It looks like you're using an Ad Blocker.

Please white-list or disable AboveTopSecret.com in your ad-blocking tool.

Thank you.

 

Some features of ATS will be disabled while you continue to use an ad-blocker.

 

Boeing 737 Next Generation software bug blanks out cockpit displays landing on certain runways

page: 1
4

log in

join
share:

posted on Feb, 17 2020 @ 05:32 PM
link   
Well Boeing stepped in it again …….

Their new 737 Next Generation ( Not 737 Max) has a software flaw which blanks out cockpit displays when landing on westward runways at several
airports

The airports. 5 in US and 2 in South America have been found to be affected by the software bug

www.theregister.co.uk...

Hope they get this fixed before any more planes crash and burn ……..



posted on Feb, 17 2020 @ 05:42 PM
link   
a reply to: firerescue

Software developer here. Also I happen to be responsible for integrating with apis for other vendors hardware at that. This has not nor would it ever happen within my development team -- it is what happens when you hire people for other reasons than their expertise. Progress tho, amirite?

ETA: I happen to write PCR software for EMS. You might know what that is gicen your handle.
edit on 17-2-2020 by drewlander because: (no reason given)



posted on Feb, 17 2020 @ 05:48 PM
link   
a reply to: drewlander

Yea happens when hire bunch of 7-11 rejects for $9 hour



posted on Feb, 17 2020 @ 05:53 PM
link   
with profit the primary focus, and a close second being diversity more than skill or knowledge base you end up with sub par products.



posted on Feb, 17 2020 @ 05:56 PM
link   

originally posted by: firerescue
Hope they get this fixed before any more planes crash and burn ……..


From the linked article:


In the airworthiness directive the FAA said it had "confirmed that the faulty version of DEU software has already been removed from all airplanes conducting scheduled airline service into the affected airports" in the US.


There are over 7,000 NGs that have been built. They've found seven airports around the world that cause this, and it appears that it's only when using autopilot on approach. The software has been removed on all US aircraft, and the AD has been spread to all oversight agencies to be aware, and most likely remove the updated software until it's resolved.



posted on Feb, 17 2020 @ 05:58 PM
link   
a reply to: Irishhaf

These examples are exactly what im saying. If you are Blizzard this is okay. I dont care if someones gaming experience is ruined by bad developers. When lives are at stake it is irresponsible to not hire the best because they are the best.



posted on Feb, 17 2020 @ 06:01 PM
link   

originally posted by: drewlander
a reply to: firerescue

Software developer here. Also I happen to be responsible for integrating with apis for other vendors hardware at that. This has not nor would it ever happen within my development team -- it is what happens when you hire people for other reasons than their expertise. Progress tho, amirite?

ETA: I happen to write PCR software for EMS. You might know what that is gicen your handle.


I have always thought that most QA processes are more focussed on documentation than functional testing.

Although, even crap code shops can go through the motions and get apparent QA approval.

Also, the business signs-off the acceptance of third-party code, which makes the business (who may not be properly qualified to do so) partly responsible when bugs are identified. And by the time the bugs are identified, the individual coders have usually moved on, as the cheap shops don't pay enough to retain staff.

So, where do you apportion blame, and more importantly, how do you properly fully remediate/prevent the problems?

edit on 17/2/2020 by chr0naut because: (no reason given)



posted on Feb, 17 2020 @ 06:03 PM
link   
a reply to: chr0naut

as long as your hiring practices are solid... I seem to remember one of the major hacks it came out that the head of IT security was a music major.

*shrugs* read into that what you will.



posted on Feb, 17 2020 @ 06:04 PM
link   
a reply to: drewlander

Software issues happen, even with the best developers involved. This is a minor issue, that has caused zero significant issues. It's affected a small number of aircraft, that only fly into those few airports. It will be resolved fairly quickly, and if it wasn't for the other issues it wouldn't be a blip on the radar.



posted on Feb, 17 2020 @ 06:07 PM
link   

originally posted by: chr0naut
So, where do you apportion blame, and more importantly, how do you properly fully remediate/prevent the problems?


