confusion and frustration with failover

[lemmy]

been trying to setup a working failover. I think at this point i'm either missing something or have a design bug.

setup : Main production site A with esxi 5.1 on network 10.15.0.0/16 Disaster site B with esxi 5.1 on network 10.16.0.0/16.

* Replication works fine. We can get our VMs updated multiple times a day.
* Our goal with replication is that if we loose site A, we can fire up a VM on site B and let it take over for the failed VMs from site A until site A can come back online.


Q: In the scenario of a simple file server, how does failover work? Once I initiate a fail over, Does the distant server take on the IP of my production VM from site A? How would it do this considering my sites must have separate IP ranges ? Right now the failover seems to work from a Veeam standpoint, but we have no connectivity with the machine running on the other site.

what am i missing here?

[Vitaliy S.]

Once I initiate a fail over, Does the distant server take on the IP of my production VM from site A?

Do you use DHCP server or static IP addresses on your main site? Replicated VMs will be using the same IP addresses as your production servers unless you assign new IP addresses to them via DHCP server (located on the remote site) or Re-IP rules configured within Veeam backup console.

How would it do this considering my sites must have separate IP ranges ?
Right now the failover seems to work from a Veeam standpoint, but we have no connectivity with the machine running on the other site.

Do you have any routing configured between two sites? When you perform failover to the DR site, you should also configure routing on the main site to forward all network packets/requests to the DR location.

[veremin]

Also, assuming that your backup server currently is deployed at production, it stands to reason to install additional instance at remote location and set it responsible for replication jobs only. Should any disaster happen (primary site goes down, for instance), the required operations (Failover, Failback) can be easily started from additional backup server. Thanks.