Does Cloudflare have plans to support cache range requests for large files to R2? Cloudfront+S3 supp
Does Cloudflare have plans to support cache range requests for large files to R2? Cloudfront+S3 supports it as well
UProbably not? The way cache works is they store the entire file in cache and then return ranges from that so if your file is larger than what can be cached it wont work.
UI do think there is a bug specifically in the R2 -> Cache behavior where if a file is too big for cache and you request a range it will ignore that and return the entire file where it should return the range and not cache (which I think it what the normal Origin -> Cache behavior is atm). But no clue whats going on around that 
USo a normal origin will also ignore the range request if the file is too big for cache? Not what I have heard from others in here who talk about this issue
UNot possible
UYou can use public urls with a custom domain but not presigned s3 urls.
https://developers.cloudflare.com/r2/buckets/public-buckets/
https://developers.cloudflare.com/r2/buckets/public-buckets/
UAfaik theres no plans to and I doubt it would happen. Why would you want it?
GCloudfront caches just the range you request. So they don’t only cache full files, but also sub-files. Seems to me Cloudflare could (should?) support that as well
GSo cache files aren’t only identified by their url, but also additionally with an optional range
SWhy don’t you want to share the R2 api domain?
CDoes anyone know how to use r2 with the cloudflare cdn
CI have a simple website where uses upload files to an r2, and can get files from an r2
CIt works but I want to use the cloudflare cdn with it and I can’t find anything in the docs that explains how
C
CIt’s weird that there isn’t much documentation since it’s one of the key features advertised
CCan you explain a little more, I haven’t set up the custom domain yet, still developing with the dev domain they give you
CLike when you set up the custom domain there’s an option for cdn integration?
UPublic Bucket is a feature that allows users to expose the contents of their R2 buckets directly to the Internet. By default, buckets are never …
UCustom domains have to use the CDN its not an option
COh I see
CThank you!
CAlso this may be a stupid question but say an r2 item is cached and 1000 users request the item, does that count as 1000 r2 read operations or just 1 for the initial retrieval from the r2 bucket
Mjust curious, I’m trying to authenticate r2 in a GitHub actions runner. But cloudflare keeps rejecting the ticket
Is this syntax correct or not?
run: |
curl -X PUT "https://api.cloudflare.com/client/v4/accounts/$ACCOUNT/r2/buckets/$BUCKET/objects/output.mp4" \
-H "Authorization: Bearer SECRET" \
-H "Content-Type: video/mp4" \
Is this syntax correct or not?
run: |
curl -X PUT "https://api.cloudflare.com/client/v4/accounts/$ACCOUNT/r2/buckets/$BUCKET/objects/output.mp4" \
-H "Authorization: Bearer SECRET" \
-H "Content-Type: video/mp4" \
UCache hits do not go to the R2 service so would not incur further reads. Cache is per data center though so not every requests will be guarantee to have it cached if they hit a different data center. Assuming 1000 requests all hit the same datacenter in a short period of time where the file is still cached it would only be 1 R2 read.
CCool, thanks
UI dont even think thats an api that exists, you would want to use the S3 API with an S3 API token
MI asked bing ai
MI suppose that’s why lol
SHey there, is there a way to enforce a max file size using S3 presigned URLs? I'm looking to upload files from a browser and using presigned URLs seems like the best way to go about this but don't want my users to be uploading massive files.
KNot currently - you can sign the
content-length header but that's only an exact length so you'd have to get the length the user plans to upload and then send them a URL for that length.
SThanks, I'll look into that :)
Dhow are people tackling post processing use cases with r2 when you can't listen to events?
DI'm thinking of an delayed action that triggers when the expected file should be uploaded
Dbut it'll be too slow
Dhoping to not have my server as the middleman handling the file upload since there's a lot of nuances to that :/
JMaybe a weird question, I assume there's no way to bind worker to R2 bucket in another account?
Jguess I'm moving some domains around, ty!
VNo info on global replication for R2 yet, right?
VSo they are single region, either EU or US based on where they are created
VDo R2 support the atomic read-after-write consistency of S3?
SYeah I think that if you create your bucket in Europe, all files are uploaded to Europe. Even when you're uploading from the USA. I might be wrong tho, correct me if I'm wrong.
VYeah so it's single region with no replcation ATM
