Skip to navigation
FeaturesCustom Storage Configurations

Cloudflare R2 Custom Object Storage

View as Markdown

Cloudflare R2 Custom Object Storage

MeetStream can write meeting recordings, transcripts, screenshots, chat logs, and manifests to a Cloudflare R2 bucket through R2’s S3-compatible API. Configure the destination in the MeetStream dashboard. Your bot and media fetch API calls continue to work as described in the custom storage overview.

1) Prerequisites

  1. Create an R2 bucket in the location or jurisdiction appropriate for your data.
  2. Create an R2 S3 API token with Object Read & Write permission restricted to that bucket.
  3. Save the token’s Access Key ID and Secret Access Key. Cloudflare shows the secret only once.
  4. Copy the 32-character Cloudflare account ID for the bucket.

R2 tokens can be scoped to a bucket, but not to an object-key prefix. Use a dedicated bucket when you need stronger isolation for MeetStream media.

2) Configure R2 in the dashboard

Open Custom Storage in the MeetStream dashboard, add a Cloudflare R2 destination, and enter the bucket name, account ID, and S3 access key pair. Choose a base object-key prefix or separate prefixes for audio, video, transcripts, and metadata, then save the destination.

SettingValue
ProviderCloudflare R2 (cloudflare_r2)
Bucket nameYour R2 bucket name
Account IDYour 32-character hexadecimal Cloudflare account ID
Regionauto, if the dashboard asks for a region; no other value is accepted
Access Key ID and Secret Access KeyThe R2 S3 API token’s key pair
Access moderead_write only
Base prefixOptional; defaults to meetstream
Category prefixesOptional overrides for audio, video, transcript, and metadata

MeetStream derives the R2 S3 endpoint from the account ID:

https://<ACCOUNT_ID>.r2.cloudflarestorage.com

R2 does not accept an endpoint URL override. MeetStream-generated presigned URLs use the account-scoped S3 endpoint. Keep the access keys in the dashboard; do not put them in client-side code.

You can save multiple destinations, including multiple R2 buckets. Manage the default in the MeetStream dashboard, and use storage_config_id to select a saved destination for a bot.

3) Permissions and validation

The R2 token needs Object Read & Write permission for the selected bucket. MeetStream treats HeadBucket as diagnostic and validates write, read, and delete access with temporary objects under the configured prefixes before saving the destination. Cleanup of R2 validation objects must succeed.

Cloudflare R2 supports only read_write access in MeetStream. A write_only configuration is not available.

4) Fetch media

MeetStream writes artifacts below the resolved category prefix and bot ID:

<category-prefix>/<bot_id>/<artifact>

Existing artifacts stay in their original bucket and prefix if you change a destination later. Keep credentials for earlier buckets available when you need to retrieve older media.

Your MeetStream media fetch calls do not change. They return R2 presigned URLs for recordings or read JSON artifacts from the bucket. Presigned URL lifetimes vary by artifact; see Usage & Retention. Call the fetch endpoint again for a fresh URL.

5) Troubleshooting

The dashboard rejects the R2 destination

  • Check that the account ID is a 32-character hexadecimal value and matches the bucket.
  • Use auto as the region if prompted; geographic region values are not valid for R2’s S3 API.
  • Use an R2 S3 Access Key ID and Secret Access Key from an Object Read & Write token scoped to the bucket.
  • Do not enter an endpoint URL override.

A presigned URL returns 403

Fetch a new URL if it has expired. Check that the configured credentials still have object read access and that the account ID and bucket name are correct.

Browser requests to R2 fail

If your application fetches presigned URLs from a browser, configure the bucket’s CORS policy for your application’s exact origins and methods. MeetStream does not change your R2 bucket’s CORS settings. Do not use an R2 custom domain for MeetStream presigned URLs.

Files are not in a newly selected bucket

Only artifacts processed after the destination change use the new bucket. Earlier artifacts remain in their original bucket.

Provider references