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.
The problem probably isn't with the time change. Airplanes use GMT so the local time doesn't matter. The problem is probably related to the longitude going from W179.99 degrees to E180 degrees.
Assuming it WAS a time issue upon crossing the International Dateline...
Design problem? Why should navigation software require "local time"? They knew they were crossing the international dateline, so they must be linked to GPS timing systems... why not just use GPS' universal time? (Sure, you want local time eventually for your displays but that's a "view" calculation, not one intrinsic to the navigation software)
Bug tracking problem? Did the testers not think of testing about a time zone change? Did they assume the above that everything would be on a universal time and therefore didn't see the need for crossing time zones?
Why wasn't this a stock reusable code module in Lockheed Martin's labs?!?
(And for a media look at this issue, check out the anime Geneshaft or the movie The Pentagon Wars)
Why do you guys give +5 to someone who doesn't know for sure how the date line works, and who merely looked up which SI prefix was small enough to cause a 64-bit overflow? Most likely the bug has to do with overflow in position, not time. Even assuming this has to do with time overflow, modern GPS electronics can only measure signals to within 10 nanosecond. Using femtoseconds (10,000,000x smaller) is complete BS to make his argument work.
The F-22 is a full stealth fighter, the EuroFighter is not in any way. If you cannot understand why you cannot base a stealth plane design in any major way on something that is not a stealth plane (hell, no stealth fighter has existed before the F-22) then why are you eve talking about this as you clearly have no idea about anything involved.
And yes the F-22 is likely worth 84% more than the Eurofighter in terms of performance due to stealth alone.
Incidentally since the F-22 is what the F-35 is based on that $70billion has technically led to the creation of two planes, the later of which is being sold quite widely.
I have worked on Commercial and DoD avionics, and this type of thing is inexcusable.
Commercial avionics software of the sort described is governed by a standard called DO-178B level A or level B. The process is so rigorous that the slogan is "no-one has ever died from software failure in a commercial airliner, yet." DO-178B level A is expensive. It is virtually impossible that a software error of the nature described could get into a certified aircraft.
Having said that, the military is not obliged to follow commercial standards, but there is a trend toward using DO 178-B in military systems in part because the Europeans are starting to require commercial JAA/FAA certification for all aircraft that enter their air space. But even in the more lax military world, every line of code is typically formally reviewed and there are independent testers. The type of error described should have shown up in simulators before the first flight of the aircraft. Test flights should have stimulated the error long before a squadron ever attempted a transpacific flight.
Even worse still, avionics systems are supposed to be isolated from each other. Navigation radios typically share nothing but power with GPS or with engine instruments etc. Great effort prevents one system from disturbing the power of another too. Aircraft typically have two or more separate primary navigation systems plus inertial guidance and old fashion compass + baring/vector navigation. Military aircraft need to survive both normal equipment failures and battle damage. Military radios (including navigation) need to be isolated from other systems for security reasons too. Those NSA guarded encryption systems can not be contaminated by software that has lower security classification (like navigation)without somebody going to federal prison for a long time.
The bottom line is that something very very wrong, negligent, and illegal needed to happen for the described error mode to manifest. That makes me doubt the story.
The return on investment is HEAVILY in favor of the F-22. There is no aircraft anywhere even close. The Eurofighter is the second best fighter aircraft ever built, but it is miles from being in the same class as the F-22 Raptor.
It sure doesn't need to be like that.
Our desktop computers crash because we can tolerate crashes. There is some redundancy - if my notebook crashes, I reboot it and, in a couple minutes, I am back to work. If it breaks, I grab another computer and continue.
A plane, on the other hand, should work at all times. When lives depend on some equipment, one should enforce much higher standards than we do on desktop or even mission-critical busines software. Nobody dies if your sales people have a 5 minute outage. Nobody dies if you can't create a patient record. People die when the computers a plane relies upon fail.
It's completely unacceptable - and quite alarming - to see a plane malfunction like that on its first deployment.
Things like that should have been exercised years ago. By now, the code should be rock-stable. Whant kind of quality assurance they did?
http://www.dieblinkenlights.com
Only if stealth is a requirement. In a real dogfight, the Eurofighter likely wins because maneuverability was foremost in its design, whereas the F-22 has stealth as the foremost design priority. The thought is that engagements are likely to be fought a distance with missles, and the low observability tech will allow the American aircraft to engage long before the enemy can return fire. This does not jive entirely with engagements of the past, which often involve close range encounters to verify enemy, or orders to wait until fired upon to return fire.
Compare this to the ability to put twice as many aircraft in the sky, carrying more munitions (while the F-22 has some stealty weapons bays, maxed out with a full bomb load involves external mounts with has a huge impact on radar visibility). Point is, whether stealth is worth 84% more has more to do with your mission profile and expected enemy/target,
And even more people die when all systems work perfectly.