8 Configuring TCP/IP

Before You Begin

This chapter requires you to enter parameters that are unique to your system in order to properly configure TCP/IP. Refer to Appendix E for a worksheet listing all of the installation parameters that you will need to know about before you complete the configuration process. Ask your network administrator to provide you with these parameters before you begin.

Configuring TCP/IP

The following configuration example is for Microsoft Windows NT 4.0. Other Microsoft Windows operating systems may have these panels in a different location or may present them differently, but the required contents will be essentially the same.

You configure TCP/IP from the network configuration tool in the Microsoft Windows Control Panel.

Select the Protocol tab and select TCP/IP Protocol. Then, click Properties to display the Properties panel.

Figure 8-1 Network Configuration Tool

Description of "Figure 8-1 Network Configuration Tool"

If the TCP/IP Protocol is not already installed, then click Add and then select the TCP/IP Protocol.

Configuration consists of assigning a host name, an IP address, and a network mask to a given network interface.

In the IP Address tab, use the drop-down list to select the adapter you will use. Your network administrator can tell you whether you will be using DHCP or a static IP address. If using a static IP, then you must enter the correct values for IP address, subnet mask, and default gateway.

Figure 8-2 TCP/IP Properties Panel

Description of "Figure 8-2 TCP/IP Properties Panel"

Additional configuration consists of defining a name server IP address or creating entries in the hosts file on the local system. Name server translate host names into IP Addresses when queried on a particular host name. The hosts file provides this same functionality, but in a non-network participating manner.

The Hosts file may be edited with a text editor of your choice. For example, in Microsoft Windows NT, the file is located in:

C:\winnt\system32\drivers\etc\hosts

where C:\winnt is the Windows NT system root.

To use a name server, you must configure the TCP/IP to use DNS. Select the DNS tab and enter a host name and domain name. Your network administrator will provide these values. Click Add below the Domain Suffix Search Order box and enter the IP Address of the name server. You may enter up to three name servers. Click OK.

Figure 8-3 Define a Name Server

Description of "Figure 8-3 Define a Name Server"

For local configuration (in other words, the gateway and the DRDA Server are on the same system), it may be desirable to use the loop-back address. The IP address is 127.0.0.1 and is typically given the local name ("localhost" or "loopback") in the Hosts file. Using the loop-back address reduces the amount of network overhead by handling the traffic internally without actually talking to the network.

The gateway is configured for TCP/IP using the DRDA_CONNECT_PARM initialization file parameter. In an SNA configuration, this parameter would be set to the Side Information Profile name (name set in Figure 6-21 or Figure 7-17). In a TCP/IP configuration, this parameter should be set to the IP address or Host name of the DRDA Server, which should be followed by the Service Port number of that server. For more information about the port number, refer to "Port Number".

The rest of the DRDA-specific parameters are unrelated to the communications protocol and may be set the same for either SNA or TCP/IP installations.

Example #1: Configuration for a DRDA Server on a host named 'mvs01.domain.com' (or IP address of 192.168.1.2) with a Service Port number of 446.

DRDA_CONNECT_PARM=mvs01.domain.com:446

or

DRDA_CONNECT_PARM=192.168.1.2:446

Example #2: Configuration for a DRDA Server on the same host as the gateway with a Service Port number of 446.

DRDA_CONNECT_PARM=localhost:446

or

DRDA_CONNECT_PARM=127.0.0.1:446

For additional information on configuring TCP/IP, refer to the Microsoft Windows installation and configuration guides.