FDDIXPress provides station management (SMT) commands (listed in Table 3-1) for managing and monitoring both the FDDI ring and the FDDI connections at each station. You execute these commands from a shell window.

FDDIXPress includes a graphical tool for maintaining FDDI rings: FDDIVisualyzer. This application displays a color-coded graphical representation of the FDDI ring, thus allowing you to see the ring status and view all the station statuses. Unlike the SMT commands that work best for monitoring a single station, FDDIVisualyzer provides complete ring status at a single glance.

You can use the SMT commands to verify that a station is reachable on the ring, to list all the stations on the ring, to monitor the status of a particular station's FDDI connection, and so forth. Table 3-1 lists the SMT commands.

Something the SMT commands do not do for you is maintain a list of all the stations that are supposed to be on the ring. In general, SMT information is always current, so only the currently operating stations can be listed and contacted. However, at times, it is convenient to have a list of all the stations that should be functioning on a particular ring. This is a task the network administrator for the FDDI ring should do.

The SMT commands are specialized versions of the IRIX networking commands. For example, a few of the SMT commands that are similar to the networking commands include smtstat (like netstat), smtping (like ping), and smtconfig (like ifconfig). Be sure to use the SMT commands for the FDDI connections; use the standard network commands for the Ethernet connections. See Appendix D for the man pages for the SMT commands.

To see if a station is on the FDDI ring:

% /usr/etc/smtping –I fddiinterface -c 2 host 

host is a name in the /etc/ethers file or a MAC address, and fddiinterface is the network interface (for example, ipg0, ipg1, and so on, or xpi0, xpi1, and so on).

This command sends a request for a response (analogous to an RSVP). The previous request is sent twice (–c 2) to a host (either a MAC address or a name defined in /etc/ethers) on the specified FDDI network connection. As the host receives each request, it responds by sending some data. If host is not responding for any reason, smtping will time out, and the screen display indicates the problem with the message ... 0 packets received, 100% packet loss.

When a station is on the ring and responds, you see a display similar to Figure 3-1.

Figure 3-1. smtping Display

For additional information about smtping and its options, see smtping(1M).

To list all the functioning stations on the FDDI ring, use the smtring command:

% /usr/etc/smtring 

smtring broadcasts to all the stations and concentrators on the ring, waits 60 seconds to collect all responses, and then lists the responding stations in logical order. The local station is at the top of the list (unless a specific host has been specified in the command line). Each line shows three items of information, described from left to right:

Figure 3-2 shows an example of responses. In this example, the smtring command has been invoked from a station named goofy whose MAC address is a:0:1b:4:0:7 and whose upstream neighbor is mickey. From this listing you can construct a ring, as shown in Figure 3-3.

Figure 3-2. smtring Display

In Figure 3-2, goofy's station ID is the default one created by the FDDIXPress SMT daemon from the MAC address.

Figure 3-3. Ring Created From smtring Display

You can use the –I option to smtring if you have more than one FDDI controller board installed (for example, xpi0 and xpi1) and want to examine the state of the second board's ring. To do this, use:

% /usr/etc/smtring -I interfacename 

With the smtinfo command, you can see SMT information for a specified remote host, including station configuration and operation information and supported SMT versions.

% /usr/etc/smtinfo host 

goofy uses SMT version 2 (supported versions: 1-2)

Display a Station's Configuration Information

To see station configuration information for a remote host (for example, 0a:00:1b:04:00:07 or goofy), use:

% /usr/etc/smtinfo -c host

host is a name in the /etc/ethers file or a MAC address.

An example of smtinfo -c output is shown in Figure 3-4 and the information is explained in Table 3-2.

Figure 3-4. smtinfo -c Display

Table 3-2. Information Displayed by smtinfo -c

Item	Description
Msg Timestamp	This information is not useable.
Station Dscrpt	Description of the station
Supported Vers	SMT versions supported and currently used
Station States	Station's states. dupa=0 indicates that there are no duplicate MAC addresses. ROOTSTA indicates that the station is connected to a dual ring, not a concentrator.
MAC3 Nbrs	Upstream (una) and downstream (dna) neighbor addresses.
Path Descriptors	Descriptions of the port or ports and the interface's MAC address in canonical order.

