Node Failure Recovery (NFR) | Teradata Data Mover - 17.10 - Node Failure Recovery (NFR) Considerations for Cloud - Teradata Data Mover

Teradata® Data Mover Installation, Configuration, and Upgrade Guide for Customers

Product
Teradata Data Mover
Release Number
17.10
Published
June 2021
Language
English (United States)
Last Update
2021-06-16
dita:mapPath
ivp1619423229496.ditamap
dita:ditavalPath
stp1591127384938.ditaval

Node Failure Recovery (NFR) is a cloud feature wherein Teradata SQL-E nodes are automatically replaced when a node fails. When NFR occurs, only static IP addresses associated with that node remains valid while non-static IP addresses are replaced. If a network fabric is defined using non-static IP addresses, that network fabric becomes invalid after NFR occurs.

To avoid this scenario, choose static IP addresses when defining a network fabric for a cloud-based SQL-E system. In AWS, these are referred to as elastic IP addresses and in Azure, these are referred to as static public IP addresses. On a SQL-E node, these are generally associated with the eth0:0 to eth0:3 network interfaces. Consult with a cloud expert when defining a network fabric to make sure that the IP addresses chosen are static IP addresses that are considered valid in the case of NFR.