Store working data directly in S3

[baby-gnu]

Hello.

Detailed Description

We want the ~/.aptly/pool and ~/.aptly/db directories to be stored on S3 buckets.

This way, even if the host doing the work crash, we just have to keep the ~/.aptly.conf in a VCS to make it working.

Context

We are actually using the Ubuntu mirror which use rsync and we want to store that mirror in S3.

We are thinking about using aptly because of its easy snapshot feature and S3 support but the actual workflow requires to store twice the data.

In case of a full Ubuntu mirror this means 2 * 1.2TB.

I saw references to see functionality in #230.

Possible Implementation

Unknown.

Your Environment

Distributor ID:	Debian
Description:	Debian GNU/Linux 10 (buster)
Release:	10
Codename:	buster

Package: aptly
Version: 1.3.0+ds1-2.2~deb10u1

[emopinata]

This would be fantastic, especially when your aptly instance runs in AWS. Then you could just use aptly in a container and only run it when you need it.

...hamster wheel starts turning...

[lubo]

I believe this could be partially solved by #1074.