iirc both should work but this is from a 2020 issue so maybe it has changed

Thank you very much for your help, I've made it, I've finished testing and am going into full production. Thanks again for all your assistance!!!

How do I get rid of these temporary files?

rclone cleanup

?

Do you need to wait or reconnect

Hi there, is it possible to create multiple R2 buckets with a unique API key for each bucket? So each API key will only be able to access files within that single bucket and not all buckets?

seems to only allow access to all buckets rather than specifying buckets?

You can’t currently scope tokens to specific buckets, correct

That's a shame

How much of a difference does bucket location make when serving files through workers in terms of latency and general upload/download times? Does it make sense to have multiple buckets in different regions and then let users choose which bucket to upload to depending on their location?

You’d need to consider how you then keep track of which file is in which bucket - I.E if you’re relying on GeoIP then that can change at any time, so you might have KV or D1 to keep track of filename -> bucket relationships.

Just wanted to see if anyone else has ideas for this. I will try uploading via the API later. Also wanted to ask if anyone knew a good way for me to automatically provide like a public directory linking for these. As right now I have been manually linking via HTML, which is a big hassle. Is there any good way to have all the folders in just a traditional file directory with pages?

That would be easy to do in my use-case. Just wondering if it's worth it.

As in, would there be a noticeable improvement in general upload/download times for my users.

Depends if your workload is write heavy - if not, just use cache

I would say it's write-heavy, yeah. I'll give it a try 🙏

Would be cool to have the region available on the UI too eventually.

To follow up on that, the object expiration worked perfectly!

Set it to 1 day and after 24 hours and 20 minutes the objects were gone.

Can I somehow get the Storage / Class A and B utilizations of a bucket via API?

You should be able to use the graphql API or at least that is what dashboard uses

When using the S3 API, what do I put in place of the AWS endpoint?

.r2.cloudflarestorage.com

Thank you

Also relevant: https://developers.cloudflare.com/r2/api/s3/tokens/

Found that already, but thank you 😄