Heartbeat Support for Sx Interface

Revision History


Note

Revision history details are not provided for features introduced before release 21.24.

Revision Details

Release

First introduced

Pre 21.24

Feature Description

In accordance with 3GPP standard, support has been added for node-level Heartbeat procedures between the Control Plane (CP) function and User Plane (UP) function over Sx Interface.

The Heartbeat procedure contains the following two messages:

  1. Heartbeat Request

  2. Heartbeat Response

Heartbeat Request

The CP function or the UP function sends a Heartbeat Request on a path to the peer node to find out if it is alive. The Heartbeat Request messages are sent for each peer with which a Packet Forwarding Control Protocol (PFCP) control association is established.

For each peer with which a PFCP control association is established, a CP function or UP function is prepared to receive a Heartbeat Request at any time, and replies with a Heartbeat Response.

Heartbeat Response

This message is sent as a response to a Heartbeat Request.

How It Works

CP function and UP function sends Heartbeat messages after configurable time duration. If the peer does not respond, the message is retried for configured number of times with the retry-interval and then the configured action is taken for the calls associated with the corresponding peer.

Recovery Time Stamp Information Element (IE), which contains the start time of the node, is supported by both Heartbeat Request and Heartbeat Response. Heartbeat Request contains its own Recovery Time Stamp value and sends it to the peer while Heartbeat Response contains the peers Recovery Time Stamp value.

Path Failure Detection

Path failure is detected in following conditions:

  1. Heartbeat failure: This condition occurs when the peer does not respond to the Heartbeat that is sent and also retires.

  2. Recovery Time stamp change in Heartbeat: This condition occurs when the Heartbeat Request or Heartbeat Response has a new larger value than the previously received value.

  3. Recovery Time stamp change in Sx Association message: This condition occurs when the Sx association message is received again from the peer with a new Recovery Time Stamp.

Path Failure Handling

When the Recovery Time Stamp value received is more than the previously received value, then the peer restart is detected. If the Recovery Time Stamp value is lower than the previously received value then the value is ignored and peer restart is not detected.

When a peer restart is detected, an SNMP Trap is generated to indicate the path failure for the peer. Also, based on the path failure configuration (refer Configuring Heartbeat for Sx Interface), all the calls connected to that peer can be cleared.

Configuring Heartbeat for Sx Interface

This section provides information about the CLI commands available in support of this feature.

Enabling Heartbeat for Sx Interface

Use the following commands under Sx Service Configuration mode to enable Heartbeat parameters for Sx Interface.

configure 
   context context_name 
      sx-service service_name 
         [ default ] sx-protocol heartbeat { interval seconds | max-retransmissions number | path-failure detection-policy { control-recovery-timestamp-change | heartbeat-retry-failure | heartbeat-recovery-timestamp-change } | retransmission-timeout seconds } 
         no sx-protocol heartbeat { interval | path-failure detection-policy { control-recovery-timestamp-change | heartbeat-retry-failure | heartbeat-recovery-timestamp-change } 
         end

Notes:

  • default : Sets/restores default value assigned for specified parameter.

  • no : Disables the followed option.

  • heartbeat : Configures Sx heartbeat parameters.

  • interval seconds : Configures heartbeat interval (in seconds) for SX Service. seconds must be an integer in the range of 1 to 3600.


    Important
    In releases prior to 24th July CUPS Lab Drop, seconds was set as 60 seconds by default.

  • max-retransmissions number : Configures maximum retries for SX heartbeat request. Must be followed by integer, ranging from 0 to 15. Default is 4.

  • retransmission-timeout seconds : Configures the heartbeat retransmission timeout for SX Service, in seconds, ranging from 1 to 20. Default is 5.

  • path-failure : Specifies policy to be used when path failure happens via heartbeat request timeout.

Configuring Detection Policy for Path Failure

Use the following commands under Sx Service Configuration mode to specify detection policy to be used for path failure. 

configure 
   context context_name 
      sx-service service_name 
         [ default | no ] sx-protocol heartbeat path-failure detection-policy { control-recovery-time-stamp-change | heartbeat-retry-failure | heartbeat-recovery-
timestamp-change } 
         end

Notes:

  • default : Sets/restores default value assigned for specified parameter.

  • no : Disables the followed option.

  • detection-policy : Specifies the policy to be used. Default action is to do cleanup upon heartbeat request timeout.

  • control-recovery-time-stamp-change : Path failure is detected when the recovery timestamp in control request/response message changes.

  • heartbeat-retry-failure : Path failure is detected when the retries of heartbeat messages times out.

  • heartbeat-recovery-timestamp-change : Path failure is detected when the recovery timestamp in heartbeat request/response message changes.

Monitoring and Troubleshooting

This section provides information about CLI commands available for monitoring and troubleshooting the feature.

Show Command(s) and/or Outputs

This section provides information regarding show commands and/or their outputs in support of this feature.

show sx-service all

The output of this show command has been enhanced to include the following fields introduced in support of this feature.

  • SX Heartbeat

    • Interval

    • Retransmission Timeout

    • Max Retransmission

  • SX path failure detection policy

    • Heartbeat Timeout

    • Heartbeat Req/Rsp Recovery timestamp change

    • Control Msg Recovery timestamp counter change

show sx-service statistics all

The output of this show command has been enhanced to include the following fields introduced in support of this feature.

  • Heartbeat Request

    • Total TX

    • Total RX

    • Initial TX

    • Initial RX

    • Retrans TX

  • Heartbeat Response

    • Total TX

    • Total RX

Disconnect Reasons

The following disconnect reason has been added in support of this feature:

  • sx-path-failure - When the Recovery timestamp changes or heartbeat failure is detected, based on the configuration, calls are cleared with this disconnect reason.

SNMP Traps

The following SNMP traps have been added in support of this feature:

  • SxPathFailure - This trap is generated when the peer path failure is detected.

  • SxPathFailureClear - This trap is generated when the path is restored for the peer.