Skip Navigation

InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)B

BitOneZero @ .world

@ BitOneZero @lemmy.world

Posts
24
Comments
88
Joined
3 yr. ago

Supporter of the art work of James Joyce, "World Wide Wake" as in year 1924 Irish Dublin story "Finnegans Wake"!!! Year 1923 "Finn's Hotel"

Joy to you!

  • Good News

    Most of these ‘attacks’ are targeted at the database

    A major PostgreSQL performance issue, logic mistake, was discovered today in lemmy_server and is an easy fix. Details: https://lemmy.world/post/2008987

  • Lemmy.ml is down right now (back online, was unreachable for about 6 hours)

    Jump
  • You can use markdown codes in titles. Bold, italic, and in my case strikethrough. Two tildes before and after the text.

  • It is back up

  • Down, "Server error" message for past 20 minutes at least

  • Lemmy.ml is down right now (back online, was unreachable for about 6 hours)

    Jump
  • Lemmy.ml is down right now (back online, was unreachable for about 6 hours)

    Jump
  • Beehaw.org was down for over 30 minutes, came back for 15 minutes, now down again another 30 minutes.

  • Lemmy.ml is down right now (back online, was unreachable for about 6 hours)

    Jump
  • Lemmy.ml is down right now (back online, was unreachable for about 6 hours)

    Jump
  • 6 hours down for lemmy.ml is really going to have a lot of items queued on lemmy.world, beehaw, and other popular servers to delivery to lemmy.ml - causing resources to climb on all those servers.

  • Lemmy.ml is down right now (back online, was unreachable for about 6 hours)

    Jump
  • Lemmy.world is now responding some at least, and some of the comments from remote servers came into this posting just now.

  • Removed

    TIL lemmy.ml is a pro-authoritarian CCP shill instance

    Jump
  • Personally, I don't think the Lemmy numbers are that big. There are some instances running bots of feeds, a lot of memes. A lot of discussion about Lemmy and other federated services. I do not really consider the volume of posts/comments about everyday topics to be that high.

  • Good to see a heavy production server taking on the scaling issues. Thank you! To discuss Lemmy performance issues, there is a dedicated community: !lemmyperformance@lemmy.ml

  • Yes, I installed a Lemmy server my own self, there is no screening, approval, or even a "terms of use" on the signup page. This is the "wild west" of social media. And some of the claims on the GitHub project page such as "full delete" are an overreach, as it has no footnote that federated servers do not have to comply with the delete of your replicated votes/comments/posts/profile

  • the comment_like database table in Lemmy also has a timestamp on it, "published" field, that discloses what time you voted. This reveals patterns of your Lemmy usage to other federated servers.

  • When the browser loads that URL, hotlinked image, that server has to have your IP address to return the results. Just browsing posts those images are being loaded.

  • Your home instance will act as a proxy and only they have access to your email and IP address.

    Your home image typically doesn't proxy image loading, those are hotlinked to the Lemmy server that the image was uploaded to. So your IP address and browser string are going to other Lemmy servers.

  • Federation protocol implementation is poorly optimized, it does SQL lookup for the person every incoming post/comment/like, checks the community if they are banned, all of this with zero caching. HTTP outbound is also very simple design that only now some logic to detect dead peer servers is being added, etc.

  • For micro/personal server runners

    Having to send out all those posts, comments, likes is killing the servers holding all the content. The project leaders really need to make an announcement about the performance problems, they are systemic.