VRA fabric during VRA

[Rajisom]

In our customer environment, we are having a hybrid cloud deployment which spans across four DCs out of which two DCs are cloud DC from the vendor and two are customer Datacentre. We are having vSphere Endpoints on

all DCs.We are having three tenants (Dev /Test /Prod) and one Fabric Group for each Tenant. Each Fabric Groups contains ESXi clusters from all the four DCs.

The VRA appliances are hosted in Primary Cloud DC and are replicated via SRM to secondory DC. In case of a disaster, the recovery plan is to failover entire VRA appliances to secondory cloud DC and the VMs will start with same name and IP.

During DR we will be loosing one DC . In that scenario, when the VRA is started on the secondory site, will there be any issue in starting the Infrastructure Fabric and Fabric groups awhichs the clusters from one site is missing?

[daphnissov]

You'd have to test that scenario in your environment to know for sure, but my thought is there shouldn't be a problem, per se, with the fabric groups. You might have issues with your reservations depending on how you have those set up because, presumably, one entire datacenter (one endpoint, one cluster) is down and so provisioning cannot take place there. Any reservations which use that endpoint become unusable, so you must take that into consideration when you design your catalog and services. Also, you want to ensure whatever authentication sources you have set up will continue to work if one whole site fails.