Common SLIP problems and error messages

Message: 0821-296 Cannot set line discipline for /dev/tty# to slip.ioctl(TXSETLD). A system call received a parameter that is not valid.

Possible Causes: This type of error normally occurs when starting the slattach process and is attributable to incorrect configuration of SLIP. The problem is most likely caused by a mismatch between the tty device number and the sl interface number. This also explains why the system reported that ifconfig had not been run before slattach.

This problem may also occur when slattach processes are dropped or killed incorrectly or when the user attempts to move a SLIP connection to another tty port and forgets to reconfigure the sl# interface to match the tty. Check for running slattach processes that may still be running (for example, ps -ef | grep slat).

Action: The tty device for SLIP is /dev/tty24 and user has created an sl0 interface. This is incorrect. The user should create an sl24 interface which matches the tty number (tty24 and sl24). If the problem continues, the user should bring down the sl interface (see "Bringing Down an SLIP Interface") and reconfigure the connection using the following commands:

Message:

 

network is not currently available

 

route to remote host not available

Possible Cause: These errors occur most often when a user attempts to ping a host over the SLIP link and the link has been improperly established. The most likely problem is that one or both tty ports associated with the sl# interface are in an enabled state. It is also possible that there is an address or route conflict between the host systems.

Actions:

• Remove the sl# interface using the smit rminet fast path This must be done on both the local and remote SLIP hosts.
• Do the following for each SLIP host:
  1. Enter pdisable | grep tty#.
  2. If the tty device is NOT listed in the output of the previous command, the tty is not disabled. Disable the tty either through SMIT or the command line. With tty ports disabled, use SMIT to recreate the SLIP interfaces on both systems. If problem persists, verify network addresses and routes (if any). Use the netstat -ir command to quickly view address, routing, and interface information.

Problem: When the remote site dials in to the local host, the modem on the local host connects but does not complete the login process.

Possible Causes: If the two modems connect and begin to handshake or exchange connection information but then disconnect, the problem may be due to modem result codes. This problem can also be caused by an improper slattach dial string. If the two modems ring but never begin the handshake process, the problem may be that the modem is not set for auto-answer.

Actions:

1. Test the modem connection first with the cu command. The modem on the remote host should allow the user to login to the system. There should not be any garbage on the screen during the login attempt; if so, it may indicate a noisy phone line which may be part of the problem. During the login, multiple login heralds should not scroll across the screen. If they are present, this could again indicate a problem phone line or incorrect modem settings.
2. Check the modem configurations and try turning off the ARQ codes if they are currently on. In most Hayes-compatible modems this is the &A0 setting. Disabling ARQ result codes does not affect error-controlled connections nor does it keep the modem from returning standard CONNECT messages (if result codes are enabled) as needed for the slattach dial string.

Problem: The user is unable to ping across a modem SLIP connection. The ping command may hang or return error messages.

Possible Causes:

1. The modems and/or tty ports may be configured to use XON/XOFF flow control.
2. The slattach process may have been terminated on the remote host or the modem connection dropped.
3. The addresses assigned to the SLIP hosts may be incorrect.

Actions:

1. Examine both the local and remote modem configurations. They should be set to use RTS/CTS (hardware) flow control or no flow control at all. The user should attempt to ping from each system. Ping systemA to systemB.
2. Verify that the slattach process is still running on both local and remote systems. Use the command: ps -ef |grep slat. Verify that the sl# interface is in a running state. Use the command: ifconfig sl#.
3. Verify that there is not a conflict between the SLIP addresses and those associated with other network interface (if any). Use the command: netstat -ir. If the address or address class is in question, reconfigure SLIP using a simpler address scheme such as 1.1.1.1 for the local host and 1.1.1.2 for the remote host.