A question regarding Kopia Repository Server

General
1

Hi, after having read Repository Server | Kopia I am still not completely sure, how it works.Will all data transfers for all clients be done via the repository server if I use one?

My use case is to back up a virtual machine, a NAS (unraid) and some macOS and Linux computers to a (single) cloud storage location:

             +-------+
NAS -------> |       |
Laptop ----> | Cloud |
PC --------> | Store |
VM --------> |       |
             +-------+

However I would still like to have an overview (and control?) over all these backups from one location (preferrably via web UI on the VM). The Kopia Repository Server seems to offer this. But will the data transfers for the backups still be as described above, or will they all go through the VM:

NAS -----+
         | 
         v
Laptop --+
         | 
         v
PC ------+
         |      +-------+
         v      | Cloud |
         VM --> | Store |
                +-------+

?

2

Anyone? The answer highly influences the final layout of the backup system. I understand from the documentation, that Metadata access is relayed through the server, but does this also hold true for the backp (and restore) data transfers themselves?

3

It looks that all backup sources are yours - I would not use server, as for what?

Server makes sense when you use repository you do not want to share details with all users. Let’s say you let your friend to backup to your cloud storage but do not want to give him your cloud credentials. Then you spin kopia server on one of your machines and let your friend to connect to it. He can run his backups, you can run yours (either via kopia server or directly) but he only knows kopia server account credentials you created for him. And can only see his own backup.

As per docs:

By default, every user of Kopia repository directly connects to an underlying storage using read-write access. If the users who share the repository do not entirely trust each other, some malicious actors can delete repository data structures, causing data loss for others.

Repository Server allows an instance of Kopia to proxy access to the underlying storage and has Kopia clients proxy all access through it, only requiring a username and password to talk to the server without any knowledge of repository storage credentials.

In repository server mode, each user is limited to seeing their own snapshots and policy manifest without being able to access those from another user account.

Without server every individual kopia client has full access to all backups including ones from other machines. So in your case you can access on your Laptop backups created by NAS and vice versa etc.

IMO in your setup adding server is unnecessary addition you will have to manage without any clear benefits.

I would definitely go for:

             +-------+
NAS -------> |       |
Laptop ----> | Cloud |
PC --------> | Store |
VM --------> |       |
             +-------+
1 Like
4

Thank you @kapitainsky. That makes the concept somewhat clearer. The goal is, to control all backups (of NAS, laptop, PC ecc.) and have a “single point of overview” in one location. All Backups should also be restorable to this location (no matter, from which computer it has been made). However I’m currently not sure, if that is (easily) doable wit Kopia.

5

This is absolutely doable and easy with kopia and no need for server as discussed above.
As long as you connect to the repo with the cli you can list snapshots from all sources and restore to the device you are browsing the snapshots with.

6

Great. Thanks a lot @xxxliqu1dxxx for the clarification. I will try that as soon as I have solved my access issue