HN user

Wallacy

42 karma

[ my public key: https://keybase.io/wallacy; my proof: https://keybase.io/wallacy/sigs/wYkDUxWG7-f7y7b4vspA17gNW_oQegsP20lCuMfu-wI ]

Posts0
Comments24
View on HN
No posts found.

Discord is the place to go. All CF Devs are there.

I got on particular fix one time in just a few minutes, and scalated pretty quickly. I got a recomendation fix on my side (a code change on my deployment that cover the problem) and a permanent fix on CF side 4h latter (time that took to changes propagate to all colos)

" Apps that want even more control can move to using full-on Durable Objects (which will soon support a SQLite database per-object)."

Can you elaborate this little bit more? Im using DO today and i have a bad time sharding my data (works, but i hate it);

So i will have the option to use the standard store or/and SQLite?

If so, i dont can keep with my DO (because i have control of everything) and use SQLite for things that is bigger than what the value store supports.

Was computed related, but he find a loop role on his business contract. He was right by the legal team os the service, but they used a another term that make they right to not work with us anyway. His document ID was banned (he cant create accounts with his name/id anymore).

I dont remember the details, but was related to the free limits per account. And if my memory serves, he was in the plan that he can create how many parent accounts he wants for free. That plan does not exists anymore.

My POV as customer for 5 years, doing mostly Video and Binary transfers and asked to move to enterprise few times (and still on the Pro plan, never had any interruption):

They will NEVER says specifics bytes or anything like that (any cloud provider that has any abuse prevention clausule will not too)!

If you says that a "large" file is 100mb people will just gamble and split the files in 100mb chunks. Same for any value here. The idea is mostly about pattern recognition (any abuse detection algorithm) and how much do you pay and how much you are costing.

Bandwidth is cheap for CF but they dont want anyone gaming the system, i will upgrade for enterprise at some point mostly because the SLA, but at the time im writing here, i pay $20 month for a service traffic that cost (to me) hundreds `per day` on AWS or GCP...

People here talk like they multi million dollar project will be blocked because some user downloads a 10mb file on a free plan. But if you really have a critical product you can get enterprise plans starting at $1500 dollars, hardly more than 3K. The enterprise plan is mostly about the SLA than actually services; Thats not the reality of mostly of the people. The basics plans are very good for almost everyone IMHO.

My rule of thumb of any cloud service / provider: You will for sure get deals that will cost very little to you, that is 10x/20x cheaper, but once you "savings" are on the 100x range you may find someone on the other side trying to make you to pay more. I have a friend that was banned from GCP for that reason (long story).

Another example: Hetzner - You can do a lot with the "free" 20TB bandwidth... But do not think that just because theres a number here theres no "catch" for some "creative flow", we got blocked once, was a mistake and is resolved but anything free has abuse protection rules (sometimes not writen) and will find you.

Sir, you are probable the best CEO of this universe.... On alternative reality you company has a Search Engine (count with me to help with that) and doesn't need any advertise to keep that running, also has "don't be evil" as code of conduct and actually apply that....

But yes, no one is perfect, but at end of the day i really prefer your business model that does not need break users privacy.

Enterprise plan can carry, but as far I know, will not make be much more than 2500-3500;

Maybe they should create a Enterprise that has a flat price and be self subscribed. And a “Corporate” for people that needs several limits more than a “basic” enterprise that they offer. Or just stop that and put a price on things.

The basic enterprise is fair just for the SLA. (Depending the size of your business)

Thats problem is for CDN product. On R2 like other Developer products (KV, Worker, DO) you are already paying the bandwidth over the storage price. All those products can be used to HTML and non-HTML product.

https://www.cloudflare.com/supplemental-terms/#cloudflare-de...

Thats the major problem of the CDN product, they dont say that are no traffic fee, they just say they will not charge you "except" on some cases (that you never know). But other products they are very specific of the egress fee, like Stream, Magic Transit, Spectrum, etc.... They should fix that matter on CND, no one like that.

Sure, i dont know what will happen if someone just try to make a youtube like over R2... They probable "want" that....

FWIW: https://www.cloudflare.com/supplemental-terms/#cloudflare-de...

