and web interface only supports file below 300MB which is much less then other providers (500MB)

Does custom metadata work with the S3 SDK?

Also great work on R2. I just hide in this Discord but I've been playing with R2 since beta last summer and I'm shocked it's already been a year.

Were you able to fins a solution to this?

For some reason when I try and upload a folder that's called accounts

wrangler r2 object put

fails:

bash
⛅️ wrangler 3.0.0
------------------
Creating object "accounts/test.txt" in bucket "mybucket".

✘ [ERROR] Failed to fetch /accounts/44321e1b7fbd29b56b65c1890a3bbee2/r2/buckets/mybucket/objects/accounts/test.txt - 404: Not found);

Sounds similar to https://github.com/cloudflare/workers-sdk/issues/2421

i dont think the api is a huge fan of things called accounts

Ah yea that could be it

There is no bucket in Australia, Singapore is the closest (fastest), depending the ISP.

You could set a policy to block access from any location other than Australia using CF rules, or from a worker. But a VPN could always bypass this by using an Australian server. If R2 adds replication at some point your data could be replicated in multiple areas. If you need the data within Australia, try Oracle, as the first 10TB/month is free then it's $0.3/GB. Or DO spaces with $0.1/GB egress from SYD.

Yeah, I think it's still unsupported for now. I just use Cyberduck.

anyone else having trouble with newer versions of aws-sdk-go (v1) and R2? any release from v1.44.265 and onwards results in silently failed uploads for me 🤔 https://github.com/aws/aws-sdk-go/releases/tag/v1.44.265

How can I have public access for my r2 bucket on my local wrangler/state/r2? I developed my application with the r2.get binding, but that's many extra requests that I could simply save by putting the bucket to public. This probably works, but I don't know how to do it locally while developing

If you add a custom domain to your bucket in the bucket settings, all files will be publicly readable on that domain with no authentication

but it's just the files that are actually on the bucket. In my local wrangler setup, the files will never be uploaded there because I have them locally in

.wrangler/state/r2

then. But someone in the #799437470004412476 channel suggested to just serve those files with any file server which I'll try now

Is there a way to trigger my nodejs server when a file is uploaded to r2? I will be using pre-signed urls.

how can I check that a bucket is really in the region i desire ?

I just created a bucket using the web interface, selected eastern europe as the location hint, and rclone still says it's on us-east-1

so it seems my "hint" is getting ignored 🤷‍♂️

It should return the hint

I don't know how to use that, i have always just used rclone to interact with r2

I even tried using a vpn to get a european ip address, went to the web interface and recreated the bucket...and rclone still says it's on us-east-1

us-east-1

doesn't 'exist' with R2, that's an S3 zone primarily so it likely just defaults to that

You can use or any of the SDKs in to do a

?location

call

i usually use aws cli https://discord.com/channels/595317990191398933/940663374377783388/1100486682593214554

I have this weird thing, not sure if this only happens to me? direct public links to r2 files, work on desktop, but don't work on IOS, no matter which browser (safari, chrome, brave) Both on wifi and on 4g. Any.. advice how to test this and what could be the issue?

Here's an example link but it happens with all R2 links it seems: ||https://pub-0a8ccdbc8ffd407ea09cfa21fdcc26ed.r2.dev/000aabe3c4602d64c3a973e56321e56e/000aabe3c4602d64c3a973e56321e56e.mp4||