Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Cloudflare took down our website

robindev.substack.com

Post reply on HN
Print view View original

411–420 of 483 posts

Re: Cloudflare took down our website

#411

Post by skrebbel »

Aeolun wrote: post #407
skrebbel wrote: post #139

Earlier quoted context omitted.

Yeah so I think it might’ve been a real system limit of sorts. Something timing out somewhere, some pipe getting clogged in a way that their edge nodes couldn’t scale their way out of the way they usually do. Eg because the scaling/monitoring code didn't detect that particular pipe getting clogged etc. We had weird long-running http requests at the time. Note, this is pure conjecture, I’m just well aware from my own…

> I don’t believe they’re that kind of business I didn’t either, but then I read this post :/

Fwiw I still don’t. Large companies mess up too sometimes. This is what it looks like when a sales team messes up.

Re: Cloudflare took down our website

#412

Post by wraptile »

bluelightning2k wrote: post #181

At the moment the account got banned, I would guess that the CloudFlare sales team had this down as a "60% likely to close, estimated close in 6 weeks". There is just no reason they would suspect that they were going to lose the deal to Fastly at this moment. They were very much the default winner. Extortion or not, I just can't fathom that they ragequit the deal at this moment, because they were about to win it. It…

It's quite a common pattern in saas. Someone gets through automated ToS checks with some niche use case and escalates some unrelated issue to support which triggers manual ToS review. That being said, a corporation as big as CF with 120k usd bills should do better and never let this happen. Very amateur.

Re: Cloudflare took down our website

#413

Post by hannasm »

HenryBemis wrote: post #339
luuurker wrote: post #118

The way Cloudflare approaches situations like this is not ideal for anyone. You start using the service and don't pay a lot, so you make plans around a certain level of expenses. Then out of the blue you receive an "urgent" email from a sales representative and suddenly you have to go from $20 or $250 to $thousands right away. Obviously it's not in CF's interest to keep a customer that doesn't pay enough, but droppin…

> but dropping a "bomb" on the customer Why on earth any company would jump from $250 to $10k per month unless they had a gun to their heads? Even if their revenue is to the millions/billions (which most likely is considering the nature of their business). They work for their own profit, not Cloudflare's.

Seems pretty reasonable if the $250/mo plan is costing Cloudflare close to or more than that amount of money due to any loophole or other unforseen expense in the plan.

What seems interesting to me is just what the loophole is and how many other business are also on the radar for this drastic pricing change. Are there other goodwill discounts Cloudflare is ready to start collecting on, or does the gambling site represent a unique situation?

Re: Cloudflare took down our website

#414

Post by sharpshadow »

They figured out that they are losing money with you and offered to negotiate a new plan which was declined so they made and offer which was ignored and you let them know that you maybe want to change the service so they magically found an violation and after still not accepting their offer they took you down.

Pretty standard behavoir from both sides with room for improvement. They should have been more clear about what's going on and you should have been more insightful what they wanted.

In my opinion they acted to fast and not really cooperative, but if you wouldn't have declined the initial offer and started to figure out why they offer it and what options there are, you would have came out with a better deal than 10k/month and likly without 1 year upfront payment which would have given you the time needed to transition to another service.

Re: Cloudflare took down our website

#415

Post by fredthedeadhead »

587846 wrote: post #66

This is my first post on Hacker News as I primarily just browse. This situation kept me intrigued, wanting to know how it would unfold. The Google Cloud situation and all these little happenings, including the proliferation of Gen AI into everything, make me long for the days when companies had their mainframes onsite, in closets or separate rooms, away from CDNs and cloud networks. It seems like a better idea to use…

> The Google Cloud situation

What's the Google Cloud situation?

Re: Cloudflare took down our website

#416

Post by pianom4n »

pclmulqdq wrote: post #200

As far as I can tell, the issue with this is: OP runs a casino/gambling site. Gambling is a regulatory mess (I have spent far too long dealing with this as an RNG supplier), and so it's very hard to comply with every jurisdiction, and each one needs you to prove compliance to operate in that jurisdiction.* Gaming companies spend a lot on compliance and tracking, but since the internet is the internet, it's pretty har…

Maybe the first couple of weeks there was a misunderstanding, but by May 7 it was clear what the situation was. They was told they need BYOIP through and enterprise plan; they just didn't want to pay for it.

Re: Cloudflare took down our website

#417

Post by johnklos »

remram wrote: post #395
johnklos wrote: post #376

Earlier quoted context omitted.

> You can't protect your website from your DNS provider or hosting provider suddenly kicking you off. You are going to be offline for a couple of days. Sure you can. Colocate in two or three places. You're your own DNS provider and your own hosting provider. If one of your colocation companies doesn't like what you're doing for whatever reason, you use the other two until you replace that provider.

"Do it all yourself" is a far cry from "don't put all your eggs in the same basket". The latter principle I agree with, the former not so much. And you are still unprotected from your DNS registrar kicking you out, directing your domain to some "customer terminated" page until you can find another registrar, and have new NS records propagate (days).

You're not really making any points, or at least none that're relevant to this discussion.

Let me summarize: businesses do silly and sometimes stupid things for irrational reasons, or for reasons they never care to divulge. To avoid what happened that led to this discussion, the most suitable solution is to not be at the whims of companies that don't communicate well.

The legal requirements of domain registrars are clearly spelled out, unlike the TOS from, say, Cloudflare which leaves tremendous amounts to the imagination. These are not the same at all.

Re: Cloudflare took down our website

#418

Post by CommanderData »

RomanAlexander wrote: post #358
wiether wrote: post #259

Earlier quoted context omitted.

> By default, every OVHcloud product is supported by the Anti-DDoS infrastructure to defend against malicious activity. https://us.ovhcloud.com/security/anti-ddos/

everyone that has used OVH and received an attack is laughing at that.

In my experience OVH DDoS protection works pretty well. It's not perfect every single time.

I have more good things to say about OVH then bad. I do wish their edge firewall worked against internal traffic too however.

Re: Cloudflare took down our website

#419

Post by eknkc »

weird-eye-issue wrote: post #168
eknkc wrote: post #85

We had a site hosted on CF business plan with fairly large bandwidth usage (completely legal, had a lot of media). They approached us with an enterprise plan but we did not have the budget for it. Asked for a little time, they said fine and we moved much of the bandwidth usage to a couple of dedicated servers on OVH I think. Never heard from them after that.

Can I ask how much bandwidth we are talking? We are doing about 3TB

This was a couple years ago so not sure but likely 50+ TB/month.

Re: Cloudflare took down our website

#420

Post by definitelyauser »

trane_project wrote: post #366
redman25 wrote: post #217

Earlier quoted context omitted.

That's one of the main reasons I'm leary about them. Such a big f-up is difficult to forget. It shows that they have a move fast and break things culture which for a company that is responsible for critical infrastructure feels wrong.

I interviewed there once and they asked me what I would do if a service broke after a deployment. I said the first step was to revert to the last known good version and then investigate. Color me surprised when that was not the answer they expected.

Was the correct answer related to cache invalidation?
Post reply on HN