Slashdot Mirror
Software Bug Halts F-22 Flight
mgh02114 writes "The new US stealth fighter, the F-22 Raptor, was deployed for the first time to Asia earlier this month. On Feb. 11, twelve Raptors flying from Hawaii to Japan were forced to turn back when a software glitch crashed all of the F-22s' on-board computers as they crossed the international date line. The delay in arrival in Japan was previously reported, with rumors of problems with the software. CNN television, however, this morning reported that every fighter completely lost all navigation and communications when they crossed the international date line. They reportedly had to turn around and follow their tankers by visual contact back to Hawaii. According to the CNN story, if they had not been with their tankers, or the weather had been bad, this would have been serious. CNN has not put up anything on their website yet." The Peoples Daily of China reported on Feb. 17 that two Raptors had landed on Okinawa.
"I'll assume they're adding a day"
... just as well it wasn't in the fire a mission flag byte ... or worse still, the ejector seat flag byte.
Unfortunately, its software sounds like it could have stored that day byte (or some other data), somewhere is shouldn't have
There are 10 kinds of people in the world... those who understand binary and those who don't.
As far as I remember the Space Shuttle not only has redundant computer systems, but also redundant software, i.e. the software has been developed twice to ensure that software bugs don't cause a catastrophe. I'd prefer to know that systems capable of carrying weapons which can kill hundreds of thousands of people were designed with the same safety in mind.
memomo: free web based language trainer DE-EN-ES-FR-IT
The Bismarck battleship had a bug also: when the main turrets would fire, the aiming radars would be disabled. That's no joke when you're in the midst of a battle and everyone of those large caliber shells counts. As I understand, the radars would be disabled by the vibrations of the turret cannons firing. Not a software bug, but bug nonetheless, and you do wonder how did this battleship pass testing.
"The agriculture ministry is not in charge of Gundam" - Japanese ministry official.
I really doubt such an advanced (and stealth) aircraft would have any "traditional" radio capabilities that could easily be intercepted. If the encryption is written such that the position of an aircraft matters, they may have no communication channel at all.
;-)
That said, I'm not sure how this bug would have escaped QA. I mean, it's an airplane. Hundreds of commercial jets fly over that line day in and day out, as do other American military planes. I wonder if the bug also exists at the Prime Meridian?
I hate to imagine what the software patch process is like on a jet. I doubt you can just ssh in and run an svn up
Yes. When Jews cross the dateline, we don't know what day it is anymore, much to the consternation of those of us who may need to travel to Japan or Hawaii.
When F16s crossed the equator, the computer would roll the aircraft 180 degrees and fly inverted:
http://catless.ncl.ac.uk/Risks/3.44.html
With the inertial navigation systems I work with, time stamping of data is very important. Clocks that are accurate down to nanoseconds aren't uncommon, synching with GPS 1-PPS signals (1 pulse per second) to determine and correct clock drift per inertial sensor read cycle, etc. Timing systems are usually custom built for the product in question as part of the design.
There are only 10 kinds of people in this world... those who understand binary and those who don't
I tried posting this on several sites but on March 11th, when the new daylight savingsregime kicks in for the first time there will probably be a lot of Java applications that will start having data issues because the latest Java version IS NOT BACKWARDS COMPATIBLE for several three character time codes that have bee removed. Several codes have been deprecated in a way that is not backwards compatible. I could be wrong about the severity, but for he last two weeks my software team has been dealing with this issue and the interaction between Oracle and Java.
"God fights on the side with the best artillery." - Napoleon, Marshal of France - speaking truth to power
You forgot one major issue with the F22 versus the Eurofighter. The Eurofighter & every other modern fighter in the world for that matter can't see the F22 on their radars. F22 is full-up stealth (assuming no external stores).
In wargames held in the US with 1 F-22 versus 5 F-15's. 5-0. The F-15 pilots never saw the F-22. Not a fair fight - but then that's the idea.
Have you compiled your kernel today??
Actually the whole story is a lot more exciting that 5 on 1...
The 27th Fighter Squadron (8 F-22s) at Langley AFB, Virginia fought against 33 F-15Cs and didn't suffer a single loss. The F-15's again didn't even detect the F-22's until they were all locked and targeted.
Then some months later during Exercise Northern Edge F-22's reached a 144-to-zero kill-to-loss ratio against F-15s, F-16s and F/A-18s. Only 12 of the F-22's accounted for nearly 50% of all kills for the Exercise.
OK - as a software development major, I have issues with this statement. I'm assuming that this is just shooting from the hip on the part of Don Shepperd, which is understandable. OTOH, if he was somehow involved (which I can't tell from the article) in the project and is stating this as a fact, then I have a bit of a problem with this. I don't know the architecture of the chips they use (motorola, or RISC based, I'd assume), but this seems like quite a bit of code. Once you get this much code, unless it is broken down over the various subsystems very well, you are asking for trouble. The fact that they fixed it in 48 hours might be due to a 'blackbox' log or some way of knowing exactly where the code dumped or that the code is much smaller than Don's saying. Anyways, a few million lines of code is very hard to maintain, (the degree of difficulty is probably a function of the language) and seems like a bit much for a fighter's onboard computers. Even if we take into account the code for handling all the geometry, HUDs, and systems monitoring, it still seems like a lot. Windows 3.1 was 6 million lines of code, according to wikipedia.
