory, whatever that PLC was controlling would come to an uncontrolled halt. Engineers would never figure out why all the processors were crashing - diagnostics don't exist to monitor this kind of attack. In the days of highly potent viruses like Nimda, it wasn't uncommon for scores of computers that connected to all the PLCs in a given facility to be infected. If Nimda or its kind carried a PLC targetting payload we would have seen disaster much greater than the biggest doomsdayers ever predicted from the millenium bug.
But how did that ISO standard get approved? Microsoft OOXML as a standard springs to mind.
Undetectable Steganography? Yep, there's an app fo
Safety control systems in the chemical industry have been used for 20+ years. These systems have: - redundant CPU modules (which can be hot plugged) - redundant IO modules (which can be hot plugged) - redundant communication systems - self diagnostics (can detect a failed output transistor) - internal diagnostics (CPU voting to detect failed CPU core) - standard algorithms for redundant transmitters Shutting down is the "safer option" however there is still risks (such as thermal stressing pipework). It is a lesser of the two evils problem. This stuff is bread & butter for the chemical industry, there are a number of control companies that refuse to deal with the nuclear industry due to the requirement for unlimited indementy. ZombieEngineer
The system as a whole *did* know the reading was bogus. The control/safety system shut down because it stopped getting "safe" indications from the monitoring/input system. It seems pretty clear that the input system itself correctly logged the reason for the error.
The interface to the control system for the tank level doesn't (or at least shouldn't) have an entire separate "error" parameter -- it probably takes a simple numeric value from the input system.
The input software knows when the reading are bogus or missing. In that case it either stops sending input, which would presumably trigger a watchdog in the control system, or it sends data that indicates a worst-case scenario. with which the control system can do whatever it does in a worst-case scenario.
The control system itself doesn't care why there is or may not be safe input parameters, it only cares that it cannot rely on the input it needs for safe operation. Giving it any more information just adds code and interface complexity to safety-critical software.
Here's the system as implemented:
level = tank.getLevel()
if (level < SANE_MIN || level > SANE_MAX)
level = 0
control.input.set(TANK_LEVEL, level)
Here's the system you describe:
error = 1
level = tank.getLevel()
if (level > SANE_MIN && level < SANE_MAX)
error = 0
control.input.set(TANK_LEVEL, level, error)
The later makes the safety-critical control software more complex, with more test cases and more input parameters, none of which add any value to the safe operation of the control system. The error parameter potentially allows for operation during transient errors, but that's a decision you can make in other ways, without adding interface complexity.
The only inconvenience of the simpler interface is that you have to check the logs from the input device in addition to the control device to determine why the error occurred. And please don't argue that consolidated error logging is worth extra code complexity -- that's probably not even true in a web app, let alone a human-safety control system.
I'm really sad you posted AC, because I'm dying to hear some cool shop stories about nuclear reactors scramming.
You can't put too much water in a nuclear reactor.
You call this a sig?
It is sort of reassuring to have seen a failure result in a controlled shutdown rather than some other, more undesirable action.
You're not kidding!
What?
You can't put too much water into a nuclear reactor
IMHO this should be examined more closely. It may have exposed a dangerous flaw in the software design.
Surely software design is diagrammed, studied and HAZOPed as much as the average P&ID?
That's not what I'm suggesting at all. I'm suggesting that the accidental reset itself could have triggered a critical system shutdown instead of relying on the reset data being out of range. What if the reset fooled the system into thinking some subsystem that was experiencing a worst case scenario was within safe operating conditions? Why a workstation on the corporate network was able to reset this data is another issue in itself.
Now I want to add a "Frequently Begged Questions" list to my site, just to piss off grammar Nazis.
That's the thing, though; it's a misuse of a phrase so much as "kick the bucket" as a literal expression of kicking a bucket would be a misuse of a phrase. The definition of it can be easily achieved by examining the words it uses and their contexts, making it much less likely to confuse a non-native speaker than many other expressions in wide use. The main source of confusion would be if someone tries to make it out to be an invalid phrase.
One of the great things about English is that one can phrase something a million different ways and still get the same meaning; banning the use of one phrase because it happens to also be the name of a logical fallacy is silly and pointless.
