P-GW Changes in Release 21 · • NEMO-LMAHeartbeat,page29 ... Release Change Reference, StarOS Release 21.2 10 P-GW Changes in Release 21.2 How CUTO Works. List of Attributes and

P-GW Changes in Release 21.2

This chapter identifies features and functionality added to, modified for, or deprecated from the P-GW inthe StarOS 21.2 software release.

The following identifies all of the P-GW enhancements included in this release:

Feature Changes - new or modified features or behavior changes. For details, refer to the P-GWAdministration Guide for this release.

Command Changes - changes to any of the CLI command syntax. For details, refer to the Command LineInterface Reference for this release.

Performance Indicator Changes - new, modified, and deprecated bulk statistics, disconnect reasons,counters and/or fields in new or modified schema and/or show command output. For detailed informationon the show commands, refer to the Statistics and Counters Reference for this release. For detailed informationon the counters and disconnect reasons, refer to the BulkstatStatistics_documentation.xls spreadsheet thatis included as part of the software companion package for this release.

This release includes enhancements that are applicable to multiple products. The following lists the variousmulti-product enhancements sections, some of which might include content applicable to your P-GW.

Important

• AAA Changes

• CF Changes

• ECS Changes

• GTPP Changes

• Lawful Intercept Changes

• NAT Changes

• SNMP MIB Changes

• System and Platform Changes

P-GW changes for 21.2 include:

• 3GPP Changes to the Gx Interface, page 2

• APN Configuration Errors Display, page 3

Release Change Reference, StarOS Release 21.2 1

• Auto Correction of VoLTE Sessions, page 5

• Cisco Ultra Traffic Optimization, page 8

• Civic Address Display Format Changed in Monitor Protocol and show subscribers pgw-only full allCommand Output, page 18

• CLI Configuration for APN String, page 19

• Error Message Provides Additional Information for TDF Ruledef Validation, page 21

• FAPA Disabled Due to Bandwidth Policy Usage, page 22

• Inclusion of APN AMBR in the Create Session Response, page 23

• Incorrect Cause Code in Create Session Response by P-GW, page 25

• Increase in ARP Support , page 26

• NEMO-LMA Heartbeat, page 29

• Piggyback Support on S2b Interface, page 32

• Spare Protocol Types Supported in RAN-NAS IE for S2b Interface, page 33

• Update Bearer Request Sends Exact Same CRI in CCA-U for Dynamic Rule Already Installed, page34

3GPP Changes to the Gx Interface

Feature Information

Summary Data

Modified FunctionalityStatus

21.1Introduced-In Release

21.2Modified-In Release(s)

P-GW, SAEGWApplicable Product(s)

ASR 5500Applicable Platform(s)

Yes, only for Customer specific Gx Dictionary.Default Setting

CSCvc99275Related CDETS ID(s)

Not ApplicableRelated Changes in This Release

P-GW Administration Guide

SAEGW Administration Guide

Related Documentation

Release Change Reference, StarOS Release 21.22

P-GW Changes in Release 21.23GPP Changes to the Gx Interface

Revision History

Revision history details are not provided for features introduced before release 21.2.Important

Release DateReleaseRevision Details

April 27, 201721.2Modified in this release.

Feature ChangesCurrently, P-GW sends 3GPP-User-Location-Info AVP in CCR-U for the event triggerSuccessful_Resource_Allocation, only if there is a change in the ULI. This feature has been implemented tocomply with the 3GPP specifications. With this feature, P-GW sends 3GPP-User-Location-Info in CCR-Ufor the event trigger Successful_Resource_Allocation, even if there is no a change in the ULI. If theResource-Allocation-Notification AVP is included in the Charging-Rule-Install AVP,3GPP-User-Location-Information is included in the CCR-U sent towards the PCRF.

LimitationsFollowing are the limitations of this feature:

1 This feature works with diameter dictionary dpca-custom8.

2 Resource-Allocation-Notification AVP should be included within a Charging-Rule-Install AVP whileinstalling/modifying rule.

APN Configuration Errors Display

Feature Summary

Summary Data


21.2Introduced in Release

Not ApplicableModified in Release(s)

P-GWApplicable Product(s)


P-GW Changes in Release 21.2Feature Changes


EnabledDefault Setting



Release Change ReferenceRelated Documentation

Revision History



April 27, 201721.2New behavior change in this release.

Feature ChangesPrevious Behavior: In releases prior to 21.2, configuration errors were not flagged for a condition where:

1 Rulebase associated with an APN contains the CLI 'billing-records egcdr' (generation of ACS Gz records)and,

2 GTPP group associated with an APN does not have a valid GTPP dictionary configuration (Gz billingcannot be done without this.)

NewBehavior: From release 21.2 onwards, errors for the abovementioned condition is flagged in the followingformat:[local]asr5000# show configuration errors section apn

######################################################################################################Displaying APN-configuration errors######################################################################################################Error : Rulebase <rulebase_name> configured for APN <apn_name> dictates generation ofeGCDRs/PGWCDR'sbut associated GTPP group <group_name> does not have a valid GTPP dictionary configuration.

