Problem: Slow or failed resynchronization with network devices

General information

When NFM-P performance is slow, especially when performing network device resynchronizations, SNMP and IP performance along the in-band or out-of-band interfaces between the network device and the NFM-P server may be the problem.

Check the following:

  • configuration of the LAN switch port and the NFM-P station port match

  • configuration of the LAN switch port and the network device management ports match

  • mediation policy SNMP timeout and retry values are sufficient to allow the transfer of data between network devices and the NFM-P

Steps
 
1 

Ensure that port configurations are compatible for the NFM-P server, the network device management ports, and the LAN switch. This is normally done by configuring auto-negotiation between the platforms, but your network may require more specific configuration.

2 

Check whether all data is being transferred between the network device in-band management port and the NFM-P server.

  1. Open a Telnet or SSH session to the device from the NFM-P.

  2. Check statistics on the in-band management port of the device:

    # monitor port 1/2/3

    Check the output for the following.

    • errors that may indicate a communication problem with the a LAN switch.

    • Over each time interval, is the number of input and output packets constant? This may indicate intermittent traffic.

    • Are there more input packets or octets being transferred than output packets or octets? This may indicate a unidirectional traffic problem.

    The types of error messages displayed determine the action to take.

    • For failure errors, consider increasing the SNMP timeout value

    • For collision errors, consider increasing the SNMP retry value

  3. Check the mediation policy for the device using the NFM-P client GUI. Check the SNMP timeout and retry value for the mediation policy.

    If the output of step 2 indicates failures, consider increasing the default SNMP timeout value and perform step 2 again.

    When the output of step 2 indicates frequent collisions, consider increasing the default SNMP number of retries value, then retest to see if resynchronizations are more reliable. Increasing the number of retries increases the likelihood that an SNMP packet is not dropped due to collisions.

    You can check SNMP timeout and retry values from the Administration→Mediation menu. Click on the Mediation Security tab.

    CAUTION:

    When LAN performance is poor, increasing timeout values may mask an underlying problem. Increasing the SNMP timeout value in an environment where collisions are frequent reduces performance. Timeout values should be based on typical network response times

    Check LAN communication issues, as specified in Chapter 3, Troubleshooting the NSP platform. If problems persist, collect the logs as specified in Before you call support and contact your technical support representative.

End of steps