If the host has an additional FDDI board installed and you want to see configuration information about that interface (for example, ipg1), use:

% /usr/etc/smtinfo -c -I interfacename host

For more information about smtinfo, see smtinfo(1).

The network administrator for an FDDI ring needs to monitor the ring and recognize faults when they occur. One type of fault, for example, occurs when a station on the network goes down. The SMT modules within the FDDI stations isolate the fault and wrap to the secondary ring. If an optical bypass switch is installed on the dysfunctional station, the switch causes the ring to bypass the station and no wraps occur. The network continues to function. Figure 3-5 shows an example of wrapped ring when one station goes down.

Figure 3-5. Fault Isolation and Ring Wrap

As you can see in Figure 3-5, when one station experiences an error, only that station is out of the loop. It is also possible for one port on one station to malfunction (as illustrated by fault 1 in Figure 3-6), in which case the ring wraps, but no station is out of the loop. Traffic on the network continues as usual. Notice that when a station fails (both ports are off the ring), the wraps do not occur on the faulty port, but on its two neighbor ports.

If two or more stations go down, the network may become fragmented into a number of smaller loops, and users may not be able to access stations that have become isolated onto a different loop. It is important to locate the faults and make the necessary repairs as quickly as possible. Figure 3-6 shows a fragmented network with two stations down.

Figure 3-6. Fragmented Ring

How do you know there is a fault on the ring? One way is to use smtping at each station to verify that each one responds. Another way is to use smtring to create a complete list, then verify that all known stations appear on the list. The easiest method is to use the FDDIVisualyzer product, which graphically indicates faults on the ring. In all of these instances, an up-to-date, complete list of all stations that are physically attached to the ring is vital. The network administrator should maintain this list. In many cases, once a station fails, it is not possible to communicate with it remotely over the ring. Its failure is deduced from its absence or lack of response. To debug the faulty station, log on to that station through a second network connection or at the station.

You can log on to any station and use smtstat to display status information about that station's FDDI interfaces. This command does not access a station remotely; rather, it displays information about the local host, even when the host is not functioning on the ring. A step-by-step verification process is described in “Verifying the FDDI Connection”.

You can display and monitor the status of a station's FDDI connection by logging on to the station and entering the following command:

% /usr/etc/smtstat 

The smtstat report is shown in Figure 3-7 and a listing of its columns follows. Table 3-3 explains the entries for the columns in more detail.

Figure 3-7. smtstat Display

In the DAS display, the first line contains the following information:

The second line contains column headings while subsequent lines display the information:

By monitoring SMT status, you can see the connection state, transmitting and receiving line states, and the flags indicating the ring state. In the smtstat display shown in Figure 3-7, the primary ring is operating (RNGOP), its transmit line is working (THRU), its receive line is idle (ILS), and the secondary ring is functional (LS, RC, JOIN) but not operating. If the ring were wrapped at this station, you would see a WRAP indication.

Display SMT Information (MIB)

To monitor continually the FDDI activity of a configured interface on the local system, use:

% /usr/etc/smtstat –s -I fddiinterface

fddiinterface is the interface (for example, ipg0, ipg1, xpi0, xpi3).

The output of this command produces the following six smtstat reports.

• MAC status

• port status

• ring management status

• configuration information

• neighbor information

• miscellaneous SMT information

This output requires a shell window or a terminal at least 80 columns wide by 24 lines high. When the window is too small, the display is not usable. Make your display area at least this size before invoking the command.

The reports are viewed one at a time. The top of each report displays the report name, date, and time. At the bottom of each report is a menu line listing all the available reports, as shown in Figure 3-8.

Figure 3-8. smtstat -s Display: General Report Format

The following user interface allows you to manipulate the reports:

• To display a different report, type the report's number. The numbers are displayed in the menu line on the bottom of the screen. For example, to view the SMT report, type 6.

