undefined | Better HN

0 pointsbrianwski4y ago0 comments

> how does the presigned URL work? I thought it was this - https://www.backblaze.com/b2/docs/b2_get_upload_url.html

If you want to use Amazon S3 APIs, you do not call ANYTHING that is documented on the Backblaze website, and you especially should not call "b2_get_upload_url" because that is a B2 native API, not an Amazon S3 API. You can always tell if you are using "B2 Native" if the call starts with "b2_" -> then that has literally nothing to do with Amazon S3 compatibility, it is the custom Backblaze protocol.

If you want to find out about Amazon S3 APIs (which you use to communicate to Backblaze's Storage Cloud or Amazon S3) then you can start here: https://docs.aws.amazon.com/general/latest/gr/signature-vers... Make sure you stay ENTIRELY on the Amazon website, and only read Amazon documentation, and use the APIs Amazon talks about (but of course you are doing all of this communicating with the Backblaze Storage Cloud backend). If any of that fails in your application, or in incompatible, PLEASE LET US KNOW!!

0 comments

segmondy4y ago

Got it! I thought the S3 API is a wrapper on B2 native API. Good to know

brianwskiOP4y ago

One interesting thing that Backblaze supports (that I don't think many people use) is that you can actually use any S3 API on any bucket, and any B2 API on the same bucket. Like every other call if you want.

So if you HAPPEN to find something is more clear with the B2 API, it's fine to use those calls on a bucket. If you find something is more clear in the S3 API, it's also fine to use those calls. The bucket won't get confused. :-)

j / k navigate · click thread line to collapse