You can get the region via the S3 APIs, but not any more specific than that

I'll try that ty

Hello, I'm getting a weird issue with R2

I'm uploading a number of files with the following command

wrangler r2 object put --content-type  "text/css" -f <filepath>/index-0e61881a.css  <bucket>/onboarding-ui/assets/index-0e61881a.css

and for this specific file... R2 returns

text/plain

while for others css files it returns the correct

text/css

this is kind of baffling because I'm not doing anything strange with that specific file

the only thing I can think of is that this is a 34Kb css file, while the others are way smaller.

The R2 dashboard displays the correct content-type

am I missing anything?

Huh, just this one file is misbehaving? How are you fetching it?

How can I obtain the checksum of a downloaded object on the client, to verify integrity? For S3, the etag is equal to the MD5 of the body, but doesn't seem to be the case on R2

Ahhh... it's because I'm using range headers and R2 sends the etag for the full object, not the range. That's an S3 incompatibility then 🙂

Is there another way to check integrity, other than putting it into the object itself?

no it's not possible to move a bucket, even if you reuse the name and create again... once created it's stuck forever

cool thank you

Yes… it’s one file… I created a cors rule and I’m accessing the file directly

As you would to from a “website s3 bucket”

It works very well for everything… except that file and another css file (both the main css files of two interfaces im building with vite+tailwind)

I’d be happy to send you the file to see if you can reproduce

Why is it that when I upload files of 3GB or more it loads ad infinitum?

@User You can use these global flags

--header stringArray                   Set HTTP header for all transactions
--header-download stringArray          Set HTTP header for download transactions
--header-upload stringArray            Set HTTP header for upload transactions

https://rclone.org/flags/

Right yeah, our ETags aren’t MD5 hashes. The only way to do integrity checks right now is to store a hash as custom metadata next to the object itself, like you said

I’m not sure I understand, you’re using multipart uploads to upload your files right? Are you referring to the greyed out rows? If so, they just represent multipart uploads that haven’t been completed yet. Have you issued a CompleteMultipartUpload for these uploads?

I have a question about R2 wrangler commands or api... is it possible to bulk upload some files like we can do for KV?

asking this again, how can i fix winscp not able to upload???

is there an eta on replication 🙏

what fastest speed could be while downloading from r2 bucket?

ive seen up to usually 65 mb/s (could be completely wrong)

using public domains, downloading straight from the bucket with s3 was 115 mb/s

If you use a public R2 domain, and the files are cached, the download speed should be similiar because you'll use the Cloudflare CDN