External Policy Server Connection Status Reporting

When the Oracle Communications Session Border Controller (SBC) loses connection to the external policy server or an attempt is unsuccessful, the SBC generates an SNMP trap. In addition to this trap, the SBC generates major or minor, non-health effecting alarms. When a connection is restored, the SBC sends a clear trap and clears the associated alarm.

Note:

The SBC does not raise SNMP traps or connection down alarms when you cause the event with a configuration change, such as disabling a server element.

Enterprise Traps

The following table identifies the proprietary traps that the ap-diameter supports.

Trap Name Description
apSysMgmtExtPolicyServerConnDownTrap

1.3.6.1.4.1.9148.3.2.6.0.74

Generated when the SBC is unable to connect to an external policy server
apSysMgmtExtPolicyServerConnEstTrap

1.3.6.1.4.1.9148.3.2.6.0.75

Generated when the SBC is able to re-establish a connection with an external policy server

These traps contain the following information:

  • Name of the policy server as it is configured on the SBC
  • FQDN of the policy server cluster (This is left empty if the policy server is entered as an IP address)
  • IP Address and port of the active policy server in the form <IP-Address>:<Port>
  • Realm to which the policy server belongs
  • Operation type (RACF/CLF) of the policy server

A connection is deemed successful when the Diameter CER/CEA transaction completes. In the case of FQDN, a connection-established trap is sent when each policy server in the top-level cluster completes its CER/CEA action.

During a reboot, a connection-down trap is sent initially. An attempt is made to connect; if the attempt is successful, a connection-established trap is sent. Upon every subsequent failed attempt to establish a connection, the SBC makes another attempt without an additional trap.

During a switchover, the newly activated SBC behaves in the same way as a reboot process. Initially, a connection down trap is sent. An attempt is made to connect; if the attempt is successful, a connection-established trap is sent. After the switchover, and upon every subsequent failed attempt to establish a connection, the SBC makes another attempt without an additional trap.

In the event that a transport protocol connection is established but a Diameter connection is unsuccessful, the existing connection is closed and a connection down trap is sent.

Alarms

Alarms that correspond with the traps above notify the user, as follows:

  • When the SBC sends a connection down trap for a connection configured within a policy server cluster, it generates a minor alarm.
  • When the SBC sends a connection down trap for a connection configured with an IP address and port, it generates a minor alarm.
  • When the diameter connection is re-established, the SBC clears the alarm.
  • When all connections associated with a single ext-policy-server are down, the SBC generates a major alarm.

The table below shows alarm information associated with a failed connection to an ext-policy-server.

Name/ID Severity/Health Degradation Cause(s) Log Message Traps Generated
APP_ALARM_EPS_RACF_CONN_FAIL MINOR/0 Connection to External Policy Server is lost. External Policy Server (RACF) connection lost!!! apSysMgmtExtPolicyServerConnDownTrap

You see this alarm using the display-alarms command.

ORACLE# display-alarms
1 alarms to show
ID      Task    Severity        First Occurred          Last Occurred
327689  97      4       2018-03-20 01:29:46     2018-03-20 01:29:46
Count   Description
1       External Policy Server (RACF) connection lost!!!

When multiple server connections are lost, the alarm appears as follows:

ORACLE# display-alarms
1 alarms to show
ID      Task    Severity        First Occurred          Last Occurred
327689  97      4       2018-03-20 01:30:56     2018-03-20 01:30:56
Count   Description
1       2 External Policy Server (RACF) connections lost!!!

The alarm definition is as follows:

  • Definition: APP_ALARM_LOST_RACF_CONN - ID: 327689
  • Text: External Policy Server (RACF) connection lost!!!
  • Definition: APP_ALARM_LOST_CLF_CONN - ID: 327690
  • Text: External Policy Server (CLF) connection lost!!!

These alarms remain cleared as long as all diameter connections are established. The alarm text provides counts of the number of connections that are down, which the system updates as connections go down or are re-established.

The alarm text differs, based on the number of servers down, as follows:

  • External Policy Server (RACF) connection lost!!!" (one connection down)
  • 2 External Policy Servers (RACF) connection lost!!!" (more than one connection down)