• Nighed@sffa.community
    link
    fedilink
    English
    arrow-up
    2
    ·
    1 year ago

    I imagine that an individual flight plan can be dropped with no problems (the airline just has to update it and send a new one).

    So there must have been some spectacularly bad coding to cause this.

    • mackwinston
      link
      fedilink
      English
      arrow-up
      4
      ·
      edit-2
      1 year ago

      Wild speculation here, but I suspect the developers were of the opinion that “We only receive data from trusted sources, so we don’t have to bother validating stuff”, which if it is the case would be a tremendously bad take by the developers because your upstream can still have bugs, even if malice is never involved.

      The spirit of Little Bobby Tables is likely alive and well in these kind of systems where messages are “only received from internal sources” and “therefore trusted”.

      • Nighed@sffa.community
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 year ago

        If there is one thing I learnt from the airline industry, it’s that no one follows message specs properly…

        Your probably right, they probably cope with bad messages, but this one was unexpected bad and somehow escaped. They should have had a fall back check for when something escapes though.

        From what they have been saying, it sounds like it was crashing and they have been spinning it as failing safely.

    • mannycalavera
      link
      fedilink
      English
      arrow-up
      2
      ·
      1 year ago

      All or nothing data ingestion. Spectacularly poor. Zero data validation by the sounds of it.