Hi!
So i am feeling sorta stupid so i figured id ask the experts.
I have a network spanning an entire city. In one end we have a datacenter with 2 vmware clusters. There is a Backup server in the building running Veeam Backup & Replication v8 with latest patch. We are now preparing a second identical Veeam Backup & Replication v8 server to place in another building on the other side of the city. The two offices are connected by a 1gb fiber link.
Question is this:
We don't have a seperate vmware cluster at the other building we simply want another backup server to be able to perform incase the first backup server dies for some weird reason. Should we setup a backup copy job to the second backup server OR should we run a backup replica (if its even possible).
Hi Wouter,
Replication is the best solution for the most critical applications in terms of RTO and RPO.
But since replication just makes a copy of the VM, in some cases like system failure or hacking it will not protect your infrastructure, so backups are also needed to be taken.
I would suggest to run both replication and backup jobs for the best availability and reliability.
Thanks.
P.S. Best practices are also described in this article.
You can not replicate to another Veeam backup server, since it`s not a hypervisor. In a case of disaster you will do a failover to a secondary site, there should be at least ESXi host, cluster is not required.
Thanks.
rinzler wrote:We don't have a seperate vmware cluster at the other building we simply want another backup server to be able to perform incase the first backup server dies for some weird reason. Should we setup a backup copy job to the second backup server OR should we run a backup replica (if its even possible).
Seems like one word is missing here. Do you want to do restores or backups if your first backup server dies?
If it is a restore operation, then use backup copy jobs to send backups offsite, then during the disaster import your backups to the secondary server and do the restore. If you need to do backups, then just deploy proxy server (managed by secondary server) on the main site and configure backup jobs pointing to existing repositories after your first backup server dies.
I guess we would mainly need to it to perform restores and backups. If server 1 which sits near the clusters dies then we want to on server 2 activate the jobs to run instead of the dead server.