I'm only speculating since I have no idea the depth or complexity of the onboard computers. Does anyone have any experience with anything like this? Does this seem like a lot to anyone else?
If I mod you up, it doesn't necessarily mean I agree with what you've said, sorry.
from tfa...
Lockheed Martin is rushing a software fix to Hawaii after 12 US Air Force F-22A Raptors en route to Japan for the stealth fighter's first overseas deployment had to turn back because an unspecified problem with their navigation systems.
well, THAT patch hasn't had much time for a burn-in/test period. how comfy would you feel flying with that in place?
-r (*shudder*)
-'fester
What you state about Airbus is absolutely correct but FADEC stands for Full Authority Digital Electronics Control but many seem to remember it as E = Engine, like you do. As far as the fly-by-wire system is concerned, I might add that it has already saved at least 300+ lives - an Emirates A340 attempted to rotate with insufficient airspeed at takeoff (but past V1 so they couldn't stop either) and the FBW system stepped in and throttled up (fortunately autothrottle was on so it was permitted to do so) and rotated as soon as the aircraft had sufficient speed to take off instead of just lift and stall (and consequently crash). Emirates training got a slap on the wrist by Airbus since the crew apparently had the attitude that if there's a problem, the computer will sort it out whilst the correct procedure is to either perform maneuvers properly manually or tell the computer what you want the aircraft to do and then monitor it - even though the computer can do a lot to correct crew errors, crews shouldn't perform poorly just because it can do that. I remember an article posted on airdisaster.com in which some first officer that wanted to remain anonymous (for obvious reasons) wrote that due to the software that outperforms any human pilot, many captains he had flown with had definitely ignored the rule that whilst you should let the autopilot land if weather conditions are extremely bad, you shouldn't force it to do so if you couldn't land in those conditions yourself too because how can you judge what the limits of the autopilot are (and how close to those it is) when it is already outperforming you? I wish Boeing adopted the same design philosophy now that they've finally switched to FBW too with the 777 - there's simply no justification to let a fly-by-wire aircraft stall due to pilot error when the system could easily be programmed to prevent it (not to mention detect better how close to stalling an aircraft is instead of just giving the pilot a list of stall speeds at certain configurations).
All complex systems have bugs that need to be ironed out....
. html ...
http://www.atimes.com/atimes/Middle_East/IB10Ak05
"Keys notes, however, that the electronic spectrum around Baghdad is polluted by the myriad jamming devices that coalition forces primarily employed to thwart remote detonations of the improvised explosive devices that have inflicted 70% of all US fatalities in that war."
"The potential problem was discovered when the first F-22s were operating near US Navy ships off the Atlantic coast. Navy radars overwhelmed the F-22's automated sensors. Even now, larger, multi-station, purpose-built electronic-intelligence-gathering airplanes encounter difficulties around the Iraqi capital because of the extreme density of jamming devices."
Here is an interesting question.
In a fight between 1 F-22 and 2 EuroFighters, who would prevail? If the F-22 prevails, then the F-22 is an excellent investment.
However, the United States Air Force has never claimed that 1 F-22 can beat 2 EuroFighters. I suspect that the 2 EuroFighters would reduce the 1 F-22 into a pile of smoking rubble.
Mind you, the EuroFighter may greatly outclass an F-15.
I talk about stuff.
That machine handles and fights so differently that when the engineers flew simulated combat against Air Force pilots, the engineers won. The first few times. As soon as the pilots got the hang of it the engineers were toast.
If the Eurofighter's radar can't detect the F-22, multiple Eurofighters won't be any great advantage compared to one.
Any sufficiently advanced technology is indistinguishable from a rigged demo
--Andy Finkel (J. Klass?)
Yeah... uh... you know, in the same way they simply do not and never would have the navigation system connected to the In-Flight Entertainment system in an airliner, likewise they would never slave the ejection system to anything other than the mechanical operation of that yellow handle between the pilot's knees.
As for missiles? First, they fly unarmed on ferry missions because ammo is dead weight that reduces range; and second, even if they were armed, what do you really think would happen if an AMRAAM missile was free launched without being turned on, much less having had targeting info downloaded? Drop like a stone, it would, right into the pacific. Bloop. All gone.
Say it's also a good thing water isn't flammable, otherwise fire trucks would show up to fires and only make the situation worse, right?
If a job's not worth doing, it's not worth doing right.
