Hi, is there any way to set a max upload size? I'm using presigned urls with the js sdk and unfortunately the Content-Length is not an option for me as I don't know the actual file size

can I use R2 for personal computer backups? I am using DigitalOcean Spaces, but I would like to use R2

You can if you want

Hello people, I've tried googling, but can't seem to find any information if versioning is planned for R2 objects any time soon?

Thanks Kian, i guess were no updates on this during developer week.

Try doing the logic of whether or not the file is acceptable via a worker, will be cheap/efficient to do so, unfortunately you can't trust the browser

That was my last resort, I was hoping to find something with less overhead but I guess that's the only way until ACLs are implemented. Thanks!

Can you rename object or directory?

No, but you can copy an object to a different key, and then delete the original.

I've stumbled across this message here, because I discovered a similar issue today, but not with pre-gzipped content, but just plain txt files. https://discord.com/channels/595317990191398933/940663374377783388/1086136363352596500 Chrome gives a "net::ERR_CONTENT_DECODING_FAILED 200 (OK)" error. Test file: https://pub-7e722c392cb24f389b2941db8812a3f0.r2.dev/test-files/file_4m.txt The files in question were uploaded in January 2023. It worked for some time, I don't know how long. Just noticed it because the part of the application is broken which uses this files.

I'm not entirely sure what encoding (if any) that is

The file content is just random output from /dev/urandom or something similar. We only use it's known size to measure the rough internet speed of a user.

Isn't this a case of Chrome (or any browser really) being told it's

text/plain

when it isn't?

Could be, but why should it break randomly? And as a counter argument: I've downloaded the files from the Dashboard and reuploaded them. They work fine

Also with content-type

text/plain

. https://pub-7e722c392cb24f389b2941db8812a3f0.r2.dev/test-files/test-files_file_4m.txt

The ETag is identical (other than one being weak) but the lack of

Content-Length

on the broken one would lead me towards an issue with transparent (de)compression.

Cache-Control

is present on the broken one but doesn't specify

no-transform

, I have no idea if `no-store`/`no-cache` influence it.

The R2 team should be able to tell you if there's been a change/regression in that area when they're next here

Thanks for looking into it @kian . Is there something I could have f'ed up here? If I remember correctly I've uploaded both versions via the Dashboard.

Nothing on the response headers looks wrong to me - I'd keep the broken file around in the bucket if you can so they can take a look at it

Will do. Those 100MB will certainly not make any noticeable difference in my invoice. 😁

Hey, someone knows how R2 bills look like? Is there is showing overall sum or if I have 2 or more buckets there is separation to have an understanding of which bucket how much you spent in the previous month

Have you created a ticket already? If not, can you do so so it can be escalated? That broken file might be of help too!

Not sure if the bill has it split out but the GQL API lets you get this info:

gql
query {
  viewer {
    accounts(filter: { accountTag: "<your account id>" }) {

      r2OperationsAdaptiveGroups(
        filter: {
          date_geq: "2023-04-01",
          date_leq: "2023-04-30"
        },
        limit: 10000
      ) {
        dimensions {
          bucketName
          actionType
        }
        sum {
          requests
        }
      }

      r2StorageAdaptiveGroups(
        filter: {
          date_geq: "2023-04-01",
          date_leq: "2023-04-30"
        },
        limit: 10000
      ) {
        dimensions {
          bucketName
          date
        }
        max {
          metadataSize
          payloadSize
        }
      }

    }
  }
}

(Note that storage groups has the date dimension there because it is

peak storage size each day / number of days

in the month)

Is there is a way to get that using S3 API/REST API, I'm not familiar with GQL API

Not that I know of but GQL is basically just a json interface and the query I sent is exactly what you would run 🤷 Writing GQL is actually very very easy especially if you use a GQL client that has auto complete and pulls up the docs in a side panel for you.

There's an undocumented (use at your own risk) /usage endpoint for the bucket that returns current size (which wouldn't exactly correspond to your bill since it's just now and not peak daily usage), but other then that, it's all GraphQL. The dashboard itself uses GraphQL to pull the Class A/Class B Operations, and all of the metrics

How to add metadata, such as Cache-Control: max-age to R2 files? Currently I'm using rclone to upload files.

Where I can get documentation for r2StorageAdaptiveGroups, r2OperationsAdaptiveGroups and if there is more?