I submit for citation the original post. In it is a perfectly valid and widely understood use of the phrase.
There are such requirements in the US, be they for SIL ratings, performing haz-op reviews, etc. Particularly in nuclear apps.
In a plant, not all control systems are SIL rated, but the safety backups usually are....though more and more operators are buying or upgrading to SIL qualified systems and extending SIL to other than just the safety and protection backups.
In this case, the engineers were probably asleep at the wheel and didn't realize the changes they made to the control software impacted the trip & protection systems, so didn't bother to even have a haz-op review prior to making the change to get updates to a control parameter (or set of parameters) from a networked device. They probably figured they were just adding a trim or tuning variable of some kind to the control loop and didn't do ANY real failure analysis.
Oops.
Oh well, time for all the governing bodies like the NRC to get out the microscopes and take a peek at the plant's operating procedures and engineers adherence to them.
Cheers
The details are trivial and useless; The reasons, as always, purely human ones.
What part of FAIL SAFE don't you understand?
The System FAILED. It is programmed to SAFE the reactor when shit happens.
Without its sensors it had no choice but to assume worse case and scram the reactor.
It did it the right way. It did it the way it was programmed to do it.
What would you have it do to determine why it is no longer getting critical data? Send out a droid to check the cat5 cables? Its a frikin computer in a rack, not R2D2.
It worked the way it was supposed to.
Take a step back and let the big boys handle the reactor, Please.
Sig Battery depleted. Reverting to safe mode.
So... you haven't updated the kernel in four years?
The details are trivial and useless; The reasons, as always, purely human ones.
Well, the auditors seem to expect it... as do the vendors when we call for support - "Oh, you say foobar isn't working... well it looks like you're 15 revisions behind; why don't you just fix that and call me when you're done. Oh, your policies state you need to test and certify them? Well I guess I won't be hearing from you for a while, then."
--
.nosig
No data = data out of range.
Data anomaly = shutdown.
Rats chewing on a wire could have been the cause. I don't care. Intelligent adults want the system put into a safe condition when things are out of line. You can always determine the problem and restart the reactor if you shut it down in time.
There is no such thing as a perfect system. But this one comes pretty close.
Sig Battery depleted. Reverting to safe mode.
What do you mean "makes no sense"?
The "controversial" use of the phrase means "requests or invites another question", which I would argue is a fairly logical interpretation of the phrase.
The way you're using the phrase means "taking for granted a principle", which isn't at all intuitive or logical. In fact, it seems to be a corruption of Aristotle's original phrase, "Petitio Principii" ("begs the principle").
So, maybe you should just start using the phrase "begs the principle" instead of "begs the question".
The details are trivial and useless; The reasons, as always, purely human ones.
Becuase it is Fail-SAFE a term you seem to not understand here. No data = very very very bad data, so the system did exactly what it should do which was put the reactor int the safest mode it could. Anything else is added complexity, in thsi case what was very likely a PLC expects a stream of data from this computer, it stopped receiving this data so it did the only sane thing it could do, it shut everything down as fast a possible.
I may agree with what you say, but I will defend to the death your right to face the consequences of saying it.
The problem wasn't caused by some sensor not sending data, it was data in the database being actively reset on purpose by a bit of software on another machine. This wasn't a case of "if some data anomaly occurs, shut down" - the article suggests this was a case of "some data anomaly occurred which happened to make the system think there is insufficient water" which caused a shutdown by chance. The data anomaly caused by the reset might not have fooled the system into thinking the water level was out of range.
It doesn't really matter in this case if the operation system is looking at plant data from a minor monitoring system. What is troubling here is that it's completely reliant upon this minor monitoring system. If this box someplace is so important as to cause a emergency shutdown in a nuclear power plant then one would think there would be a backup system that comes into place when the primary monitoring system goes down. Did they think this box would never have a hardware failure? That it would last forever as some kind of cosmic perpetual motion machine? I am very worried that operations management systems like this even get implemented in high security and important locations such as a nuclear power plant. Looks like it's time to higher a better and more intelligent Information Systems and Network Manager.