Anyone else remember the incident where the US Navy tried to automate a warship with Windows NT and it crashed shortly after leaving port... Divide by Zero error left it dead in the water... Software glitches leave Navy Smart Ship dead in the water The US Navy's so-call "Smart Ship technology" left the Aegis missile cruiser USS Yorktown dead in the water off the coast of Cape Charles, Va. for several hours. The shutdown of the ship's propulsion was credited to a database overflow in a Windows NT system. The crash was caused by the inability of the OS to properly handle division by zero. Said Anthony DiGiorgio, a civilian engineer with the Atlantic Fleet Technical Support Center, "Using Windows NT, which is known to have some failure modes, on a warship is similar to hoping that luck will be in our favor." The Navy is still expected to spend $138 million expanding the "Smart Ship" program to the entire Aegis class, and to other ships in the fleet. (Government Computer News, 13 July 1998)
Luckily they found it during simulations of the F-16. A bug in the fly-by-wire software caused the plane to think that it was upside-down whenever it crossed the equator. It would try to correct the problem immediately -- A maneuver that the plane could probably survive, but that would probably kill the pilot had it occured in real life.
Free Software: Like love, it grows best when given away.
Yep, the "F"-117 is actually a small bomber, but "B-117" just doesn't have the sex appeal that "F-117" has. Cognressmen are stupid, and the contractors and the Air Force wisely names teh craft f-117 so they craft could get secret funding in secret appropriations hearings. More ninja like.
The F-117 just went in and knocked out air defenses and communications and maybe a few really crucial surprise targets. That's really its role. Then the bigger bombers can fly in (more) safely.
But it's no fighter. I know an F-117 shot down that President in season 4 of 24, and I've shot planes down in my F-117 on my playstation, but no F-117 has actually shot anything down in real life. The F-22 is so superior to any other fighter because it actually is like a ninja, unseen. The new russian and european fighters are more maneuverable, but they really have no chance when they learn of the F-22 by the missile that's closing in. that stealth and supercruise are worth what we paid. I just hope we don't sell these planes to the extent we have in the past (Iran still has a few F-14's, a very lethal interceptor and one reason we might need a "counter-interceptor").
May not help much in Baghdad, but it makes a difference at the diplomatic table. China, Russia, Iran, France. None of these countries really want to fight the USA, but this sort of thing keeps that proposition safely moot. And that superiority probably saves lives. The US has a lot of detractors in the world, and I imagine our constantly new abilities have prevented at least some conflict.
Keep in mind that these fighters are often tested in comptuer simulations together, and the US has a long standing tradition of drastically cutting back its capabilities. I recall one international military simulation I took part of (not airforce, but Army). and the range of what I did was cut back ridiculously. Nearly 90% reduced range. It much harder to win, I guess.
Also remember that experience counts. The US Air Force has experience rivaled only by Britain. Our Pilots in our planes would simply not make the mistakes other nations would. When will the eurofighter be flying over the international date line? Never, probably. The US is stretched throughout the world and fighting some hard fights every day. Maybe it's a lousy policy in general, but we work our kinks out of our weapons a hell of a lot faster than most others.
The F-22 is more than stealth. It has the capability of obfuscating radar signals in general, making it difficult for enemy fighters to use their radar at all. They don't know whose their friend or foe, they don't know that a missile is approaching, etc. Few really know the full maneuverability of the F-22, and it has more control surface than the eurofighter or almost any other aircraft save drones so the widespread stories of its low maneuverability are probably not confident stories.
I smell a lot of "likely" coming off this thread - mostly from Amurricans justifying the F22.
Has anyone seen the results of exercises btn the F22 and Eurofighter? I thought not. Most of the combat exercises people have mentioned have been btn F15s and F22s, and even then under test conditions. Give it a more agile opponent, the F16 or a more modern opponent, and a mixed-mode operation.
Remember, expensive is not always better, specs don't always relate to combat. Interesting that the Eurofighter's turning circle is tighter, but the ability to sneak up is good. Costing less is also good, as are the training costs. With extra fuel and more weaponry - always an addition in war - I reckon the stealth capabilities will be shot. I suspect there will be difficulties with maintenance as well, particularly with repair facilities operating at a war-time standard, sometimes %50 of peacetime. Stand-off is *BAD* as IFF is always assumed to be good - which it never is - so the F22 could end up a friend-killer if used as stand-off. Politicians always see missiles as a cost-saver, which they never are, so I'm thinking most of the figures I've seen as responses are DOD-minted bullshit. I figure close-combat (the place where fighters are judged) is this aircrafts weak point.
Given that, after 25 years of development, the USAF and their contractors failed to foresee cross the meridian as a problem - yuck, yuck, yuck. The Chinese Airforce must be pissing themselves laughing. This from the only world super-power?
Patriotism is a virtue of the vicious
Nobody says here that there exist anti-stealth technologies on going which could eventually make all the stealthy worth nothing.
Right as of yet no one has figured out a way to beat stealth.
When they do the bar will be raised.
Until then the F22 is invisible to radar.
And as far as a missile failing fine the pilot of the F-22 just slides in below and behind and hits with a simple AIM-9 from it optimum position.
Kill ratio 90% or better when fired like this.
And even if by some chance it misses or fails you have enough time for a follow up shot before retreating out of LOS to take another shot. Not sure on how many 90% or better shots you are trying to say MIGHT miss.
As of right now on one can beat stealth and you can't hit what you can't see.
It really is that simple.