As you are all painfully aware, kbin has been my nemesis pretty much from the start. Unlike Lemmy, Mastodon, Firefish, writefreely, akkoma, synapse, pixelfed, and peertube, I simply cannot competently run kbin. It’s a complete goat rodeo of database errors, kbin and lemmy aren’t getting along, and so on. Though I love the idea and trajectory of Kbin, it simply needs a more time to cook in the oven before being ready.

I will contrast lemmy (infosec.pub) with kbin on fedia.io: fedia.io runs an separate app server and database server. Both servers are larger than the single server that infosec.pub runs on, yet infosec.pub has about 10x the traffic, and kbin is struggling under the load.

If this were all I did, I could likely sort out the various database layout issues and make contributions to fix the code, since I am somewhat familiar with php. Unfortunately, I don’t. And more than that, I have observed a general slowdown in the rate of contributions to the code base of kbin, leaving me to think that it’s not going to get better any time soon.

I don’t take this decision lightly, and I kicked the can down the road for a long time hoping to find a way through so that I didn’t have to do this, but I have to face facts: it’s not getting better and I see nothing that is going to change that.

Most unfortunately, kbin has no options for account migration, which makes this all the more painful. My intention is to shut fedia.io down at the end of November.

I am intending to resurrect it as a lemmy instance, assuming I can sort out how to ensure there are no issues with account keys.

My sincere apologies for this…

Jerry

  • simple@lemm.ee
    link
    fedilink
    arrow-up
    2
    ·
    1 year ago

    Unfortunate. Kbin is a really cool project but it just isn’t as mature or battle ready as Lemmy. The lack of an API especially is a big sore spot for adoption. I almost joined Fedia when Kbin was under load, but Lemmy’s mobile apps were getting really good at the same time.

    I am intending to resurrect it as a lemmy instance, assuming I can sort out how to ensure there are no issues with account keys.

    Sounds like a good option, thanks for keeping it alive.

    • TheVillageGuy@fedia.io
      link
      fedilink
      arrow-up
      4
      ·
      1 year ago

      As said by @teppic, mbin is the answer to the long wait for progress being made on kbin. It’s a piece of cake to change to it and instead of being forced to wait for one person to do something, the contributors can now focus on actually developing and fixing issues. With this new development, I’d say give it some more time.

    • Teppic@kbin.social
      link
      fedilink
      arrow-up
      3
      ·
      1 year ago

      The kbin code does now have an API, kbin.social just isn’t running the latest version of the code.
      On the rate of development and backlog of pull requests - it looks like kbin has just been forked (mbin) in an attempt to address this.

    • jerry@fedia.ioOP
      link
      fedilink
      arrow-up
      2
      ·
      1 year ago

      It’s hard to express how painful it is to do this. Every day I keep hoping there will be some breakthrough, but every day there is just another source of error 500’s

  • e569668@fedia.io
    link
    fedilink
    arrow-up
    2
    ·
    1 year ago

    Both servers are larger than the single server that infosec.pub runs on, yet infosec.pub has about 10x the traffic, and kbin is struggling under the load.

    I wonder what the difference is. One thing I have always wondered is if the double comment / post issue is causing tons of double traffic, not to mention double storage / bandwidth / network costs. I really wish there was an easy solution but I haven’t been able to find any answers. The random thing I thought was like, it looks like everything federates with both fedia.io and www.fedia.io, so I was like maybe everything is double sending on its side? (E.G. if you look at lemmy.world or lemm.ee etc they all list fedia twice.) But people didn’t seem that was likely to be the cause, as hopefully everything is deduped by signing keys rather than hostnames. I don’t know enough to investigate further myself.

    • jerry@fedia.ioOP
      link
      fedilink
      arrow-up
      3
      ·
      1 year ago

      The database server absolutely hammered non-stop. Now that we are using mbin instead of kbin, I will be working with the developers on a fix

    • jerry@fedia.ioOP
      link
      fedilink
      arrow-up
      2
      ·
      1 year ago

      I hadn’t noticed that till now - I knew about the duplicate posts but you are the first to point out www and non-www. I just set www as a redirect

  • troed@fedia.io
    link
    fedilink
    arrow-up
    2
    ·
    1 year ago

    Thanks for all the time and effort you put in to trying to get it to work.

  • chris@fedia.io
    link
    fedilink
    arrow-up
    1
    ·
    1 year ago

    Shame, but understandable. I’d pretty much come to the same conclusion that the interface and concept is really nice, but the backend just isn’t working properly. I’ve migrated most of my communities off already due to issues.

  • Jerry Lerman@hear-me.social
    link
    fedilink
    arrow-up
    0
    ·
    1 year ago

    @jerry@fedia.io
    Interesting because I too announced that my kbin instance, feddit.online, is shutting down on November 4th for a different set of goat rodeo reasons.

    No, it’s just not maintainable at this time.