"The Cloudflare Developer Platform consists of the following Services: (i) Cloudflare Workers, a Service that permits developers to deploy and run encapsulated versions of their proprietary software source code (each a “Workers Script”) on Cloudflare’s edge servers; (ii) Cloudflare Pages, a JAMstack platform for frontend developers to collaborate and deploy websites; and (iii) Workers KV, Durable Objects, and R2, storage offerings used to serve HTML and non-HTML content."

You can use non-HTML content on R2, DO and KV.Thats was updated, if i'm not mistaken few months ago.

If the update is:

set -e -u -o pipefail

(dos-make-addr-conf | tee config.toml) && dosctr set template_vars config.toml

well....

`tee config.toml` will still produce a empty config.toml

(set -e - u -o pipefail;false | tee config.toml) && cat config.toml

But you can use the '>' operator, that will create the file only if the command runs successful:

dos-make-addr-conf > config.toml && dosctr set template_vars config.toml

The terms os service is still in place:

2.8 Limitation on Serving Non-HTML Content The Services are offered primarily as a platform to cache and serve web pages and websites. Unless explicitly included as part of a Paid Service purchased by you, you agree to use the Services solely for the purpose of (i) serving web pages as viewed through a web browser or other functionally equivalent applications, including rendering Hypertext Markup Language (HTML) or other functional equivalents, and (ii) serving web APIs subject to the restrictions set forth in this Section 2.8. Use of the Services for serving video or a disproportionate percentage of pictures, audio files, or other non-HTML content is prohibited, unless purchased separately as part of a Paid Service or expressly allowed under our Supplemental Terms for a specific Service. If we determine you have breached this Section 2.8, we may immediately suspend or restrict your use of the Services, or limit End User access to certain of your resources through the Services.

But realistic we can serve a lot of content (any content) before trigger any problem, that includes video. I have a one App that does several GB per month (including images/video) on Free plan, and other one using few TB/month on Pro Plan... I will probably upgrade to the business plan after when i get close the 1PB, but no one ask me too.

My experience says that no one will contact you about any egress if you are not trying to implement any file host site, youtube-like VOD service, etc.

Its not like that.... If you send you code to github, then deleted that code in our computer, and github goes down before any other person cloned your repo and decided not delete, then you code is lost.

You same question can be applied here: Why someone will download a random repo on the github and store just for fun? No one will do that without any reason. So, on git the only people that keep any file/repo offline, has any interest on that content to be hosted with then.

The same here, if I send any file using IPFS, the same file will be accessible using any gateway, my own gateway, my neighbor, or some other that i dont know who owns in other part of the globe.

The gateway is just a gateway. Its one way to find some asset on the network, and has a optional cache. Months ago I send one file using IPFS, then accessed via cloudflare (and others) gateway, then I deleted my file, and in few hours the file was inaccessible in every gateway that i tested. No gateway keep you file forever....

But then can if they want, anyone can keep any block if they want. So, the same way as my previous example using git! The asset will exists util everyone that hold that asset want to keep, but because you cant control the others nodes, once some asset is public over the network, thats can be "forever" or just few hours.

So... IPFS is more about in how you can share assets privately than the lifetime of the object. And with or without cloudflare IPFS is decentralized because at any point i can setup any gateway that can listen my assets in the same way (same hash) than any other gateway.

IPFS gatway is much more about node discovery + cache.... No centralization happens here. You can store third party blocks in you node if you want, or rely on other nodes if want, but thats is up to you. Its the same as upload anything to any cloud then delete the local copy.

"At launch, R2 will cost $0.015 per gigabyte of data stored on the service, about half as much as AWS charges for customers on its frequent access tier. Infrequent access to R2 will be free, the company said. "

https://www.protocol.com/enterprise/cloudflare-r2-storage-aw...

So.... I still need to test, but if that is true can be a game changer. Of course... $0.00099 is almost free, but if everything is done automagically will be awesome.

If tou put the CF in between will not pay the egress. But will be better to Hetzer be part of the alliance.

In this case is because they actually does not charge bandwidth. I have few services on CF and never paid a dime for bandwidth. Even on free plan I push few terabytes.