> Network Administrator Guide > Understanding Network Transport Technologies

8.2.6b

Chapter 6b - (2/2)

Understanding Network Transport Technologies

The application-to-application network data communications message transport uses network protocols like TCP/IP and Novell SPX to establish connections between cc:Mail Routers and cc:Mail Mobile applications. This is sometimes referred to as native transport packet protocol support. This means that the cc:Mail Router has built-in support for network protocols or native packet transport protocol support.

Unlike the file I/O message transport, which is network independent, application-to-application network data communications involves the direct use use of specific network protocols to establish a dialog between cc:Mail Routers and cc:Mail Mobile applications over a network in realtime. Remember that the cc:Mail user mail applications are database applications that access files on a file server rather than data communications applications. The cc:Mail Router, however, is both a database application and a true data communications application.

Figure 5.3 illustrates application-to-application network data communications showing the cc:Mail Directory entries, Call List entries, cc:Mail addresses and drive mappings for asynchronous communications using modems.

Figure 5.3: Peer-to-Peer Message Routing Using TCP/IP (Type II connection using a network protocol)

Setting up the Router Call List

The call list is administered with the cc:Mail administration program. You must run the cc:Mail administration program with the command line parameter /CALLLIST/Y as shown below to enable its call list management feature.

ADMIN /Nponame /Ppassword /Dpo-path /CALLLIST/Y

You only need to use the /CALLLIST/Y parameter the first time you use the cc:Mail administration program to manage the call list. Once you have enabled this feature for a given post office, call list administration will automatically be available when you run the cc:Mail administration program.

Verifying the post office addresses

Before setting up the call list at the originating post office, you must verify that there is a directory entry in the post office for each of the post offices to be called, and that the address for each one is correct for the connection method you plan to use. A typical directory might look something like that shown in Figure 5.5.

Adding entries to the Call List

Once you have verified the addresses for the post offices you want to call, select "Manage Call List" from the cc:Mail administration program’s Main Menu to bring up the call list entry screen like that shown in Figure 5.6.

To add an entry to the call list, follow these steps:

1.Enter a call time. This can be one of the following formats:

2. Select the post office to call from the directory as shown in Figure 5.7.

The entry should now appear in the call list as shown in Figure 5.8.

The call list entry has default values automatically set for the other fields which appear across the top of the call list. These fields are described as follows:

If you set the priority to Urgent, normal priority messages will not cause the cc:Mail Router to initiate a call.

Modifying Call List Entries

If you want to change any parameters associated with an existing call list entry, highlight the call list entry and press Enter to display the call list menu as shown in Figure 5.9.

Using this menu you can modify or delete the selected entry from the call list. For a detailed description of any of the items, highlight the menu item and press the F1 key. When you have finished entering and modifying the call list, press Escape to return to the Main Menu.

Entering Multiple Call List Entries for the same Post Office

There are many instances where it makes sense to include multiple call list entries for calls being placed to the same post office. You can set up each call list entry with different settings for message size, hours, etc., depending on how you want to customize the characteristics of cc:Mail Router calls based on certain conditions. Here are some examples where multiple call list entries would be useful:

1. Exchange only messages smaller than a certain size during the day, and larger messages at night, when the telephone rates are lower. (See entries for ENCORE-FRANCE in Figure 5.10)

2. Exchange messages 24 hours a day, but exchange directory updates only during non-peak hours. (See entries for ENCORE-SF in Figure 5.10)

3. Deliver urgent messages within 5 minutes, but normal or low priority messages only every 20 minutes. (See entries for ENCORE-HQ in Figure 5.10)

Changing the Auto Call Interval and Retries

If you want to change the automatic call interval or the default retry settings for the call list, select "Change

Post Office Profile" from the Main Menu. The lower left corner of the post office profile shows the call list defaults as shown in Figure 5.11.

Changing the auto call interval will affect all call list entries that have the call time specified as only "Auto" rather than "Auto/n". Retries and Retry Interval specify how many times the cc:Mail Router should retry if a connection was unsuccessful (if the modem gets a busy signal, for example), and how long it should wait between retries.

