PRIMECLUSTER Global Link Services Configuration and AdministrationGuide 4.3Redundant Line Control Function

3.11.2 Self-checking feature

3.11.2.1 How to set up the self-checking function

The self-checking function can be enabled as follows.

1. Enable the self-checking function

   # /opt/FJSVhanet/usr/sbin/hanetparam -e yes

2. Check the changed parameters.

   # /opt/FJSVhanet/usr/sbin/hanetparam print
   [Fast switching]
      Line monitor interval(w)           :5
      Line monitor message output (m)    :0
      Cluster failover (l)               :5
      Cluster failover in unnormality (c):OFF
      Line status message output (s)     :OFF
   
   [NIC switching]
      Standby patrol interval(p)         :15
      Standby patrol message output(o)   :3
   
   [Virtual NIC]
      LinkDown detection time (q)        :0
      LinkUp detection time (r)          :1
      Link monitor starting delay (g)    :5
   
    [Common Setting]
      Hostname resolution by file(h)     :NO
      Self-checking function(e)          :YES

3. Reboot the system. After reboot, the self-checking function will be enabled.

   
   

The self-checking function can be disabled as follows.

1. Disable the status monitoring function

   # /opt/FJSVhanet/usr/sbin/hanetparam -e no

2. Reboot the system. After reboot, the self-checking function will be disabled.

3.11.2.2 Error detection of the self-checking function

The following describes how the monitoring is performed with the self-checking function. The virtual driver and control daemon are monitored periodically.

Figure 3.25 Error detection of the self-checking function

The monitoring targets are as follows. A system wide hang or error status cannot be detected.

Monitoring target	Error type	Error detection method
Driver	Hung-up	No response from the virtual driver for 15 seconds
	I/O Error	Information is not received from the driver five times in a row
Daemon	Hung-up	There is no response from the control daemon for 300 seconds
	I/O error	Information is not received from the control daemon five times in a row
	Stopped process detection	There is no control daemon process

If an error has been detected, a message similar to the following will be output to the system log. After this, the monitoring function will be stopped. To restart monitoring, reboot the system after collecting troubleshooting information.

ERROR: 97427: sha driver error has been detected. code=xxx

xxx: error type (hung-up or I/O error)

ERROR: 97627: hanetctld error has been detected. code=xxx

xxx: error type (hung-up, I/O error, or stopped process)

Note that placing a script in the following location allows the script to be executed when an error is detected. For more details, see "3.12.2 Setting user command execution function".

/etc/opt/FJSVhanet/script/system/monitor