Disabling the option Enable Service Engine Self-Election on the SE group configuration can cause health monitor failures for the pool members if the pools are not configured correctly.
In scale environments, there might be delay on the CRUD (Create, Read, Update, Delete) operations of the virtual service like Disable virtual service, Delete virtual service, enable the virtual service after disabling it) since this involves updating the underlying network configurations like Static Address and Service Insertion rules.
Preserve Client IP (IPv6) is supported with NSX version 4.1.2.1.1 and from NSX 4.2.
Preserve Client IPv6 is only supported for end-to-end IPv6 (i.e., VIP, Pool members, FIP should be all IPv6). Mixed mode of v4 and v6 is not supported.
Service Engine Data NIC and Pool member network segment should belong to the same NSX Tier-1.
NSX API rate limit might get hit in scale environments, it is recommended to increase or set appropriate limit as per the NSX API Guide.
When clients want to access Virtual Services hosting the Preserve Client IP service and connect to servers that are pool members of these Virtual Services, consider the following caveats when configuring your environment:
Clients cannot be connected to any segment of a Tier-1 gateway to which the Service Engines hosting Preserve Client IP services are attached.
Clients can be outside of the NSX-T overlay network provided that the VIP is reachable.
Clients can be connected to a segment of another Tier-1 gateway provided that the VIP is reachable.
Pool members (servers) must be attached to a segment of a Tier-1 gateway to which the Service Engines hosting Preserve Client IP services are attached.
Pool members cannot be attached to the same segment that the Service Engines themselves attach to.
Different pool members may be attached to different segments of this Tier-1 gateway if required.