I think you're missing the real point, which is that the central safety systems are being fed data from a 'business network'. What would happen if that computer had an issue that caused it to send the same data continuously even when the coolant level had really dropped? WHY are any safety systems receiving data from an insecure network?
It's bad enough that most reactors use regular PC's to do the data collection and reporting, given the security risks posed by such systems (especially if networked), but I never realized they would be so stupid as to feed data in the other direction like this!
I disagree. In this case the failure mode of the error was not foreseen, and while the code reacted in a safe way, there is no reason to believe that it was specifically coded to do this. From the description it appears to me that the system requires continous input of coolant levels, without that input the level appears to be zero (a continous input of 0). What would have happened if the software upgrade had caused the same levels to be reported continuously? A real drop in coolant might not have been detected! Not foreseeing a failure mode in which a data-provider becomes corrupt seems like a serious problem to me.
I think it makes sense to shutdown when you cannot verify that your data is 100% available. The software did interpret "Lack of Data" as exactly that "Lack of Data" and shut down accordingly.
I think you're missing the real point, which is that the central safety systems are being fed data from a 'business network'. What would happen if that computer had an issue that caused it to send the same data continuously even when the coolant level had really dropped? WHY are any safety systems receiving data from an insecure network?
It's bad enough that most reactors use regular PC's to do the data collection and reporting, given the security risks posed by such systems (especially if networked), but I never realized they would be so stupid as to feed data in the other direction like this!
Obviously you have -zero- experience with power plant networks. Allow me to enlighten albeit anonymously.The reason machines like this receive data from networks that could be considered 'less secure' is because telemetry is required from a multitude of sources to actually ascertain any useful realtime information. Aggregation machines have to speak many different protocols and translate between them while communicating with other machines that belong at other plants, cities, states, and even companies to effectively get an accurate picture of the entire grid's current conditions.
The world of plant control machines themelves is very vendor-driven. Most facilities have turnkey solutions brought in by the few major players in this field. ABB, Hathaway, GE, etc. Those players don't even use the same SCADA protocols. Some use ICCP, some use DNP, and others prefer Etherpoll. I've seen RS232 data encapsulated into everything from fully-meshed TCP connections via OSI-Soft's PI to barely encoded into modbus and slapped onto ethernet with only an understanding of ARP.
The solutions are required because electricity is not just one powerplant pumping watts blindly. Instead, you have a multitude of plants all pushing power onto ISO-controlled grids that all have to work in concert with each other. This requires -- yes, you guessed it -- networking! The world of plant networks is pretty complex despite the hype you see in the media. The business of making actual watts appear magically at your house at a nice, consistent 60Hz is vastly more involved that most people realize.
Telemetry comes from secured networks, business networks, and other companies and controlling agencies. That is how it works. Period.
If you are actually interested in seeing the way these are regulated to be secured, the information is cleverly hidden in plain sight at the NERC website.
"Personally, I don't think letting devices on a critical control system accept data values from the business network is a good idea.""
Yeah, well that's why you obviously wouldn't succeed in business. You can't seem to grasp that things like time-to-market and pleasing focus groups are far more important than piddly little things like that. Geez.
Oh, you're not stuck, you're just unable to let go of the onion rings.
yes the system ultimately made the right choice, shutting down with a perceived loss of critical information.
however, this was a best choice response to a poorly engineered shutdown system.
a properly designed critical shutdown system would have completely independent sensors, for exactly this reason. by design, no external system (i.e. business network data collection) should be able to compromise the integrity of a safety system in any way. Safety systems are designed to be redundant within themselves on many levels so that even if some link in the chain were to fail, there's another link waiting to take it's place until repaired. Business systems, and often standard control systems, do not have that sort of availability/reliability, and so should have not part of the safety system.
Well they just might have been running a modern OS.
Yeah, those bastards, the way they used THE SLIGHTEST AMOUNT OF CARE in designing a system that shuts down in response to unexpected data so as to avoid RECKLESSNESS with the SAFETY OF OTHERS. And to top it off they had the gall to report it instead of covering it up.
One needs an odd number of monitoring systems, so two will not suffice. If there is only two and they both report two different things then one still has to shut down the plant until things are sorted out. In fact, now that there is two systems, there is twice the chance of something going wrong. If there are three systems then the majority wins. There is still the problem of what to do with the bad system as hot swapping a new one in has the potential to bring everything down. It would also be a nightmare to test all the failure conditions. For example, one of the early shuttle launches was scrubbed because when all the redundant computers tried to synch up, the clock signals edge appeared and it had only been designed to deal with the high and low states of the signal and all the testing never encountered this condition. Redundant systems can also give a false sense of security if they are not maintained independently. For instance, if someone makes the same mistake to all of them or a batch of defective parts is used on all of them then they can easily all fail for the same reason within a short timespan of each other.
Why is it every time anything goes even slightly off-optimal in the nuclear industry it's news? Every day thousands are injured and killed in and by other industries and nuclear power just keeps on quietly pumping out the mega-watts but if somebody sneezes in the control room everyone in the world knows about it.
With all the updates that came out for VAX/Unix in the last four years?!?
Haven't they ever heard of redundant systems? I would have thought that having more than one controller on vital equipment was obvious. Of course, there is another kind of redundancy that might become relevant for the responsible engineer; although I am not sure I think the guy should be fired - knowing how finances trump security, safety and common sense in most companies, he probably wasn't given the resources necessary.
um...
I swear I don't have anything but a trolling myspace page but:
Me Too!
That line is *too* good.
whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
You just rephrased what he said more eloquently. :)
Business. Numbers. Money. People. Computer World.
"And then it rebooted" What? But that means.. Oh my god, please please do not tell me that in our nuclear power plants they run WINDOWS on critical parts of the system? Are they f...ing stupid? It is even dumb enough to run it at home, when you want a stable system, but there.. I try to ask our plant what they use and if I get the same answer I really considering to move my home to the place where I am most far away from it (and another). :(
Ubuntu, a terminal, Python and Slashdot. Thats all you need.
In French, and maybe in American English, but I live in Western Canada and was writing in Canadian English. In that context, it's civilization.
Laissez lire, et laissez danser; ces deux amusements ne feront jamais de mal au monde. - Voltaire
The point is that it's not a person that begs the question. It's the situation or the argument.
Let me be more explicit. In either interpretation, be it classical or contemporary, "one" never begs the question unless you're talking about circular theorems regarding numbers to the power of 0.
Laissez lire, et laissez danser; ces deux amusements ne feront jamais de mal au monde. - Voltaire
see this post
Laissez lire, et laissez danser; ces deux amusements ne feront jamais de mal au monde. - Voltaire
How was it 'left to chance'? If it doesn't know what the water level is it doesn't knwo that it's dangerous, but it doesn't know that it's safe either. Seems it erred on the side of caution which is correct behabiour, at least to anyone with a clue.
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
Don't forget the gun.
Patents Drive Free Software as Hurricanes Drive Construction Industry
In any case I don't see the relevance of that to this discussion, which boils down to this: do you assume no news is good news, or bad news?Well that much is true.
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
While that is interesting, it doesn't really explain why a PC is feeding water level data to the safety system. Surely the water level isn't something other power plants need to know? Unless I'm missing something, the water level sensors should be directly connected to the safety systems and not through an intermediary PC.
Actually, it's BRITISH English that uses the "s" - American uses "z" instead.
For a site about things like basic rights, Slashdot users sure do like to censor "dissent".
Uhm, some reactors are cooled with liquid sodium, and trust me, you CAN put too much water into those ...
Potential cost: Only that the business-critical system they're controlling stops working. Nothing important, y'know.
Potential benefit: Well, if they're on a very isolated network and access to the systems is already very tightly restricted, that's a good question. A lot of Vax systems never even had TCP/IP installed - you use a serial console to connect, which by definition restricts how easy it is to get to the system in the first place.
It's not a question of if the shutdown was the right think to do. It definitely was, the system thought there were a problem, it shutted down.
The problem here is that the system acted on corrupt data thinking it was the right data. It could (almost) as easily keep running when it should fail, if the opposite data was fed in a cooling failure situation.
_Renewable energy_ is the way of the future. Cue nuclear energy fanboys flames in 3.. 2.. 1..
But it will be when Windows 7 comes out.
