Anyone know why? Any way we can promote sh.itjust.works to lighten their load possibly?

  • Barbarian@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    21
    ·
    1 year ago

    Only way to do that is have redditors understand that they don’t need to be on the most popular instance in order to have a great experience.

    In other words, nothing we can do, it’s gonna take a while :)

  • PrimordialSoup@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    12
    ·
    1 year ago

    i think in future perhaps instance owners would have a temp place holder page; stating that the instance is down - please select an alternative from the list below. that way at least people get some idea of what’s going on and how to move forward from it.

    • Slashzero@hakbox.social
      link
      fedilink
      English
      arrow-up
      3
      ·
      1 year ago

      Supposedly they are working on major performance improvements for the next release. No idea when it is scheduled to come out though.

      • Barbarian@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        14
        ·
        1 year ago

        The database stuff was solved today. The bigger job, happening as we speak and hopefully finishing in a week or two, is ripping out all the WebSocket stuff and replacing it with simple HTTP. No more crazy updating feeds and weird buggy vote numbers, on top of the performance stuff.

        • Slashzero@hakbox.social
          link
          fedilink
          English
          arrow-up
          8
          ·
          edit-2
          1 year ago

          That’s great news! Is there a new release candidate with the DB changes?

          What was it? Missing indexes? Bad query plans? Bad queries in general?

          Glad to see websockets getting ripped out.

          • Barbarian@sh.itjust.works
            link
            fedilink
            English
            arrow-up
            11
            ·
            edit-2
            1 year ago

            The hot_rank query was basically hitting the server like a truck. You can read more about it in this github issue, but basically calculating the hot rank in the database has been moved to rust.

            EDIT: For clarity, the database issue was solved a bit ago, but today it was released as a Lemmy version which this server (sh.itjust.works) is now on.