• To quit, type q or Ctrl+c.

• To refresh the screen, type Ctrl+l.

• To change the time interval for the report statistics display, press one of the these keys:

  r		Each second, shows totals accumulated since last reboot.
  d		Resets fields to zero every second, and each second shows totals accumulated within that last second.
  z		Resets fields to zero when z is pressed, and every second displays the accumulating totals.

For a description of all the fields in each report, see Appendix B, “smtstat Reports.” For a complete definition of each field, refer to the ANSI standard entitled ANSI FDDI Station Management (SMT). For information about command usage, refer to the smtstat(1M) man page.

As network administrator, you may need to display or temporarily alter a station's network interface parameters. The command smtconfig allows you to do this.

Display the Configuration

To display the current configuration of the FDDI interface, follow these steps:

1. Use smtstat to determine the name of the FDDI network interface(s) on your station. The FDDI interface name is listed under the column heading Name. Examples: ipg0, xpi0.

2. Use smtconfig with the name of the interface discovered in the first step:

   % /usr/etc/smtconfig interfacename

   
   

Sample output from smtconfig is shown in Figure 3-9. A description of the fields follows, starting with the first item on the left of the top line and moving right.

Figure 3-9. smtconfig Display

• First line:

  • The interface name.

  • Flags that are enabled, with the following meanings:

    
    

• Second line:

  • The address family configured for this interface: inet is the default.

  • The address assigned to this network interface: The default is the IP address associated with the hostname in the /etc/hosts file.

  • The network mask (netmask) used for the IP subnetwork: The default is a value that, when applied, implements no mask.

  • The address used for broadcasting to all stations on the ring: The default is an IP address where the host portion is all ones.

Change the Configuration

As superuser, you can enable and disable the FDDI interface or (re)configure the FDDI interface parameters manually using the smtconfig(1M) command. Most of the time, it is not necessary to set interface parameters manually using smtconfig because FDDI is configured automatically with default settings (as described in Table 2-3) when the system is started or rebooted.

---

Note: Changes made with smtconfig are not saved; at reboot they return to their default settings (or the /etc/config/ifconfig-#.options and /etc/config/netif.options files). To change the configuration so the changes survive rebooting, follow the instructions in Chapter 2 or in IRIX Admin: Networking and Mail.

---

To change the configuration of a network interface, perform the following steps:

1. Become superuser:

   % /sbin/su Password: thepassword #

   
   

2. Use smtstat to determine the name of the FDDI network interface(s) on your station. The FDDI interface name is listed under the column heading Name. Examples of FDDI interfaces include ipg0, ipg1, xpi0, xpi3.

3. Use smtconfig with one of the following formats.

   Superuser can use the smtconfig command to change the FDDI operational parameters listed below, in much the same manner as ifconfig. For information about the creation and purpose of these parameters, refer to IRIX Admin: Networking and Mail.

   • Enable or disable the interface:

     # /usr/etc/smtconfig interfacename up # /usr/etc/smtconfig interfacename down

     
     

   • Set a network mask value for an IP subnetwork:

     # /usr/etc/smtconfig interfacename netmask value_or_name

     
     

   • Change the address used for broadcasting:

     # /usr/etc/smtconfig interfacename broadcast address

     
     

   • Enable or disable driver debugging messages:

     # /usr/etc/smtconfig interfacename debug # /usr/etc/smtconfig interfacename -debug

     
     

   • Enable or disable Address Resolution Protocol. ARP is enabled when there is no flag about ARP:

     # /usr/etc/smtconfig interfacename arp # /usr/etc/smtconfig interfacename -arp

     
     

   • Set the routing metric used by the route daemon (routed):

     # /usr/etc/smtconfig interfacename metric digit

     
     

   • Change the IP address for this interface:

     # /usr/etc/smtconfig interfacename IPaddress

     
     

   • Select a different interface to be the primary network interface for this station:

     # /usr/etc/smtconfig interfacename primary

     
     

Examples

This section describes some scenarios in which the different smtconfig command options can be used.