Slashdot. Unreadable news to annoy nerds. - wonkey_monkey
According to the article a business network computer reset data on the control network, thus the safety system shut the plant down. The control system was not relying on data from the business network, but it was possible to alter control system data from the business network, which is the part that is bad.
Later in the article it is noted that all physical connections between the business and control networks have been removed, which further supports the idea that the control side doesn't rely on receiving data from the business side.
And any error logic to deal with bad field devices or high tank levels or whatever else probably looks more like:
|--[I:3.0 > 62]------------------(O:4.4) alarm---------------|
|--[/]I:3.1------(O:4.5) permissive to safety system-|
Or as best I can ASCII draw ladder logic with this stupid fucking lameness filter.
"Sacrifice for the good of The State" - The State
Why are they running a nuclear reactor on Windows?! Are there any other operating systems that automatically reboot after an update?
You are quoting a line from a Washington Post staff writer.. how do you know that he's an expert in Nuclear Powerplant IT? Does he really know why the safety controls shut the plant down? Maybe when the IT guys were explaining it to him over the phone, they used poor terminology and the writer in turn interpreted it as being in error.
I don't see why people are arguing over this.. one of two things happened:
Come to think of it, that second situation doesn't even seem to be all that "incorrect;" if they truly hadn't thought of the possibility of this happening when doing the coding, then the software still failed-over in the safest manner possible.
Aikon-
so your systems don't have most of the updates install?
These days, the "wrong" usage is far more common then the "right" one. English doesn't have a central body that defines it (unlike some languages), it's defined by how people use it, so if more people use it incorrectly than correctly, the incorrect usage actually becomes correct.
Someone tell me how I can protect myself and my family from stupid fucking people, please?
----
"We regret to inform you that we've melted half the planet because we were stupid."
In my experience, the situation is almost always more complicated than what a journalist reports. Having no information about how the sensors were setup, what the hardware involved is like, I would say the software reacted as it should have. We don't know, it might have even reported that it is getting no data.
But despite what the reason for it is, if Water Level=0, shut the plant down, is a safe baseline to go by. The shut down sequence was probably a separate module in the system, independent of what other modules in the software are doing. Its only job is a watchdog, assume the worst and shut the plant down. The engineers can figure out why that happened later...because this way the plant wont turn into a hot radioactive carnage-dump.
Essentia non sunt multiplicanda praeter necessitatem.
All nuclear power plants have simulators to train the operators. They should all be required to test all changes and software updates in the simulator environment before installing them on line.
That implies that the scope of the simulators may need to be expanded. In addition to training operators, they should become a sandbox for testing any and all things connected directly or indirectly to operational systems.
If business systems become interconnected with operational systems, then the business systems too must be replicated in the simulator environment. That might become a very onerous requirement, but that very difficulty could have a benefit. Architectures that prove to be very onerous to duplicate in a simulated environment, should be rejected and redesigned.
"Part of the challenge is we have all of this infrastructure in the control systems that was put in place in the 1980s and '90s that was not designed with security in mind, and all of sudden these systems are being connected to [Internet-facing] business networks" said Brian Ahern, president and chief executive of Industrial Defender Inc., a Foxborough, Mass.-based SCADA security company
No, the problem is putting 'computers' on the Internet that were most certainly designed with security in mind, something the 'computers' most certainly fail at. To put in bluntly, running your SCADA units on Windows over the Internet is the dumbest thing I ever heard of. And that they are still running such designs five years after the great blackout of 2003, demonstrates incompetence and neglience boarding on the criminal
davecb5620@gmail.com
"Good enough evidence for me! Microsoft caused a nuclear meltdown! Quickly, to the Blogo-Sphere!"
.. include explicit mention of some unknown 'computer problems' at FirstEnergy, the Ohio utility thought to have triggered the regional power failures, in those preceding hours"
That's only funny if it wasn't even partly true. But here's something really funny:
"The Slammer worm penetrated a private computer network at Ohio's Davis-Besse nuclear power plant in January and disabled a safety monitoring system for nearly five hours"
"TRANSCRIPTS of telephone conversations between utility operators
davecb5620@gmail.com
"This is why you keep the IT nerds away from the process network"
.. :)
What's a 'process network' and who exactly do you get to fix your 'process network' ?
