Cloudflare is turning off the internet for me

Post by laughinghan » Wed, Jan 22, 2020, 1:34 PM UTC

yc12340 wrote: post #276

Analemma_ wrote: post #127

Earlier quoted context omitted.

I think you're confusing what you wish was true with what is actually true. For instance, here was a post from a few weeks ago about how one annoyed user was able to take down a Mastodon instance until the admin gave up and put it behind CF: https://news.ycombinator.com/item?id=21719793 . Bear in mind, if you're running a Mastodon instance, you're probably well-aware of the downsides of centralization and would only…

> I think you're confusing what you wish was true with what is actually true. And you are cherry-picking poorly sourced anecdotes to better suite your position. A VPS with 100Mbps virtual adapter physically can't withstand DoS from single attacker with fiber connection (or equivalent of it). This does not have much to do with anatomy of DoS attacks, just simple math. Cloudflare subsidizes their free users by giving a…

Post by htfu » Wed, Jan 22, 2020, 1:39 PM UTC

why-oh-why wrote: post #267

htfu wrote: post #220

Earlier quoted context omitted.

At least on iOS Safari there's this bug (feels like it's always been there, maybe content blocker related?) where sometimes a search simply gets eaten. The browser somehow thinks you went from nowhere straight to the page you're on, even though you went past a Google results page. I assume it's somehow redirect-related and that's why these sites tend to trigger it.

I’m thinking of two situations: - you tapped the “Siri suggestion” result which completely skips the SERP. I hate that “back” doesn’t bring you back to what I typed in the search/URL bar - I regularly visit The Verge, open a story and then the back button doesn’t take me to the homepage but to the page before it. I blame their crappy JavaScript but maybe we’re experiencing the same thing.

Post by deogeo » Wed, Jan 22, 2020, 2:02 PM UTC

tomc1985 wrote: post #231

deogeo wrote: post #212

Earlier quoted context omitted.

> malicious actors and abuse It's hard to consider simply viewing content to be malicious or abusive, no matter how automated.

I used to work for a data-scraping firm and very often we would accidentally knock many web sites offline when we pointed our crawlers at them. I'd love to agree with you, but the crawler problem is 100x worse today than it was a decade ago

Post by 5Qn8mNbc2FNCiVV » Wed, Jan 22, 2020, 3:11 PM UTC

missblit wrote: post #250

filoleg wrote: post #95

Earlier quoted context omitted.

Yes, and Cloudflare provides a fairly reasonable solution to this problem. Or, at least, it seems reasonable in the eyes of someone like me who never had to work on something that attempts to solve this problem, so there might be some serious caveats, but I am not aware of those. If someone with more domain knowledge can chime in on this, that would be appreciated as well.

Caveats: Won't protect against scrapers that execute JS (like ones based on headless browsers -- This includes some modern search engines!) Won't protect against anyone who takes some time to read their 80-some lines of minified "obfuscated"* JavaScript and hook up a simple text transform to their crawler of choice. So basically it'll only protect against truly trivial scrapers, but not against anyone who wants to ge…

Post by shadowgovt » Wed, Jan 22, 2020, 4:05 PM UTC

incompatible wrote: post #244

shadowgovt wrote: post #98

Earlier quoted context omitted.

Bad actors who are bad at being bad actors, which is actually the bulk of bad actors. It's maddening, but it's true. I've seen tale of people having to modify resource auto-generators that created URLs with hexadecimal identifiers in them because the sequence "ad" in a URL would trip ad-blocking browser plugins. You might ask yourself "how many ad companies worth their salt have 'ad' in the URL path?" and the answer…

There's somebody who can build a custom browser but can't figure out how to change the user agent string?