Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

 
 

Advanced Security and Data Center Interconnect Configurations

Use these examples to configure advanced security and DCI on your collapsed spine data center architecture.

Configure Advanced Security for Inter-Tenant Traffic

The SRX Series is a next-generation firewall that can provide advanced security services for inter-tenant traffic. Use this section to route inter-tenant traffic between JNPR_1 and JNPR_2 in DC1 through the SRX chassis cluster.

Requirements

Overview

The SRX Series devices in your chassis cluster operate as a single device to provide device, interface, and service-level redundancy. Use this section to separate the chassis cluster into zones and configure the routing policies so that the correct traffic is routed through the security devices.

Topology

Both spine switches are physically connected to both SRX nodes as shown in Figure 1.

Figure 1: Physical Topology of SRX Cluster Physical Topology of SRX Cluster
Note:

This example is based on SRX345 devices. Once placed into a HA cluster, the interfaces on node 1 are associated with FPC slot 5. This means the ge-0/0/11 interface shown for node 1 is actually configured as ge-5-0/11 once the cluster is formed. The FPC number for node 1 in a HA cluster can vary by SRX model type.

Reth1 is a logical interface in the SRX cluster. It is active on one of the nodes of the SRX cluster. If the primary node or interconnect link between the SRX devices and the spine switches fails, Reth1 will failover to the secondary node. Figure 2 shows the logical interfaces between the SRX devices and the spine switches.

Figure 2: Overlay Topology of SRX Cluster Overlay Topology of SRX Cluster

Each spine switch establishes separate EBGP peerings with the SRX cluster in each routing instance or tenant as shown in Figure 3. For example, Spine 1 has two peerings with the SRX cluster, one in each routing instance: JNPR_1 and JNPR_2. Reth1.991 peers with the JNPR_1 routing instance on the spine switches and belongs to the JNPR_1 security zone. Reth1.992 peers with the JNPR_2 routing instance on the spine switches and belongs to the JNPR_2 security zone.

The SRX device advertises a summary route that covers all prefixes (for example, 192.168.0.0/16). The spine switches advertise specific subnets in each routing instance.

Figure 3: Topology of SRX Cluster with EBGP Peering Topology of SRX Cluster with EBGP Peering

Configure the Interfaces

Configure the SRX Device

Step-by-Step Procedure
  1. Configure the group for the logical interfaces on the SRX device.

  2. Configure the logical interfaces. Reth1 is a tagged Layer 3 interface on the SRX cluster. Reth1.991 peers with the JNPR_1 routing instance on the spine switches. Reth1.992 peers with the JNPR_2 routing instance on the spine switches.

  3. Place the logical interfaces into separate security zones. Reth1.991 belongs in the JNPR_1 security zone and Reth1.992 belongs in the JNPR_2 security zone.

  4. Check the status of the chassis cluster.

Configure Spine 1

Step-by-Step Procedure
  1. Configure the SRX device interconnected interfaces on Spine 1.

  2. Configure IRB interfaces.

  3. Configure the VLANs.

  4. Configure VNIs as part of the EVPN MP-BGP domain.

Configure Spine 2

Step-by-Step Procedure
  1. Configure the SRX device interconnected interfaces on Spine 2.

  2. Configure IRB interfaces.

  3. Configure the VLANs.

  4. Configure VNIs as part of the EVPN MP-BGP domain.

Configure EBGP

Configure the SRX Device

Step-by-Step Procedure
  1. Configure the EBGP interconnect.

  2. Configure the routing options.

  3. Configure the policy options.

Configure Spine 1

Step-by-Step Procedure
  1. Configure the EBGP peerings in the JNPR_1 routing instance.

  2. Configure the EBGP peerings in the JNPR_2 routing instance.

  3. Configure the import and export policies for interconnect with the SRX device.

Configure Spine 2

Step-by-Step Procedure
  1. Configure the EBGP peerings in the JNPR_1 routing instance.

  2. Configure the EBGP peerings in the JNPR_2 routing instance.

  3. Configure the import and export policies for interconnect with the SRX device.

Configure the SRX Device Security Policies

Step-by-Step Procedure
  1. Configure the security policies in Zone 1 for JNPR_1.