Setting up the cc:Mail Router Workstation

The cc:Mail Router workstation should have a network card installed and the ability to log in to the file where the originating post office is located. Depending on how you plan to communicate between post offices, you may need one or more modems. In the recommended and most common configuration, the cc:Mail Router runs on a dedicated OS/2 or DOS workstation on the same LAN as the file server where the cc:Mail post office is stored. Alternatively, the cc:Mail Router can be run concurrently on a machine acting as a file server, such as Artisoft Lantastic, or IBM Lan Server. However, a non-dedicated configuration cannot be recommended for more than a handful of users.

Multiple cc:Mail Routers can originate calls from a single post office using multiple Call Lists or sublists for parallel cc:Mail Routers. If you use the DOS version, you will need one workstation per cc:Mail Router. You can minimize the number of physical machines needed for cc:Mail Routers by using the OS/2 version. The cc:Mail Multisession Router for OS/2 supports several sessions and supports serial communications on up to 8 COM ports.

The machine you use as your cc:Mail Router should be a 386 or faster, especially if the cc:Mail Router will be processing directory updates. A local hard disk is recommended but its size is unimportant and more than 4MB or RAM would be wasted. Ideally, the cc:Mail Router would not be slower than the average user’s workstation because slower machines tend to slow down access to the post office files.

Installing the cc:Mail Router Software

The cc:Mail Router software can be installed on the file server or on the hard disk of the cc:Mail Router workstation. The OS/2 version of the cc:Mail Router program is named

ROUTER2.EXE and the DOS version is named ROUTER.EXE.

Although the program file can be located either on the cc:Mail Router workstation’s local hard drive, if any, or on a file server, I recommend installing the files on a local hard disk if possible. This allows the cc:Mail Router to recover somewhat more gracefully if a network connection to a file server is lost. If the program file is on the local hard disk and the network file server where the post office was stored went down for some reason, the cc:Mail Router should display the error similar to "Post Office cannot be accessed." If the executables were on the same server when it went down, you could see unexpected errors depending on what the cc:Mail Router was doing when the connection was lost.

Starting the cc:Mail Router

Once you’ve set up the Call List, post office addresses and the cc:Mail Router workstation, you’re almost ready to fire up the cc:Mail Router. However, there are a couple of details left.

Verifying Network Access

The cc:Mail Router workstation must be logged into the file server where the post office files are stored it and must have a drive letter that is associated with the file server volume where the originating post office is located. Depending on your network operating system, you may use "map," "net use," "mount" or other commands to associate a drive letter with the location of the post office on the file server.

Note: There are a couple of ways to do dynamic drive assignments. For now, we’ll keep things simple and assume the drive mappings are permanently set up on the cc:Mail Router workstation.

The login or user ID used for the cc:Mail Router workstation must have privileges or access rights that allow it to read and write to the post office files. In the UNIX operating system, for example, these privileges are read and write. In the Novell NetWare network operating system, they are read, write, filescan and create. Similar privileges or rights must be assigned to the the login or user ID used for the cc:Mail Router workstation for each post office that the cc:Mail Router will access. In other words, the cc:Mail Router must have equivalent privileges or access rights for the directory containing the originating post office and for the directories containing post offices for which the file I/O message transport will be used.

Rights to the directory where the originating post office is located and to the directories of any other post offices which the cc:Mail Router will connect to via a Type 1 connection.

Running the cc:Mail Router

Change to the directory where the ROUTER.EXE or ROUTER2.EXE and associated files are located, and type the cc:Mail Router command as shown below. The command line differs depending on whether or not you’re using a modem.

Using Only File I/O Connections

If all of your post offices are on the same LAN the cc:Mail Router can transfer messages directly between the post offices using the file I/O message transport. In this case, the command line would be similar to:

ROUTER Q:\DATA MODEM/NONE

The post office path on the command line must be the location of the originating post office. The parameter "MODEM/NONE" indicates that the asynchronous message transport will not be used. If you had post offices with drive and path addresses in your call list, and you ran the cc:Mail Router command as shown above, your cc:Mail Router display would look something like Figure 5.12.

If the call list were empty, or none of the conditions that would trigger a connection in the call list were true, the cc:Mail Router displays a message indicating the next time it is scheduled to make a connection.

