The docs say:
Not planned: KV (a different consistency model), R2 (celld runs on blob storage; celld does not provide blob storage; declared r2_buckets bindings load, but each method throws) (…)
Although I fully understand the reason, since we're already deploying to AWS and using S3, implementing R2 bindings on top of S3 sounds like an effort worth pursuing.
For people that use R2 to dump DO SQLite data (either for backups, or for snapshotting to another DO instance), this is invaluable.
Any thoughts?
The docs say:
Although I fully understand the reason, since we're already deploying to AWS and using S3, implementing R2 bindings on top of S3 sounds like an effort worth pursuing.
For people that use R2 to dump DO SQLite data (either for backups, or for snapshotting to another DO instance), this is invaluable.
Any thoughts?