• SuddenDownpour@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    126
    ·
    1 year ago

    Turn the question back, actually. Do the DDOSers even think they’re achieving anything relevant, beyond being a bit annoying? .world is down from time to time, wow, how terrible. It’s not like I can’t log onto any other server and interact with world’s channels and threads from there, only to have my posts and upvotes become globally available as soon as world is back- Oh, wait, that’s exactly how it works.

    • socsa@lemmy.ml
      link
      fedilink
      arrow-up
      13
      ·
      1 year ago

      The ironic part is that unless they have access to a botnet or are operating their own colo facility, they are probably paying more to run the attack than the target is.

      • SuddenDownpour@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        37
        ·
        1 year ago

        Each instance keeps a local copy of the communities they’re federated with. Users of that instance interact with the local copy, and instances communicate with each other to tell what’s going on.

        This means that, if .world crashes at 5PM, and you have an account on lemm.ee, you can go to lemm.ee’s local copy of world->worldnews, find a thread started at 4PM, comment on it, and other lemm.ee users will see and upvote/downvote your comment.

        When .world is back on at 7PM, lemm.ee will tell it: “Hey, Cold_Brew_Enema said [Thanks a lot for the gold kind stranger] and got 6 downvotes”, and .world will update the thread with your comment and downvotes for users of all instances to see.

        This is my understanding of how it works, and I’m not familiar with the actual code, but it seems to be close to reality.

        • z3k3@lemmy.pt
          link
          fedilink
          arrow-up
          1
          arrow-down
          1
          ·
          1 year ago

          I’m not sure it works like that but it makes sense that way.

        • Cosmos@lemmy.today
          link
          fedilink
          arrow-up
          2
          ·
          1 year ago

          There may be some things that don’t get synced if the instance holding the pending federation messages also goes down or restarts because AFAIK they are stored in memory, not in persistent storage.