Cloudflare

I got this in r2OperationsAdaptiveGroups, who knows why I'm getting actionType: unknown?
https://cdn.discordapp.com/attachments/940663374377783388/1111340642153533521/image.png

GQL docs are automatic so I would just use a query explorer that supports it... I use altair. As far as the unknown action type no clue never seen that 🤔

I found site where there is nice documentation, can I post it here?

is it <https://pages.johnspurlock.com/graphql-schema-docs/cloudflare.html> <:meowlul:364104895638536202>

https://cfdata.lol/graphql/ has all of the schema

Yea that's by a community champ, @ James, a useful tool

An actual client with schema introspection like GraphiQL is still more useful though, but that be helpful as well

I prefer the docs inside my query editor personally 🤷
https://cdn.discordapp.com/attachments/940663374377783388/1111352524671287346/image.png

Yeah GraphQL inspector is wildly more useful. My dump at cfdata.lol is useful for a quick "what was this field again" without opening up an entire explorer though 😄

looks like the unknowns are all errors if you add the actionStatus dimension

(guessing it couldn't process the action because it failed)

Ah probably a request without a valid action or something bad body cant parse action etc

nvm I'm blind, in his screenshot he shows a few that succeeded, interesting

this is the part you use dev tools and steal from the dashboard for

Class A Query:
```graphql
query {
  viewer {
    accounts(filter: { accountTag: "b9e510810e1e12fcd40826d266a1fb12"}) {
      r2OperationsAdaptiveGroups(
        filter: {
          date_geq: "2023-04-01",
          date_leq: "2023-04-30",
          actionType_in:["ListBuckets","PutBucket","ListObjects","PutObject","CopyObject","CompleteMultipartUpload","CreateMultipartUpload","UploadPart","UploadPartCopy","PutBucketEncryption","ListMultipartUploads"],
          actionStatus_in:["success","userError"]
        },
        limit: 10000
      ) {
        dimensions {
          bucketName
        }
        sum {
          requests
        }
      }
    }
  }
}
```
Class B query:
```graphql
query {
  viewer {
    accounts(filter: { accountTag: "b9e510810e1e12fcd40826d266a1fb12"}) {
      r2OperationsAdaptiveGroups(
        filter: {
          date_geq: "2023-04-01",
          date_leq: "2023-04-30",
          actionType_in:["HeadBucket","HeadObject","GetObject","ReportUsageSummary","GetBucketEncryption","GetBucketLocation"],
          actionStatus_in:["success","userError"]
        },
        limit: 10000
      ) {
        dimensions {
          bucketName
        }
        sum {
          requests
        }
      }
    }
  }
}
```

Replace Account Id with your own. That's the same queries the dashboard uses for Class A/ Class B showing

this is the part you use dev tools and

I just made a script that collects all this information and gives me more generalized statistics
https://cdn.discordapp.com/attachments/940663374377783388/1111355624597508218/image.png

I know this is an old post but did you get the redirect rule working? Or did you wind up going with worker?

hello everyone im having an issue deleting an r2 it wont let me due to failed uploads and the failed uploads check box is grayed out :/

Hi,

Is it possible to directly upload files from browser to R2?
If its not supported, what is the preferred way to upload user files from browser to R2?

The only way we can upload the user files to R2 is by using a Worker as proxy?

Do you want users to upload files to your R2 bucket or you uploading files to your bucket?

For the first one you could use the S3-compatible API with Presigned URLs to let users upload files to your bucket. 
For the second one there is a possibility to upload files (up to 300MB) via the Dashboard.

<@385840518782910465> - Thanks for the reply.

I want my application users to upload files to R2 bucket. 
With reference to "S3-compatible API with Presigned URLs" :
This will be a 2 step process, where we generate presigned URL first and using the presigned URL, upload the actual file. Is my assumption correct?

It depends. You can do two things: 
- Let the users upload the files directly from the browser to the R2 bucket -> Presigned URL
- Let the users upload files to your server and from there upload them to the R2 Bucket -> R2 credentials

The difference is: With the presigned URL you generate a kind of "restricted" access token to your bucket with your R2 credentials. The restricted token can only access the exact path and method you specify while your R2 credentials can access all of your buckets. You never want to expose your R2 credentials anywhere to your users.

With a presigned URL you say for example: Your user is allowed to do a `PutObject` command on my bucket `appbucket` for the key (filepath) `user-uploads/123456.jpg` for the next 120 seconds.

Makes sense. Thanks <@385840518782910465>

hi there how can i tell where my r2 storage is and is it possible to move it somewhere else?

i expect its western europe but if i could find out the exact location i'll rent more servers there to get the latency faster

Have you created the bucket with location hints? If so this would be the closest guess to where the bucket is regionally located.

no i didn't sadly but even then "western europe"! is a bit general

Latency within western Europe is generally not a big difference. I don't think CF gives any more details in general which pops can be origins for R2 buckets.

Yeah I mean i'ts good about 1 something milliseconds but it's a gaming thing so every ms counts lols