• Before performing computer maintenance work that will require rebooting a station, disable the FDDI network interface, xpi0, with the command:

  # /usr/etc/smtconfig xpi0 down

  
  

• On a router with an FDDI interface named ipg0, you want to decrease the number of packets going through the station enroute to other networks. One way to do this is by advertising the route as less favorable. Use this command:

  # /usr/etc/smtconfig ipg0 metric 5

  
  

For information about creating netmasks, route metrics, broadcast addresses, and IP addresses, or about altering network interface configurations, see “Changing Settings for the Operational Parameters” in Chapter 2, the smtconfig(1M) man page, or IRIX Admin: Networking and Mail.

The smtmaint command allows you, as superuser, to set the Physical Connection Management (PCM) line state for each PHY on a station's FDDI board. An SAS has one PHY (PHY0); a DAS has two PHYs (in most cases, PHY0 for port B and PHY1 for port A).

Use smtmaint judiciously and only if you are an FDDI expert. Setting a line state can make the ring nonfunctional.

You can set these six line states:

To change a line state, follow these steps:

When an FDDI station is removed from the ring, the SMT modules reconfigure the ring. If the removal is done in a way that makes it impossible for the optical signal to complete its loop around the ring, the ring wraps. Generally, it is best to avoid wrapped rings for long periods. The procedures described in this section describe how to avoid or minimize a wrapped ring when removing a station from the ring.

The situations listed below constitute taking an FDDI station off the ring; any of these has the potential to cause the FDDI ring to wrap (sometimes for a very short period of time).

The procedure for safely removing a station from the ring differs, depending on three factors:

Remove a DAS Without an Optical Bypass Switch

Removing a DAS that does not have an optical bypass switch always disrupts the ring; removal of a station causes the ring to wrap in configurations where the secondary ring is a backup. There is no way to avoid this; hence, your primary concern is the length of time the ring will be disrupted. After you remove the station, you need to do one of the following as soon as possible to reestablish the integrity of the ring:

• Insert the same or another station at the location.

• Insert an optical bypass switch at the location.

• Patch the ring (for example, by inserting a barrel connector or section of fiber optic cable at the location).

Temporarily Removing a DAS

To temporarily remove a DAS without an optical bypass switch from its ring, follow the instructions below.

---

Note: This procedure causes the ring to wrap.

---

1. Become superuser and shut down the system. For example:

   % /sbin/su Password: thepassword # /etc/shutdown

   
   

2. Power off the station by pressing the station's power switch, and disconnect the station's FDDI cable(s).

3. As soon as possible, do something to reestablish the ring's integrity. A list of possible actions is provided at the beginning of this section.

4. You can now perform system maintenance or physically move the station. Since the removed station's identity as an FDDI station remains in the system files, the station should be returned to the ring at some point. When reinserted into the ring, the station does not have to be reconnected at the same location.

Permanently Removing a DAS

To permanently remove a DAS station without an optical bypass switch from the ring, follow the instructions below.

---

Note: The ring will wrap as soon as you disable the FDDI interface.

---

1. Become superuser:

   % /sbin/su Password: thepassword #

   
   

2. Edit the configuration files to remove or comment out the lines referring to this station. One or more of the following files may need to be edited: /etc/hosts, /etc/ethers, /etc/config/netif.options.

3. Disable the FDDI interface:

   # /usr/etc/smtconfig interfacename down

   
   

4. Shut down the system using the shutdown or halt command.

5. Power off the station by pressing its power switch.

6. Disconnect the station's FDDI cable(s).

7. As soon as possible, do something to reestablish the ring's integrity. A list of possible actions is provided in the beginning of this section.

8. At this point, you can move the workstation or server. If you want to reinsert this station into this ring, you will need to reconfigure the station.

For more information, see the smtconfig(1M), halt(1M), and shutdown(1M) man pages or ”Change the Configuration” in this chapter.

To permanently remove an FDDIXPress board and software from a station, follow the steps below: