November 16, 2025
Bots vs. Blogs: Round 999
The Internet Is No Longer a Safe Haven
Small site swarmed by bots; commenters demand gates, country bans, or a brand-new web
TLDR: A small indie site was knocked offline by bots pretending to be regular visitors, sparking a brawl over how to save the open web. Commenters are split between proof-of-work gates, hiding behind big providers like Cloudflare, or even blocking countries — with some saying the old internet dream is dead.
A cozy personal site just got steamrolled by wave after wave of “AI” scrapers pretending to be normal Chrome browsers, and the logs point to big cloud IP ranges. The owner considered dropping a trap called Anubis, but sighed that every bot fight kills a little of the hobbyist joy. Cue the comments section turning into a town hall with pitchforks and memes.
The hottest take? Gate the open web. One camp, led by voices like BinaryIgor, wants to hide content behind proof-of-work puzzles — little speed bumps for computers — so bots pay in CPU before they can binge. Another camp says the uncomfortable quiet part out loud: centralize behind giants like Cloudflare, because they see more traffic and can spot bad behavior better; small sites can’t keep up. That sparked a freedom-vs-safety brawl, with one commenter dropping the nihilist mic: “Entropy always wins.”
Meanwhile, others are going full scorched earth. Time4tea admits to blocking whole countries just to breathe, lamenting “enshittification” by Big Tech and AI. Another user simply declared, “the internet is over,” and folks half-joked we’ll need a “Web 3.0 but not that kind” to start fresh. Between grim humor about fake user agents and links to ipinfo.io, the mood is clear: people love the open web — and they’re terrified it can’t stay open without a price.
Key Points
- •A small server hosting the Brain Baking website was temporarily knocked out by scraping bots.
- •System monitoring showed Gitea and Fail2ban consuming most CPU while Nginx access logs were flooded with repeated GET requests.
- •Many requests returned HTTP 502 responses, and the presented user-agent string mimicked a mainstream browser.
- •User-agent checks failed to identify the traffic as bot activity, highlighting that headers are easily spoofed and unreliable.
- •Attacking IP ranges were primarily 47.79.*, which ipinfo.io linked to AS45102 owned by Alibaba (US) Technology Co., Ltd.; stronger measures like Anubis are being considered.