Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Some models of Airbus A350 airliners need to be hard-rebooted after 149 hours

theregister.co.uk

Post reply on HN
Print view View original

21–30 of 77 posts

Re: Some models of Airbus A350 airliners need to be hard-rebooted after 149 hours

#21

Post by raverbashing »

raxxorrax wrote: post #17
dahartigan wrote: post #2

It's a scary thought that such issues exist, regardless of how common or not.

I think planes already scrutinized to a very high degree. What I am more concerned about is airlines doing the reboot in flight to save time. Planes are often on a very tight schedule (maybe not cargo planes).

There's plenty of time to do the reboot between turning the planes (disembarking and boarding) - a good amount of A350 planes spend the day in the sun at their destinations since the operator can't fly anywhere else with them between flights.

Re: Some models of Airbus A350 airliners need to be hard-rebooted after 149 hours

#23

Post by mangatmodi »

viraptor wrote: post #8
mangatmodi wrote: post #7

It is a very well know issue with every plane. Sometimes there are no solutions to a problem. You need these hacky solutions. The title is clearly catchy with everything that is going on with Boeing. But the point is, this reboot process is very well managed and known. So I won't call it scary.

I agree it's not scary and it's a good, known workaround. But it's software - we shouldn't say "there are no solutions". The solution is: fix this problem and add long runtime testing to the qa process. Especially if this is a known issue in other planes.

When dealing with physical systems it is impossible to have no bugs. To give an extreme example, there is a not a small chance that the cosmic rays can change a bit in a system's memory - https://stackoverflow.com/questions/2580933/cosmic-rays-what...

The issue here is overflow due to time. The time is saved in a variable (don't know how much bits), which overflows after the gives period. Now there are two options 1. Upgrade circuits of every plane. These planes were designed/built a long time back. Bigger registers were not practical due to costs. 2. Document it and have a process for it.

Re: Some models of Airbus A350 airliners need to be hard-rebooted after 149 hours

#25

Post by NKosmatos »

JustSomeNobody wrote: post #13

Occasionally rebooting a system is actually a good thing. “Hello, IT, have you tried turning it off and on!?”

The opening line from the IT Crowd [0], one of the cult British sitcoms and a favorite of mine. Highly recommended :-)

[0] https://en.wikipedia.org/wiki/The_IT_Crowd

Re: Some models of Airbus A350 airliners need to be hard-rebooted after 149 hours

#27

Post by JorgeGT »

raxxorrax wrote: post #17
dahartigan wrote: post #2

It's a scary thought that such issues exist, regardless of how common or not.

I think planes already scrutinized to a very high degree. What I am more concerned about is airlines doing the reboot in flight to save time. Planes are often on a very tight schedule (maybe not cargo planes).

Don't worry, the AD (airworthiness directive) specifically calls for complete, ground power cycles. These cannot be done in flight. https://ad.easa.europa.eu/blob/EASA_AD_2017_0129_R1.pdf/AD_2...

Re: Some models of Airbus A350 airliners need to be hard-rebooted after 149 hours

#28

Post by jakobegger »

BrentOzar wrote: post #11

I think the comments here say more about us as an audience than they do about the Airbus. We're totally accustomed to hardware & software problems whose solution is, "just restart it," and we don't even find it all that disturbing - mostly just humorous. Our standards as developers & testers have gotten pretty low. I think the only way you'd see outrage here at HN is if the restart involved a real physical crash (as…

The thing here is that the issue is well understood and the root cause is known. We know that rebooting prevents the error.

The problem is when "have you tried rebooting" becomes the first troubleshooting step and you stop investigating.

But if you realise after the investigation that a reboot reliably prevents the issue, it's a worthwhile approach, especially if fixing the error would require many complicated changes that may lead to further errors.

Re: Some models of Airbus A350 airliners need to be hard-rebooted after 149 hours

#29

Post by pbhjpbhj »

mangatmodi wrote: post #23
viraptor wrote: post #8

Earlier quoted context omitted.

I agree it's not scary and it's a good, known workaround. But it's software - we shouldn't say "there are no solutions". The solution is: fix this problem and add long runtime testing to the qa process. Especially if this is a known issue in other planes.

When dealing with physical systems it is impossible to have no bugs. To give an extreme example, there is a not a small chance that the cosmic rays can change a bit in a system's memory - https://stackoverflow.com/questions/2580933/cosmic-rays-what... The issue here is overflow due to time. The time is saved in a variable (don't know how much bits), which overflows after the gives period. Now there are two options 1.…

Environmentally induced errors aren't software bugs, just because there's a problem elsewhere doesn't mean we shouldn't seek to mitigate other problems.

In plane investigations I've looked at (not many) the issue has always been a compounding of several errors or shortcomings .. that strongly suggests you shouldn't let small errors build up in different systems, to me. [1]

If it's a register which takes down the whole system then surely they'd know that (and could fix it with a watchdog that returned the effected systems to the boot state without reboot) -- other comments seem to be saying "meh, it's complex, doesn't matter what the error is as long as reboot fixes it"; that seems really dangerous in safety critical systems.

[1] but I acknowledge the "better the devil you know" issue and that pragmatism and cost take over at some point.

Re: Some models of Airbus A350 airliners need to be hard-rebooted after 149 hours

#30

Post by pcurve »

AuthorizedCust wrote: post #20
BrentOzar wrote: post #11

I think the comments here say more about us as an audience than they do about the Airbus. We're totally accustomed to hardware & software problems whose solution is, "just restart it," and we don't even find it all that disturbing - mostly just humorous. Our standards as developers & testers have gotten pretty low. I think the only way you'd see outrage here at HN is if the restart involved a real physical crash (as…

That's not necessarily a "low" standard. It's that the value of preventing a reboot every few days is much lower than other things we could spend our time on. Mandatory oftware perfection would be a huge drag on valuable new features.

I agree with you to a certain degree but appropriateness of the trade off should really depend on where the software is used. Airplane, powerplants, and train signaling systems? I would want longer uptime.
Post reply on HN