R2 runs on Cloudflare's own infrastructure, not on AWS

Cache Reserve is on R2 itself, perhaps you mean tiered caching? Afaik neither of them work on r2 custom domains, and tiered caching doesn't work with the Worker cache api either

if r2 used aws it would be unusable

oh hey there's a tag for it

?r2-tiered-cache

> R2 Public Buckets do not currently work with Tiered Caching because of the way Public Buckets work internally. You can still apply regular cache, but Tiered Caching will not be applied. > > The team has mentioned there might be a fix eventually: however there are higher priorities at this time.

any way to change the R2 404 page?

This isn't a very good 404 page

That is not the 404 page for R2. That's your browser saying the site seems deceptive from user reports

This is the 404 page

I know, but why is it triggering

nvm it got fixed

how do you setup edge cache for R2 custom domains?

these settings don't seem to work

Plus these cache rules

And yet cache still returns

EXPIRED

after merely a minute

Setting those rules means it'll be cached for up to 1 year - not always 1 year. If your assets are getting requested infrequently, they're not going to remain in cache for long, because the cache can only fit so much in it

hi, If I delete a file from the R2, does Cloudflare also clear the cache?

I don't think they do, but you can purge the cache manually yourself

ok, thanks

how to make sure it gets cached as long as possible

Have a busier site

Hi, I am a bit new to R2 and I am facing an issue in the workflow I am trying to implement : I am building debian packages and store them to R2 through a github action. My issue : I want to make my packages available for users to make "apt install" I connected a domain in the settings tab of my R2 bucket I see everything is ok (status active, access to bucket allowed) But when I try to access the URL that should serve the R2 I have a 404 error. Any thoughts ?

The doc I am following is here : https://developers.cloudflare.com/r2/data-access/public-buckets/#connect-your-bucket-to-a-custom-domain

FYI : I am able to install using "apt install", but when I try to reach the URL bucket it is failing ...

Is the path to your object correct? Do you have Page/Transform rules on your zone? Also note that hitting the root path (/) will not show you a listing of all files, you must hit the pathname of a single object (so like /my/key/here)

hey, there's no support for migrating things from s3 compatible stores like https://wasabi.com/ yet right?

Is there any way to be picky/set rules about what type of assets CF will put into R2 cache reserve? I would love to put all of our images onto it, but it seems to be all-or-nothing, which ended up increasing our overall spend because of all the writes for small assets (like search autocomplete API responses)

you can't control it at all beyond using a cache rule to mark the path not eligible for cache entirely