"I've had a whole plant lose view of it's system because some well meaning retard in IT decided to push updates onto a SCADA system without qualifying the updates....... never had it KILL the control side of things though....well done whoever you were, you've done well"
Assuming the above anecdote was even true, such an incident would never occur on a live system, and I'll tell you for why. You never update a life system - got that - never. At least in any competently run IT department.
What I suspect happened in the Georgia nuclear power plant was that some automatic patch process broke the 'computer', the rest of the story is just so much smoke screen.
re: 'qualifying the updates': Just how exactly do you qualify an update. Is an update the same as a patch or a bug fix. What motivates you to apply such qualifying updates. I mean if the computer ain't broke, then don't fix it.
If it's security updates then why bother, I mean if the 'process network' is secure you wouldn't need to. I would have thought they used end to end gateways running on embedded hardware, providing a VPN connection to the SCADA units.
But then again I only ever provided IT services to the double glazing sector, and what do I know
davecb5620@gmail.com
Patching is entirely appropriate in many situations, but SCADA systems are certainly NOT one of them.
A public facing server needs to be patched frequently to prevent the 'sploit of the day. It may cause other issues, but they won't be as bad as having your server turned into a spambot of online casino scam behind your back.
As you say, SCADA systems should be protected by an airgap. At most, a translator system should provide read-only data to the business side. In that case, any patching is a bad idea. Long term stability and having everything using exactly the same very well tested version is a far batter strategy for avoiding problems. Rollout of a new version is (and should be) a long process of careful testing. Certainly not something to do monthly sight unseen.Do learn the difference between an example and a definition.
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
Loss of coolant is perhaps the most serious event that can happen to a reactor. That's where meltdowns come from. That is one of those systems that requires immediate feedback, and so shouldn't allow an override.
Had the reactors at Chernobyl not allowed manual overrides like that, nobody would know that name today.
The mistake was allowing a system that important to be affected by a machine in the BUSINESS NETWORK (that is, a plain ol' LAN in a cube farm). This is exactly the sort of thing that can happen when you connect the red network to the black network.
Ideally, the data should have been gathered by a machine in the control network and published to a machine on the business network via a one way serial connection (That is, the Rx pin on the control network machine is actually not connected).
Actually yes you are correct. You would need at least three monitor sensors for each data value. However for the computer reading data from these sensors you only need a single backup system. It should be high enough priority to replace and bring online the original computer system in at most a 24 to 36 hour time frame. If you were very paranoid then yes perhaps three computer systems might be needed. However to triple check data values you only need an odd number of sensors not necessarily an odd number of computer systems who read the values of the sensors. Good eye however in pointing out the odd number needs.
Yes, because clearly they want to be using a closed-source virtual machine reading all of their mission-critical code in ways that can't be predicted.
It's better to vote for what you want and not get it than to vote for what you don't want and get it.
- E. Debs
I saw UAC, I was thinking ID's DOOM game... but-- Microsoft UAC-- much better fit for the situation...
every day http://en.wikipedia.org/wiki/Special:Random
There are several measures of feedwater flow and reactor water level that a chemical monitoring program should never have been able to override.
I am a name troll of Westlake. Visit my homepage to learn why.
You're right I don't have professional experience. I'm not entirely without knowledge though. My University thesis for Software Eng. was to code a nuclear power reactor monitoring system. So, yes I do have some inside knowledge of their workings from all the research I had to do.
While your explanation of the inter-networking systems and protocols was all very interesting, it isn't really relevant to what I posted previously. I already know that the monitoring and reporting systems are highly networked, often using 3rd party software and hardware (writing this kind of software was what my thesis was about!).
The problem as I see it, is that the localized safety system at the plant was receiving data from what amounts to an insecure terminal. Now, I know that there are other back-up systems acting redundantly, but as I see it, the fact that there is a failure mode that depends on an insecure server is a threat that doesn't need to exist.
Ah, didn't realize they'd fully open sourced it... last I heard they had only open sourced portions of the libraries. And I didn't know about that last bit at all.
It's better to vote for what you want and not get it than to vote for what you don't want and get it.