Total 1 error(s) in this section !

Additionally, warning logs are triggered when a call is brought up on such APN's.



Auto Correction of VoLTE Sessions

Feature Information

Summary Data



Not ApplicableModified-In Release(s)



DisabledDefault Setting




Command Line Interface Reference


Revision History

Revision history details are not provided for features introduced before Release 21.2.Important


April 27, 201721.2New in this release.

Feature DescriptionWhen dynamic rules for IP Multimedia Subsystem (IMS) sessions are lost after a switchover, VoLTE callsare impacted. To recover the calls, the IMS sessions have to be cleared manually to re-establish the PDN withthe correct dynamic rules. The Auto Correction of VoLTE Sessions feature enables a dynamic rule checksuch that when the P-GW receives an RAR (Re-Auth-Request) message, it automatically identifies and rectifiesthe issue without manual intervention. This features only applies to the APN that is configured on that usesthe "RAR" message as a trigger for the check.


P-GW Changes in Release 21.2Auto Correction of VoLTE Sessions

How It WorksAfter the APN receives any RAR message from the Policy Control and Charging Rules Function (PCRF), aRe-Auth- Answer (RAA) message is immediately sent. When the feature is enabled, an additional check isdone at the P-GW to verify if there are any dynamic rules associated with the default bearer. Assuming theSession Initiation Protocol (SIP) rule on the default bearer is recovered, other dedicated bearers are alsorecovered. If dynamic rules are not associated with the default bearer, the call is terminated. Then, a DeleteBearer Request is sent for the default bearer with the cause code - Reactivation Required. To ensure the reasoncode is sent, the APN must be configured with "pdn-behavior ims". Subsequently, aCredit-Control-Request-Type (CCR-T) is sent to PCRF and other diameter interfaces (s6b, Gy, and Rf).

Thereby, the automatic recovery procedure involves termination of a subscriber connection when an anomalyis detected. The subscriber has to then reconnect to the network. The mobile originated or terminated call isrejected for the subscriber where the dynamic rules are lost after a switchover.

Configuring the Auto Correction of VoLTE Sessions FeatureThe following section provides the configuration commands to enable or disable the feature.

Enabling or Disabling the Dynamic Rule CheckThe new CLI command, pdn validate-post-switchover, is added to enable the dynamic rule check for theauto correction of the VoLTE session. To enable this feature, configure the command at the base APN. Thisfeature should be configured only for the VoLTE/IMS APNs for which auto recovery is required.

This feature is disabled by default.

To enable or disable the feature, enter the following commands:

configurecontext <context_name>

apn <apn_name>[no] pdn validate-post-switchoverend

Notes:

• no: Disables the configured Auto correction of VoLTE sessions on the base APN.

• pdn validate-post-switchover: Validates the dynamic rules for automatic recovery after a switchover.

Monitoring and TroubleshootingThis section provides information regarding show commands and/or their outputs in support of the AutoCorrection of VoLTE Sessions feature.

Show CommandsThis section lists all the show commands available to monitor this feature.


P-GW Changes in Release 21.2How It Works

show configuration apn

The above CLI command is introduced to check if the feature is enabled at the APN level. If "pdnvalidate-post-switchover" is present then the feature is enabled.

show active-charging service statistics

This command has been modified to display the following output:

show active-charging service statisticsACS Data Statistics:Packets Dropped due to System-Limit L4-Flows: 0Packets Dropped - Invalid Len in IP Hdr(Dwlink): 0Packets Dropped - Invalid Ver in IP Hdr(Dwlink): 0Packets Not Processed due to Flow-limit: 0Packets Not Processed due to CLP not found: 0Packets Dropped CLP in Preservation Mode: 0Total Pkts: 0Total Collisions in data session hash: 0

ACS Reject Reason:RuleBase Mismatch : 0Bandwidth-Policy Mismatch : 0CBB-Policy Mismatch : 0CF Policy Mismatch : 0No RuleBase configured in APN/Subs: 0No active rule in Rulebase/Subs: 0No Bandwidth-Policy configured in APN/Subs: 0No Resources: 0Max Sessions: 0Reject Probability Exceeded: 0Rule Recovery Failed: 0CDR Flow Control Initiated: 0

Protocol Reject stats:WTP Non-initial PDU: 0WSP-CO Non-initial Connect PDU 0

Dynamic Rule Statistics:Total Subscribers: 0 Current Subscribers: 0Charging Msg Received: 0 Rule Defn Received: 0Installs Received: 0 Removes Received: 0Installs Succeeded: 0 Removes Succeeded: 0ADC Rules Received: 0 Total ADC Rules: 0ADC Install Succeeded: 0 ADC Install Failed: 0ADC Custom Mute Received: 0 ADC Custom Unmute Received: 0ADC Start Sent: 0 ADC Stop Sent: 0L7 Rules Received: 0L7 Install Succeeded: 0 L7 Install Failed: 0Installs Failed: 0 Removes Failed: 0Install Failure Reason:No Resources: 0 No Rulebase Match: 0No RuleName Match: 0 Rulebase Count Exceeded: 0Local Copy Failed: 0 Invalid Protocol: 0Invalid Source Mask: 0 Invalid Dest Mask: 0No Grp-of-Rdef Match: 0ADC Invalid Rule: 0 ADC Invalid Readdress: 0L7 Rule Invalid: 0L7 Protocol Invalid: 0 L7 Field Invalid: 0L7 Operator Invalid: 0 L7 Value Invalid: 0L7 Case-Sens Invalid: 0

