I use Firefox whenever I can.
On first install of the browser I usually end up following a hardening guide which includes stuff like blocking cross site cookies, setting a few things in about:config to disable Pocket/etc, and installing uBlock Origin. I’ve taken what I consider a relatively balanced approach, I don’t use anything like noScript, uMatrix, etc that ultimately just cost a lot of time fiddling to get the 10th website of the week working.
I’ve been more or less fine browsing the web this way for years, but around the start of 2024 I’ve started seeing way more “Access Denied” pages than I used to. I think part of it is Cloudflare or similar, but I don’t know exactly what’s changed or what’s triggering it to occur.
It usually goes away and I can re access the site in 10-30 minutes as usual, but I’ve had it occur in really weird instances, such as trying to change my Minecraft skin and getting blocked by the website. The server block often goes away immediately if I switch my user agent, so I know that it has something to do with how I’ve got everything set up.
Not sure what anyone else’s experience with this has been. I’d like to hear some of your thoughts and tips
I for one want to offer a heartfelt apology. As someone that works in this space, bots are becoming more and more sophisticated. I can’t speak for Cloudflare, but we’re definitely not interested in your personal information. As someone who also prefers their privacy on the web, the fact that bot signatures overlap with privacy-centric signatures sucks. I myself have experienced it on my mobile device with Ghostery. It’s frustrating, I know.
Would you mind sharing the guide you used for hardening your Firefox? I’m curious to see what could potentially be triggering the issue.
Also, I just want to say, I think it’s hilarious that a site blocked you but then allows you to continue browsing after changing your user agent. That right there is bot behavior.
To circle back around to the actual block, I bet changing your skin executes JavaScript which flags something from the anti-bot software.
I had my suspicions that the issues I’ve been running into are mostly because of the worsening botting/scraping situation, and in part due to the general very slight preferential treatment Chromium browsers get on the wider Internet, where anything weird coming from Firefox automatically looks more suspicious because it’s an underrepresented browser already.
I typically just look up “Firefox Hardening Guide” and follow what looks like the best of the first few results every time I do a fresh install. Because of that, I don’t know exactly which guide I followed last, but this one echoes a lot of the steps I remember taking. I’ve since turned webRTC back on because it kind of broke discord(… I know, I know, discord is terrible for privacy but it’s where all my peeps are at!) Didn’t tweak everything outlined in guides such as the one linked, but pretty much whenever there was privacy to be gained seemingly without significant website breakage, I’d toggle it.
The user agent thing was bizarre, especially since it was also on Minecraft.net! I swapped to a generic Chrome on Windows agent and it instantly started working again and let me use the site as normal again. That said the user agent thing doesn’t always work… But the fact that it does sometimes may be a clue to why websites seem to hate my configuration.
Thanks for sharing that!
Truthfully, Firefox is fairly easy to detect. Several facets of the API it uses makes for quick identification. For example, Firefox should be able to report its build ID. Also, it won’t report specifics about the WebGL renderer you’re using like the vendor and architecture.
The link you shared is great and really highlights something I was thinking about today regarding this subject. The more you harden and change things the more you stand out. You’re also more likely to trigger bot detection when you alter specifics about your browser like the major version you’re on. I’ve seen some extensions change the user agent to much older major versions like Firefox 60. That’s a big red flag.
Yes that is bizarre 😂 It’s not clear to me if Microsoft is using their own anti-bot solution or a third party one, but it doesn’t sound really successful with the way it’s reacting.
Overall, I can’t help but thinking the best route is to use the same thing as everyone else but roll your own VPN and change MAC addresses. Ideally, we would have some laws against all of this but I don’t foresee that anytime soon.
I wish I could do more to help. I’m happy to answer questions you might have, though.
Librewolf + AdNauseam on linux
Mull, Librewolf, Mullvad Browser, Arkenfox user.js
Its basically privacy.resistfingerprinting, a generalized useragent, maybe blocked javascript or ads.