Because DNS-based GLBs are only in path for DNS, they are only able to act on data in the DNS queries it receives when categorizing the client-side of connections that it is load-balancing.  The source-IP address of the DNS queries that it receives for the FQDN’s of GLB’ed services is the primary mechanism for that type of categorization.  In this architecture, the IP addresses of the recursive resolvers (not the clients) are exposed to the GLBs.  The upshot of this is that with session persistence enabled, the GLBs are required to make the same  load-balancing decision for each recursive resolver that it receives DNS queries from. This is not a problem in the short term, with relatively low requirements for multi-active site-level load-balancing, but that use-case is expected to increase in the near future.