Will talk about Linux, plants, space, retro games, and anything else I find interesting.

Also mesa@piefed.social over on Piefed.

  • 377 Posts
  • 1.23K Comments
Joined 2 years ago
cake
Cake day: June 11th, 2023

help-circle






  • This has happened to me personally. I have a very small git server sitting on a very small instance. I noticed my traffic was spiking for the past few days and thought nothing of it.

    After I looked at the logs, it was 6 ip addresses. 1 Amazon LLM search and open AI with the other 5. I added them t the fail2ban and thought nothing of it. About 10 min later, openAI had a multitude of new IPs hitting the server again, specifically the git repos. I again looked up to confirm that, yes it totally was OpenAI.

    I had to create an AI blackhole with python and create new rules just to stop the LLM madness. It worked but im tempted to put a very small capta like give me 2+2 on a form in order to see my code.

    Worst part, I had a robot.txt that totally blocked the indexing of my sites. But they just didnt care.












  • I use Mastodon to follow people and Lemmy to follow topics.

    I actually created a Proof of concept where each hashtag would be a subreddit like interface a couple of weeks ago. Its VERY rough but links2 compatible :D. The idea is the first hashtag will be the “lemmy community” like interface. Its just mastodon.py with a very small fast api front end. Popped it up in an hour or so. I need to work on the UI badly but I keep not having the spoons.




  • My wife and I were looking at Croc and just talking about getting it for SD. My biggest issue is controls. I would like to know if it is easy/hard to get the controls working. Im tired of spending weekends getting a couple games working not off of steam.

    Ill be looking into heroic! Thanks for the tip.







  • I had the same issue. OpenAI was just slamming my tiny little server, ignoring the robots.txt. I had to install a LLM black hole and put a very basic password protection around my git server frontend, since it kept getting slammed by the crawler.

    As much as I dont like google, I did see them come in, look at the robot.txt and no other calls for a week. Thats how it should work.