Adding Windows server to multiple Veeam Servers

[unsichtbarre]

Hi all,

I was wondering about the possibility of using one high-capacity Windows server and adding it simultaneously to several Veeam Servers (as separate repositories on separate drive letters, of course)?

I am in need of a high-capacity solution to service several independent Veeam servers (running as VMs), for GFS storage only. I was wondering about getting a DL380 type server w/lots of disks, running Windows W2K16 and adding that server to several Veeam Server VMs for GFS Storage?

Any input or ideas?

THX in advance!

[csydas]

It's doable, but as you noted, keep everything separate! Learned this one the hard way and lost a lot of data >.< Annoying, but understandably my bad.

Also, a fun note that the Veeam Servers will apparently "be aware" of each other's task count, so you'll need to plan accordingly for concurrent tasks. You're also gonna have to handle upgrades to all three servers at the same time just to ensure that the components match.

So you can do this. But just be aware it might be a headache. What about presenting shares to the Veeam Servers as like SMB3 shares or iscsi shares? Maybe a green poster can comment on this, but in my mind this should keep stuff pretty safe and separate, give you the benefit of the big beefy server without having to juggle component versions, concurrent tasks, etc.

[unsichtbarre]

The recommendation is strongly in favor of adding repositories directly as Windows Servers. I could build the device as iSCSI, NFS, or Windows SMB - however the goal remains the same: providing vast quantities of storage for GFS jobs originating from several Veeam Servers

[nitramd]

Have you considered using a Cisco S3260? Anton gives a thumb-up for this piece equipment.

See this link: https://forums.veeam.com/veeam-backup-r ... 49703.html

[foggy]

Hi John, yes, you can do that, just as mentioned above, keep in mind the overall load of the server and, in case different Veeam B&R instances share the same server, maintain the same patch level of both instances.

[unsichtbarre]

We are considering an HP Apollo for this purpose. This is, or is similar to the Cisco SS3260 suggested above by nitramd. As an unrelated aside, I will never-again be burned by Cisco or their extreme dedication to throwing customers under the bus!