- E. Debs
Actually, this was a fail-safe incident. Something in one of the monitoring systems screwed up - resetting data. in this situation the only logical safe thing to do is shutdown, because you no longer no what the real state of your system is.
Example: 3 Mile Island had a water sensor in a drain trap (yeah I know BRILLIANT). This sensor is the one the engineers were reading to "know" they had water in reactor. Meanwhile all the water boiled out due to a jammed pressure relief valve. Had the engineers bothered to check one of the other water sensors earlier, they would never had been within 45 minutes of a total and complete meltdown - far worse than Chernobyl. So, I'm rather glad that this reactor took the Human element out and forced them to look at more than just the one gauge they look at, because "that's the way we've always done it".
Actually, it interpreted a critical parameter perfectly. It reports an existent water level iff there is adequate justification (sensor data) to draw that conclusion.
In the absence of evidence that there is adequate water (the safe condition), it assumes that there is not.
By the same token, in a situation where water present would be bad, the system assumes there IS water unless it has adequate justification to accept that there is NOT.
That in a nutshell is failsafe design.
So you haven't been installing the patches. I count reboots in October, September, November, December, January, February, April, January being particularly interesting as the TCP/IP stack was updated. Care to share your IP address? Enquiring minds want to know.
meh
Ah, it would seem that's a rare case where the Canadian spelling is like the American spelling and not the British then. Doesn't change that daybot's attempt to correct the corrector fell flat. You had better success; thanks.
Laissez lire, et laissez danser; ces deux amusements ne feront jamais de mal au monde. - Voltaire
When questions start handing out change to people, I'll start using the expression like jo42.
Laissez lire, et laissez danser; ces deux amusements ne feront jamais de mal au monde. - Voltaire
"A process network is the network the SCADA/DCS system and it's physical controllers sit on, usually segregated from corporate LAN"
You do seem to have implimented a solution, which begs the question as to why the rest of the power industry haven't also done it. See here where their still using the Internet to relay SCADA data. They obviously don't use your methodology. Just who do you work for again?
"the vendor usually checks out stuff like windows updates, and assesses the impact in the system"
How do you checkput a service pack without installing it on a live system, when a service pack breaks something, the usual solution is ti reinstall, reinstall, reinstall. Again you do seem to have thought of a solution that would be of use to the rest of the IT industry.
"Updates are usually installed to fix/improve system operation"
My understanding is that service packs are provided by the software vendor in response to general issues, and not specifically to correct problems in a specific installation. In nix land, if there's a bug then you can directly contact the programmers and get specific solutions to your problem. I guess it's a different mind set.
davecb5620@gmail.com
"When somebody screws up an expression in a way that makes no sense, we should just accept it. In addition, since people on Slashdot constantly misuse pairs of homonyms like then/than, effect/affect, their/they're, we should just ignore historical usage differences and use them interchangeably. We should just accept sloppiness and mediocrity because that's how Western civilization was built."
Your statement needs qualifying.
Solar. Wind. Hydroelectric. Geothermal. Four other choices. These are available but do not produce the sheer quantity of energy as a nuclear reactor. Fusion will, theoretically, but it is stil in the theory stages.
Though you are right, the economical, low carbon emission choice in practice now is nuclear.
TANSTAAFL GIGO Acronyms to live by!
I am afraid that I fail to see the irony since spelling civilization vs. civilisation is not an issue of homonyms (words with different meanings that sound the same) but instead one of regional spelling differences for a single word, and I spelled it correctly for the region in which I live.
Laissez lire, et laissez danser; ces deux amusements ne feront jamais de mal au monde. - Voltaire
Unlike most posters here, I have actually been a Reactor Operator, albeit many years ago. Yes, you want any questionable signal from primary systems to initiate a scram. But I question why any primary system was connected to or using data from such a questionable source. In my experience, all such sources were hard cards - no software involved. Obviously "modern" plants have become too modern for my tastes.
For those who wonder what a scram is - it's from the early early tests where the rods were actually pulled by a human tugging a rope attached to a pulley. Once pulled, the rope was tied off, and he stood buy with an axe. If the pooh hit the rotating blades, he chopped the rope. Super-Critical Reactor Axe Man = SCRAM.