The new S3 storage - confusing fields?

fribse

Member
Feb 2, 2022
12
1
8
57
I was very excited to see the new S3 storage option, but the configuration seems overly complicated

From the PBS
1755326936868.png

Why do I have to create a pattern in the 'Endpoint' field?
I mean I have several backup systems going to the s3 storage, I've never seent that before?
It is quite confusing.
The rest is logical I guess.
But what of the local cache I have to set up, how big is that supposed to be?
 
Last edited:
I use BackBlaze S3, and it's very straight forward.
  • Endpoint is the FQDN of my Backblaze bucket
  • Access/Secret key are directly from BackBlaze
  • I had to check "Path Style" for it to work

2025-08-16_11-09-43.jpg
 
I was very excited to see the new S3 storage option, but the configuration seems overly complicated

From the PBS
View attachment 89512

Why do I have to create a pattern in the 'Endpoint' field?
I mean I have several backup systems going to the s3 storage, I've never seent that before?
It is quite confusing.
The rest is logical I guess.
But what of the local cache I have to set up, how big is that supposed to be?
The reason to use the patterns in the endpoint is so the same endpoint configurations can be reused for multiple buckets/datastores. Since the bucket is part of a datastore configuration, it is recommended to use the {{bucket}} pattern as placeholder (if you use vhost style bucket addressing that is).

The region is also part of the AWS request signature scheme, used to athenticate requests to the api. Therefore it is required to set it in it's dedicated field, the {{region}} pattern currently only being for convenience. You can set it manually as well, but it's recommended to use the pattern.