Using the Asynchronous Message Transport with a Modem

For modems, the cc:Mail Router has default settings dating back to the early days of cc:Mail. The defaults are for a 2400 Baud Hayes or compatible modem connected to the COM1 serial port of the cc:Mail Router PC. Alternatively the COM port, Baud rate and modem configuration can be specified on the command line (see below). If a Hayes or compatible modem is attached to the COM1 serial port you can start the DOS version of the cc:Mail Router as follows:

ROUTER Q:\DATA

The path on the command line should of course specify the location of the originating post office. The command line parameter "DIAGNOSTICS" is optional, but always recommended for troubleshooting, especially when you first set up a cc:Mail Router. The diagnostics option causes the characters sent and received from the COM port to be displayed (see Figure 5.13).

Alternatively the COM port, baud rate and modem configuration can be specified on the command line. A more typical DOS command line for use with a high speed modem supporting the V.42 standard might look like this:

ROUTER COM2 57600 Q:\DATA MODEM/AT&FS0=0Q0V1M0X4&C1&D2*M0*E9*F3 LIST/4 DIAGNOSTICS

In this case the speed of your modem is 57,600 Baud. The MODEM/ parameter allows a string of modem commands and settings to configure the modem for use with cc:Mail Router. The modem commands should immediately follow the MODEM/ parameter with no spaces separating them. Specific modem commands are different for different modems. A list of modem initialization strings is available from Lotus and new modems are constantly tested. All combinations of modems, however, are not always tested so as a rule of thumb, try to use name brand modems and use the same model for all your cc:Mail Routers.

Express Calling

The cc:Mail Router has an feature that allows you to immediately call a specific post office by specifying the post office name on the command line. This is called express calling. Express calls can be used to override the schedule in the Call List or to call a post office which is not in the Call List. To initiate an express call, insert the name of the post office you want to call in the cc:Mail Router command line as shown below for a post office called ENCORE-SALES using the file I/O message transport.

ROUTER ENCORE-SALES M:\CCDATA

Address Override

The post office you are making an express call to must of course have a valid cc:Mail Address. If you want to temporarily override the address listed for the post office with a different one, you can use the "ADDR/" command line parameter as shown below:

ROUTER ENCORE-SALES COM2 9600 MODEM/AT&F Q:\DATA ADDR/767-7777

In the above example, the cc:Mail Router would make an express call to another post office called ENCORE-SALES using the file I/O transport. The ENCORE-SALES post office is located in the Q:\DATA directory. If you were doing a type 1 express call to a post office on the same local area network, the command line might look like this:

ROUTER ENCORE-MKTG M:\CCDATA ADDR/S:\CCMKTG

In this example, the ENCORE-MKTG post office which is being called is located on the S: drive in the \CCMKTG directory. Once the express call has been completed, the cc:Mail Router will automatically begin placing calls from the call list, if a call list exists.

Note: The command line for cc:Mail Router express calling assumes that you want to establish a send/receive connection and that you do not want to limit the size of messages which can be exchanged. See below for additional command line parameters which allow you to further customize the call.

cc:Mail Router Command Line Parameters

Just as you can set parameters in the call list, you can also set the same parameters for an express call using the cc:Mail Router command line. Here are the cc:Mail Router command line options you would use to restrict an express call:

These command line parameters may be used in combination to restrict the express call as shown in the example below:

ROUTER ENCORE-SALES M:\CCDATA S/O PRI/U MSGSIZE/1024

The command line shown above would cause the cc:Mail Router to place a Type 1 express call to the ENCORE-SALES post office, establish a send-only connection, sending only urgent messages which are under 1024k in size.Other useful cc:Mail Router command line parameters include the following:

 

Back to Contents

 

©1996, 1997 by Global System Services Corporation (GSS) Portions of this material are ©1995 by Ron Herardian

DISSEMINATION, DISTRIBUTION OR COPY OF THIS INFORMATION IS STRICTLY PROHIBITED. NO PART OF THE INFORMATION CONTAINED HEREIN MAY BE REPRODUCED IN ANY FORM BY ANY ELECTRONIC OR MECHANICAL MEANS, INCLUDING PHOTOCOPYING, RECORDING, OR INFORMATION STORAGE AND RETRIEVAL, WITHOUT THE WRITTEN PERMISSION OF GLOBAL SYSTEM SERVICES CORPORATION.

©1995-2005 by Global System Services Corporation (GSS). Portions of this material are copyright ©1995-1999 by Ron Herardian