Still having some trouble getting my head around a good replication design. I hope you can give me some answers on this one.
Using the documentation, page 134 Veeam Backup & Replication for VMware | USER GUIDE | REV 7:http://veeampdf.s3.amazonaws.com/guide/veeam_backup_8_userguide_vmware.pdfOffsite replication over a direct path requires the following backup infrastructure components:
1. At least one backup proxy in the source site. The backup proxy must have access to the
Veeam backup server, source host, backup proxy in the target site and backup repository
holding replica metadata.
2. At least one backup proxy in the target site. The backup proxy must have access to the Veeam
backup server, target host and backup proxy in the source site.
3. Backup repository for storing replica metadata. The backup repository must be located in the
source site, closer to the backup proxy, and must have access to it.
When I would like to do this with the least possible number of VMs and make sure that target and source ESXi DON'T have a direct connection, can I then implement the following design (and get it working)?
At the source site, I will only have a backup proxy that can see the backup proxy at the target site. This backup proxy at the target site, is also the VEEAM B&R Server and a Backup Repository.
- Am I correct that the Backup Repository can be a local Windows disk on the target backup proxy / B&R Server?
- I read: "Veeam Backup & Replication writes information about every data block to the VM replica metadata stored in the backup repository. The more data blocks there are, the more metadata is written to the backup repository." Is there a ball park figure for 5 TB data? Should I think of 1 GB / 10 GB / 100GB ?
- In this scenario only the VMs that are proxies can see each other on the other site, no other cross-site traffic is needed. Correct?