Best practices for setting up a new Windows machine?

Only have a single drive with truly critical data on it. Anything important goes there. And we’ve always backed up that drive to Amazon S3.

I’d like to revamp this old setup to use Kopia while maybe retaining the versioning features other S3 clients provide? What is the best way to do this?

And if I already have KopiaUI on this Windows box, how could I use Kopia to save routine (less important backups) to networked RAID1? I would think this would require a separate repository? Could I use the existing repository already on that RAID from another box/client as a backup target for this new Windows box? In the future, could I also store Linux backups to this same repository?

Not really sure of myself, so I’m asking if this is safe. Feel free to tear these ideas up.

3-2-1 backup:

  1. One copy, is primary, working file
  2. Second copy at local backup (external drive, NAS)
  3. Third - out of premises copy

Make local backup to a network, then use : sync to s3. Both copies will identical. You can sync repository(ies) to any supported backends as much as needed. Basically sync as just a coping

Thanks!

I assume that local copy is for speed, and that any transmission anomalies will face eventual correction via “sync to S3”?

Not only for speed. If you coping repository to multiple places, you want it be the same everywhere. Second, you wouldn’t load subject for backup (host you backuping) multiple times since it is expensive operation, while “sync to S3” works kinda like rsync, it won’t push the whole repository, but the only accumulated changes

1 Like

Thanks. You saved me a lot of time with this. Don’t have a feel for Kopia’s true “envelope” yet, but I’ve had some failures which did help outline places where “not to go”. This comment saved me much additional experimentation.