is this Little Bobby Tables?

  • jonathan@piefed.social
    link
    fedilink
    English
    arrow-up
    19
    ·
    1 day ago

    My latest example of this was a memory leak in some old software I wrote. It grew linearly with traffic, so it wasn’t until traffic randomly spiked that I started seeing problems.

    • vrek@programming.dev
      link
      fedilink
      English
      arrow-up
      7
      ·
      1 day ago

      Dylan bettle talks about this in one of his presentation. He made a website that did actor and actress recommendations, I believe for like extras to be hired. They kept getting alerts around 7-715am and then around 12.30pm to 1245pm.

      Eventually they figured out that casting agents typically woke up, checked email, and submitted the requests on the website. The actors and actresses typically had other jobs and had to wait till lunch to check and apply.

      They had to redesign the system and buy more bandwidth to handle these spikes.

      He’s a great presenter and I highly recommend watching all of them, even if you’re not a programmer. And I probably got most of the details wrong… But it was a talk at one of NDC conventions.

      • jonathan@piefed.social
        link
        fedilink
        English
        arrow-up
        1
        ·
        3 hours ago

        My actual “solution” was to do nothing and just let Kubernetes restart it when it OOMs 😅

      • zorro@lemmy.world
        link
        fedilink
        English
        arrow-up
        2
        ·
        21 hours ago

        Cron job too complicated, just buy one of those timed light controllers to power off the server every night for an hour.