With Layer 2 VPN (L2 VPN), you can extend Layer 2 networks (VNIs or VLANs) across multiple sites on the same broadcast domain. This connection is secured with a route-based IPSec tunnel between the L2 VPN server and the L2 VPN client.
The extended network is a single subnet with a single broadcast domain, which means the VMs remain on the same subnet when they are moved between sites. The VMs' IP addresses do not change when they are moved. So, enterprises can seamlessly migrate VMs between network sites. The VMs can run on either VNI-based networks or VLAN-based networks. For cloud providers, L2 VPN provides a mechanism to onboard tenants without modifying existing IP addresses used by their workloads and applications.
In addition to supporting data center migration, an on-premises network extended with an L2 VPN is useful for a disaster recovery plan and dynamically engaging off-premise compute resources to meet the increased demand.
L2 VPN services are supported on both Tier-0 and Tier-1 gateways. Only one L2 VPN service (either client or server) can be configured for either Tier-0 or Tier-1 gateway.
Each L2 VPN session has one Generic Routing Encapsulation (GRE) tunnel. Tunnel redundancy is not supported. An L2 VPN session can extend up to 4094 L2 segments.
VLAN-based and VNI-based segments can be extended using L2 VPN service on an NSX Edge node that is managed in an NSX-T Data Center environment. You can extend L2 networks from VLAN to VNI, VLAN to VLAN, and VNI to VNI.
Segments can be connected to either Tier-0 or Tier-1 gateways and use L2 VPN services.
Also supported is VLAN trunking using an ESX NSX-managed virtual distributed switch (N-VDS). If there are sufficient compute and I/O resources, an NSX Edge cluster can extend multiple VLAN networks over a single interface using VLAN trunking.
Beginning with NSX-T Data Center 3.0, the L2 VPN path MTU discovery (PMTUD) feature is enabled by default. With the PMTUD enabled, the source host learns the path MTU value for the destination host through the L2 VPN tunnel and limits the length of the outgoing IP packet to the learned value. This feature helps avoid IP fragmentation and reassembly within the tunnel, as a result improving the L2 VPN performance.
The L2 VPN PMTUD feature is not applicable for non-IP, non-unicast, and unicast packets with the DF (Don’t Fragment) flag cleared. The global PMTU cache timer expires every 10 minutes. To disable or enable L2 VPN PMTUD feature, see Enable and Disable L2 VPN Path MTU Discovery.
- Between an NSX-T Data Center L2 VPN server and an L2 VPN client hosted on an NSX Edge that is managed in an NSX Data Center for vSphere environment. A managed L2 VPN client supports both VLANs and VNIs.
- Between an NSX-T Data Center L2 VPN server and an L2 VPN client hosted on a standalone or unmanaged NSX Edge. An unmanaged L2 VPN client supports VLANs only.
- Between an NSX-T Data Center L2 VPN server and an L2 VPN client hosted on an autonomous NSX Edge. An autonomous L2 VPN client supports VLANs only.
- Beginning with NSX-T Data Center 2.4 release, L2 VPN service support is available between an NSX-T Data Center L2 VPN server and NSX-T Data Center L2 VPN clients. In this scenario, you can extend the logical L2 segments between two on-premises software-defined data centers (SDDCs).
|L2 VPN Server Version (NSX-T Data Center)||L2 VPN Client Version (NSX-T Data Center)|
|3.1.2||3.1.2, 3.1.1, 2.5.3|
|3.1.1||3.1.1, 3.1.0, 3.0.1|
|3.1.0||3.1.0, 3.0.1, 3.0.0|
|3.0.3||3.0.3, 3.0.2, 3.0.1|
|3.0.2||3.0.2, 3.0.1, 2.5.2|
|3.0.0||3.0.0, 2.5.0, 2.5.1|