I’m looking at setting up the replication so that we have a backup appliance on another server in case the first server has a catastrophic failure.
Our backup storage is on Minio on a separate server. The VM’s system drive and cache drive are on the same SSD dedicated to UrBackup on that Hyper-V server.
So my questions are:
- I believe that even if replication is enabled, the clients will only backup to one Server (let’s call it
App-1). Is this correct?
- Do I have to pay for 2 licenses or just 1?
- Will it replicate the config and the cache from App-1 to App-2? As the backups are on a “shared cloud storage” I’d prefer to have it just replicate configs/DB but I am not sure if that’s possible.
- The documentation doesn’t explain how to “switch-over” the client to backup to App-2 if App-1 is down.
- Will clients automatically switch back to App-1 if they were using App-2?
Apologies for all the questions, but I couldn’t find the answers in the admin_guide_app.pdf and I want to avoid any “gotchas” once we are fully invested into using UrBackup.
2 licenses (depending on if you have <10 clients of course otherwise you might be able to use the community edition).
It’ll replicate the data. If both appliances store to S3 you’d store double the amount of data…
Select a client on the status page (drop down), then select “migrate selected”.
I guess you could add the second appliance to the
Internet server name/IP setting separated by
server1.example.com;server2.example.com). If it can’t connect to the first it’ll connect to the second in the list (only then).
To clarify: The cache is only a cache. I.e. a subset of the data stored on S3 to speed up the appliance performance. If your S3 storage get’s corrupted/destroyed you can’t rely on the cache having any meaningful amount of data. The appliance can also be completely restored from S3. I.e. setup a new appliance on your Hyper-V server, point it at the S3 bucket and after it indexed the S3 bucket all backups and settings will be immediately available again.
An option is to use a local cached RAID configuration, then mirror that RAID to S3. That is a relatively new feature. Then, you’d have your data stored locally in the RAID and as complete mirror in the S3 bucket.