  2. Configure the security policies in Zone 1 for JNPR_2.

Verify BGP on the SRX Chassis Cluster

Step-by-Step Procedure
  1. Ensure that all BGP peering sessions with the spine switches are established.

  2. Verify the SRX device received the BGP routes from the JNPR_1 tenant.

  3. Verify the SRX device received the BGP routes from the JNPR_2 tenant.

  4. Verify that the SRX chassis cluster is advertising a summary route to the spine devices.

  5. Verify inter-tenant traffic through the SRX chassis cluster.

    In this example, Endpoint12 is part of VLAN 212 and tenant JNPR_2. Endpoint12 is pinging Endpoint2, which is part of VLAN 201 and tenant JNPR_1, as shown in Figure 4. Since this is inter-tenant traffic, this traffic goes through the active member of the SRX chassis cluster. SRX-Node0 is the active member of the SRX chassis cluster and SRX-Node1 is the passive member.

    Figure 4: Inter-Tenant Traffic Through the SRX ClusterInter-Tenant Traffic Through the SRX Cluster

    Confirm that the flow table on the SRX device shows this traffic traversing the SRX chassis cluster.

    You have configured advanced security for your data center and confirmed that inter-tenant traffic is routed through the SRX chassis cluster.

Configure Data Center Interconnect (DCI)

Overview

Now that you have configured a collapsed spine architecture for both data centers and added advanced security to DC1, it is time to connect DC1 and DC2 using Data Center Interconnect (DCI).

Topology

In this example, there is no need to stretch Layer 2 between data centers. Inter-data center communication is routed through the SRX chassis cluster in DC1, as shown in Figure 5. The spine switches each have a WAN routing instance and are connected to the WAN between data centers. The spine switches hand off the Layer 3 routes to the WAN router (not shown in this figure).

The SRX chassic cluster is advertising a 192.168.0.0/16 subnet. The DC2 spine switches Spine 3 and Spine 4 are advertising the two subnets 192.168.221.0/24 and 192.168.222.0/24.

Figure 5: Data Center Interconnect TopologyData Center Interconnect Topology

Each SRX device is configured with three zones that correspond to the JNPR_1, JNPR_2, and WAN routing instances. All inter-tenant traffic between JNPR_1 and JNPR_2 is routed through the SRX chassis cluster. All traffic between DC1 and DC2 is routed through the SRX chassis cluster using the WAN routing instance. Each SRX device has individual EBGP peering with Spine 1 and Spine 2 in each of the routing instances. Figure 6 shows the EBGP peering between the spine switches and the SRX chassis cluster in DC1.

Figure 6: SRX Chassis Cluster EBGP Peering TopologySRX Chassis Cluster EBGP Peering Topology

Configuration

Configure the SRX Device

Step-by-Step Procedure

Each SRX device must be divided into three zones that correspond to the three routing instances: JNPR_1, JNPR_2, and WAN. You already created the JNPR_1 zone and the JNPR_2 zone in Configure Advanced Security for Inter-Tenant Traffic.

  1. Add a new sub interface on Reth1 for the WAN interconnect.

  2. Configure the WAN security zone.

  3. Configure EBGP for the WAN security zone.

  4. Configure the security policies. For simplicity, the security policies in this example are open. In your setup, modify the security policies as necessary.

Configure Spine Switches

Step-by-Step Procedure
  1. Configure the routing instances and irb interface on Spine 1.

  2. Configure the routing instances on Spine 2.

  3. Configure EBGP on Spine 3.

  4. Configure EBGP on Spine 4.

Verify DCI Routes

Step-by-Step Procedure
  1. Verify the routes on the SRX chassis cluster. The SRX should learn all the specific routes for the different subnets.

  2. Verify the routes on Spine 1 and Spine 2. The SRX cluster advertises the 192.168.0.0/16 summary route to the spine devices on all the VRFs. All inter-VRF traffic and DCI traffic goes through the SRX chassis cluster.

  3. Verify the routes on Spine 3 and Spine 4. The DC2 spine devices receive the aggregate route from the WAN VRFs on the DC1 spine devices. All traffic between the two data centers is routed through the SRX chassis cluster.

    You have connected your collapsed spine data center networks with DCI.