List the Machines in the Cluster or Appliance

Before the SAS High-Performance Analytics infrastructure can be installed on the analytics cluster, you must create a file named gridhosts that lists all of the machines in the cluster. The SAS High-Performance Analytics environment, SAS Plug-ins for Hadoop, and the SAS High-Performance Computing Management Console all use the gridhosts file for Message Passing Interface (MPI) communication. (The gridhosts file is copied to each machine in the cluster during the installation process. For more information, see Deploying SAS High-Performance Computing Management Console.)
Tip
You can use SAS High-Performance Computing Management Console to create and manage your gridhosts file. For more information, see the SAS High-Performance Computing Management Console: User's Guide.
On blade 0, create a file named gridhosts in /etc. (On Greenplum, blade 0 is known as the Master Server.)
In the gridhosts file, list one machine per line. You can use IP addresses or fully qualified domain names (FQDNs). However, all FQDNs must resolve to IP addresses and must be in the same DNS domain and sub-domain.
CAUTION:
The gridhosts file must contain only those machines that are members of your analytics cluster or data appliance. These machines are the NameNode (or Root Node) and its DataNodes (or Worker Nodes). If the management console is located on a machine that is not a member of the analytics cluster, then the console machine must also contain a copy of /etc/gridhosts with its FQDN added to the list of machines.
The root node is listed first. Depending on your data provider, the root node is also the machine that is configured as the following:
Here is an example of a gridhosts file:
grid001.example.com
grid002.example.com
grid003.example.com
grid004.example.com
...
Note: Make sure that there are no whitespace characters in your gridhosts file. The SAS High-Performance Analytics environment can skip entries when it encounters whitespace characters (such as tabs).
Last updated: June 19, 2017