This alarm is generated when the CPU usage of the CG9812 server is greater than the threshold (95% by default).

This alarm is cleared when the CPU usage of the CG9812 server is smaller than the threshold (50% by default).

Alarm ID	Alarm Severity	Auto Clear
0027	Critical	Yes

Parameters

Name	Meaning
Server No.	It specifies the number of the server of which the CPU is overloaded. Its value is 0.

Impact on the System

The performance of the CG9812 is degraded in processing CDRs. On the GSN/UGW server the alarm about the link interruption between the GSN/UGW and the CG9812 may be generated. When the traffic is heavy, the disks of the GSN/UGW may be full and the GSN/UGW do not send the CDRs to the CG9812 in time.

The CG9812 exits abnormally.

Possible Causes

• Other processes consume a large number of CPU resources.
• The system is attacked by viruses.
• The system is attacked by network storms.

Procedure

Analyse the alarm information

1. Determine the CG9812 server on which the CUP usage is high according to the IP address of the CG9812 generated the alarm.

Check related alarms

1. Press Ctrl+Alt+Del, in the displayed dialog box click Task Manager, click Process tab check which process consumes a larger number of CPU resources.
   http://localhost:7904/pages/31184372/01%20%282010-03-10%29/31184372/01%20%282010-03-10%29/resources/public_sys-resources/icon-note.gif NOTE:

   The processes of the CG9812 includes the pwd.exe, cls_proc.exe, knl_proc.exe, ap_proc.exe, om_proc.exe and cfg_proc.exe.
   • Yes => 3
   • No => 5
2. Save the screen snapshot of the task manager and record the CG9812 processes that consume too many CPU resources. Then restart the CG9812 and observe whether this alarm reoccurs.

   Restart the CG9812: run the KillCluster.exe program in the installation directory (by default in the directory C:\iGWB)of the CG9812 and then click Yes in the dialog box displayed. After two to five minutes, the CG9812 is stoped. Wait two to five minutes, and check whether the CG9812 is restarted again. If the CG9812 is restarted normally, you can see http://localhost:7904/pages/31184372/01%20%282010-03-10%29/31184372/01%20%282010-03-10%29/resources/help/image/ALM_0001_fig_01.png in the task bar of the window desktop. If the CG9812 is restarted abnormally, you can see http://localhost:7904/pages/31184372/01%20%282010-03-10%29/31184372/01%20%282010-03-10%29/resources/help/image/ALM_0001_fig_02.png in the task bar of the window desktop.

   If the CG9812 is restarted again, choose Start > All Programs > Administrative Tools > Service and in the Service window check whether the running status of the Process Watch Dog is Started. If the running status of the service Process Watch Dog is not Started, set the startup type of the service Process Watch Dog to Automatic and then start the service. If the running status of the service Process Watch Dog is Started, start the service. If the CG9812 is not restarted, run the cls_proc.exe program in the installation directory.

   • Yes => 4
   • No => End the procedure for handling this alarm

Yes you are right the alarm is cleared when the CPU usage of the CG9812 server is smaller than the threshold (50% by default).

Very useful documents................

Thanks a lot.........................