Multihomed Environment

Has anyone deployed a RR instance multihomed to multiple networks? We currently do this in order to grab backups from different networks and it seems to me that I tried this with RR a few years back and could never get it to work correctly.

If this is going to be viable in my current org a each node we deploy will need to be multihomed. Curious if anyone here has done this.

  • Morning. Without sounding naive here, I would define some of our RR instances as multihomed. As a couple of our RR cores connect to completely separate VLANs. Now, in our circumstance it made it relatively easy to expose the other agents/vCenters as they are indeed on separate VLANs so when the call to those IPs are made the network arrives at the correct location. What are you trying to get working/stood up? Now I'm curious as unless I'm missing something obvious I would assume that as long as the network is configured appropriately, or the correct routes/VLANs/vNIC/whichever are exposed I don't see why it wouldn't work. There very well could be a variable I am missing there though. 

  • So as an example one thing I was trying to do at my previous org was push replication over a dedicated network (using a dedicated physical NIC on physical servers) while the backups themselves pulled/pushed from another network on another NIC. It never worked. Everything was reachable but it simply failed to even start as I recall. I think Veeam does this using "proxy servers" to manage traffic.

    Currently we backup over many disparate networks. For example one for DR, one from prod, one for dev, offsite, etc.Some of the Veeam proxy servers sit in vlans that can reach both and some move by script to backup shares.

  • Gotcha. Yup, within RR itself there is not a proxy setting. However where we have multiple RR instances underneath a single router we just add firewall exceptions in-between vlans (if I am understanding you correctly). I'd be curious what issues you'd run into - technically as long as you have port 443/8006 open and the servers are patched RR tends to work for us. I understand what you mean in regards to the Veeam proxy, I'm quite familiar with that as well, we've been able to roughly achieve the same function of that by VLAN'ing and firewall rules. Happy to have a discussion, should be able to function though I'm curious where you got stopped.