Adding Virtual Route Reflectors (VRRs) After Upgrading
To support high availability (HA) for Virtual Route Reflectors (VRRs), you must add VRRs and create redundancy groups after you upgrade to Contrail Service Orchestrator (CSO) Release 4.1.0.
To add VRRs:
- Log in to the installer VM as root.
- Navigate to the CSO Release 4.1.0 directory in the installer VM.
root@host:~/# cd Contrail_Service_Orchestration_4.1.0
- Run the add_vrr.sh script.
The existing VRR details are displayed.
=========== Existing VRR Details ======== host-name | redundancy-group vrr-126.96.36.199 0 =========================================
By default, VRRs that are created with CSO Release 3.2.1 or later belong to the redundancy group group 0.
- To add VRRs, you are prompted to answer the following
Specify whether VRR is behind NAT.
y—VRR is behind NAT. If you are deploying a VRR in a private network, the NAT instance translates all requests (BGP traffic) to a VRR from a public IP address to a private IP address.
n—VRR is not behind NAT (default).
Specify whether you want to use a common password for all VRRs.
If you want to use a common password for all VRRs, enter y and specify the common password.
If you want to use a different password for each VRR, enter n and specify the password for each VRR.
Specify the number of VRR instances.
For non-HA deployments, you must create at least one VRR.
For HA deployments, we recommend that you create VRRs in even numbers, and you must create at least two VRRs. Each VRR must be in a different redundancy group. If the primary VRR fails or connectivity is lost, the session remains active as the secondary VRR continues to receive and advertise LAN routes to a site, thereby providing redundancy.
For each VRR instance, specify the following:
Specify the public IP address for each VRR that you create. For example, 188.8.131.52/24.
Specify the redundancy group for each VRR that you have created.
For non-HA deployments, specify the redundancy group of the VRR as zero.
For HA deployments, the VRRs must be distributed among the redundancy groups. There can be two groups—group 0 and group 1. For example, if you have two VRRs, specify the redundancy group for VRR1 as 0 and the VRR2 as 1.
Specify the username for each VRR.
Specify the password for each VRR.
If you have chosen a common password for all VRRs, you are prompted to specify the common password only for the first VRR instance.
You can view the newly added VRRs through the APIs: routing-manager (GET: https://IP Address of Administration Portal/routing-manager/vrr-instance) or ems-central (GET: https://IP Address of Administration Portal/ems-central/device).
Each hub or spoke device establishes a BGP peering session with VRRs that you have created and assigned to different redundancy groups, thereby providing redundancy.