You can't prevent the problems. As long as people write the software mistakes will be made and things like this will happen. The best you can do is to improve your testing and oversight and try to test for every possibility you can think of.



posted on Feb, 17 2020 @ 06:25 PM
link   

originally posted by: Zaphod58

originally posted by: chr0naut
So, where do you apportion blame, and more importantly, how do you properly fully remediate/prevent the problems?


You can't prevent the problems. As long as people write the software mistakes will be made and things like this will happen. The best you can do is to improve your testing and oversight and try to test for every possibility you can think of.


I both agree and disagree. Back in the bad old days of the '90s software dev was hit and miss, but there are increasing numbers of AI tools and mandatory security constraints that mean software dev is a lot better, even for the semi-amateurs.

The thing is that no human or machine can cover every possible outcome, especially with process control of the physical.

So, really, the issue actually needs resolution but we don't have any tools that work, even in relatively limited scenarios. It may be true that we'll always have human stuff-ups (or even computer ones) but it is unhelpful in protecting ourselves against it.

Boeing uses the DOD's/NASA's CMMi certification for software, and they are at the highest level (CMMI level 5). There's still room for improvement.

Capability Maturity Model
From Wikipedia, the free encyclopedia


edit on 17/2/2020 by chr0naut because: (no reason given)



posted on Feb, 17 2020 @ 06:29 PM
link   
a reply to: chr0naut

There's always room for improvement. We've gotten years of 0 commercial accidents, but they don't last. The goal is always 0 accidents or fatalities, and the work is always ongoing, but there's always room for improvement.



posted on Feb, 17 2020 @ 09:11 PM
link   

originally posted by: drewlander
a reply to: firerescue

Software developer here. Also I happen to be responsible for integrating with apis for other vendors hardware at that. This has not nor would it ever happen within my development team -- it is what happens when you hire people for other reasons than their expertise. Progress tho, amirite?

ETA: I happen to write PCR software for EMS. You might know what that is gicen your handle.


Tsk, tsk, tsk....could you possibly be referring to diversity not being the single most important criterion to use in hiring, not competence ?
Let me be the first to say "how politically incorrect of you"....."shame"....

/sarc off

Keep up the good work !!
Hire for competence, alone, and damn the torpedoes...



posted on Feb, 18 2020 @ 02:10 AM
link   
There could be many reasons why it possibly affected only a few airports and their runs i'd imagine but perhaps its a lookup table error where they use a hash to find the index position and the function returns an invalid number and thus theres nothing for the screens to display? If so then a change of the function to use a better hash and a test added to the checklist to run through all the possibles and look for collisions and bad values.

Theres probably so many systems with quite low level of parameter passing validation (as it could slow the cpu right down) and timings may be more important than validation as you hope that the calling function has done its checking



posted on Feb, 18 2020 @ 02:18 AM
link   
I picked up a brand new 737 800 for the airline and was flying it back to Houston when all the screens went blank. Both mine and the first officers. It was a night flight and I was just passing over Junction Tx. when everything went FUBAR. Normally you can boot strap from one pilot station to the other but since both CRTs went blank no amount of switch flipping could fix the problem. I could visually see for maybe 150+ miles and as I had flown that route 100s of times I basically knew ever light on the ground. I informed ATC what was going on and that I could not navigate other than visually..

"No I do not want to declare an emergency but neither do I want a violation for getting off an airway."

Landed in Houston and had maintenance and avionics crawling all over the aircraft while every new face told me this can not happen ! I think I made a wise ache remark along the lines of, "Tell that to the next guy this happens to !"

This whole situation could have been different if I had been in weather when everything went blank. I honestly do not remember if the auto pilot was still functioning but I think it was as I do not remember descending out of altitude which would have been around 37,000 ft. (easier to fly the bird when lower).

I flew all the 737s we had which were the 300, 500, 700, 800, and the 900.




top topics



 
4

log in

join