Conversation

Notices

1. Terence Eden (edent@mastodon.social)'s status on Wednesday, 06-Sep-2023 12:09:00 UTC Terence Eden

   in reply to

   So the system got confused because it looked like the flightplan was impossible.

   > the software within the FPRSA-R subsystem was unable to establish a reasonable course of action that would preserve safety and so raised a critical exception.

   Good! That's what computers are supposed to do. So it placed itself into maintenance mode.

   The backup system noticed that the primary was down. It sprang to life instantly. It read the flightplan. And then also crashed.

   The whole thing took 20 *seconds*!

   • Santa Claes 🇸🇪🇭🇰🎅 likes this.
   • Terence Eden (edent@mastodon.social)'s status on Wednesday, 06-Sep-2023 12:09:00 UTC Terence Eden

     in reply to

     This wasn't even a one-in-a-million error.

     The software has successfully processed 15 million flightplans since 2018. And this is the first one it choked on.

     Incidentally, for all the tabloid blather about remote working - that's what enabled it to be fixed so quickly:

     > The process deliberately utilises remote support technology to ensure that issues can be investigated immediately without time lost to travel by support engineers.

     Santa Claes 🇸🇪🇭🇰🎅 likes this.
   • Terence Eden (edent@mastodon.social)'s status on Wednesday, 06-Sep-2023 12:09:01 UTC Terence Eden

     The UK's Civil Aviation Authority have a brief report about what caused all the flight cancellations the other day:

     https://publicapps.caa.co.uk/docs/33/NERL%20Major%20Incident%20Investigation%20Preliminary%20Report.pdf (440KB PDF)

     A few interesting thoughts about their findings…

     The major cause was:

     > The plan included two waypoints along its route that were geographically distinct but which have the same designator.

     I don't know enough about ICAO4444 / ADEXP to be sure, but it sounds like an understandable, albeit avoidable, error.

     (1/n)