Remove Failure Reason:No RuleName Match: 0 No Grp-of-Rdef Match: 0Local Copy Failed: 0

Bandwidth Limiting Statistics:ITC Drops:Uplink Packets: 0 Uplink Bytes: 0


P-GW Changes in Release 21.2Monitoring and Troubleshooting

Downlink Packets: 0 Downlink Bytes: 0Dynamic Rule Bandwidth Limiting Drops:Uplink Packets: 0 Uplink Bytes: 0Downlink Packets: 0 Downlink Bytes: 0

Per-Bearer Bandwidth Limiting Drops:Uplink Packets: 0 Uplink Bytes: 0Downlink Packets: 0 Downlink Bytes: 0

Credit-Control Group Statistics:CC Dropped Uplink Packets: 0 CC Dropped Uplink Bytes: 0CC Dropped Downlink Packets: 0 CC Dropped Downlink Bytes: 0

Readdressing Failure Statistics (Packets):Non SYN Flow: 0 Duplicate Key: 0Dropped Pkts: 0

First-request-only redirections: 0

Fallback Statistics:Bandwidth Policy Applied: 0Bandwidth Policy Failed: 0

Bulk StatisticsThis section lists all the bulk statistics that have been added, modified, or deprecated to support this feature.

ECS Schema

This section displays the new bulk stats that have been added to indicate dynamic recovery failure :

• dyn_rule_recovery_failure:The total number of sessions terminated due to dynamic rule recovery failure.

• dyn_rule_recovery_num_sess_not_terminated:

The total number of sessions that are not terminated after switchover because dynamic rules were notinstalled on the default bearer.

Cisco Ultra Traffic Optimization

Feature Information

Summary Data

New FeatureStatus





P-GW Changes in Release 21.2Cisco Ultra Traffic Optimization






Command Line Interface Reference Guide


Revision History




OverviewIn a high-bandwidth bulk data flow scenario, user experience is impacted due to various wireless networkconditions and policies like shaping, throttling, and other bottlenecks that induce congestion, especially in theRAN. This results in TCP applying its saw-tooth algorithm for congestion control and impacts user experience,and overall system capacity is not fully utilized.

The Cisco Ultra Traffic Optimization (CUTO) solution provides clientless optimization of TCP and HTTPtraffic. The CUTO solution is integrated with Cisco P-GW and has the following benefits:

• Increases the capacity of existing cell sites and therefore, enables more traffic transmission.

• Improves Quality of Experience (QoE) of users by providing more bits per second.

• Provides instantaneous stabilizing andmaximizing per subscriber throughput, particularly during networkcongestion.

How CUTO WorksThe CUTO solution achieves its capabilities by:

• Stabilizing the TCP session at given or optimum target bandwidth.

• Monitoring the TCP session and minimizing bursts.

• Timing adjustment without discarding packets, and with limited buffering.

• Stabilizing and reducing TCP performance jitter.


P-GW Changes in Release 21.2Overview

• Increasing the number of simultaneous sessions served and reducing connection latency.

• Providing adequate logging information for debugging and operability.

ArchitectureStarOS has a highly optimized packet processing framework, the CUTO engine, where the user packets(downlink) are processed in the operating systems user space. The high-speed packet processing, includingthe various functions of the P-GW, is performed in the user space. The CUTO engine is integrated into thepacket processing path of Cisco’s P-GW with a well-defined Application Programming Interface (API) ofStarOS.

The following graphic shows a high-level overview of P-GW packet flow with Traffic Optimization.

Handling of Traffic Optimization Data RecordThe Traffic Optimization Data Record (TODR) is generated only on the expiry of idle-timeout of the CUTOengine. No statistics related to session or flow from P-GW is included in this TODR. The data records are aseparate file for the Traffic Optimization statistics and available to external analytics platform.


P-GW Changes in Release 21.2How CUTO Works

List of Attributes and File Format

All TODR attributes of Traffic Optimization is enabled by a single CLI command. The output is alwayscomma separated, and in a rigid format.

Standard TODR

The following is the format of a Standard TODR:

instance_id,flow_type,srcIP,dstIP,policy_id,dscp,flow_first_pkt_rx_time_ms,flow_last_pkt_rx_time_ms,flow_cumulative_rx_bytesExample:

1,0,173.39.13.38,192.168.3.106,0,0,1489131332693,1489131335924,342292Where:

• instance_id: Instance ID.

• flow_type: Standard flow (0)

• srcIP: Indicates the source IP address.

• dstIP: Indicates the destination IP address.

• policy_id: Indicates the Traffic Optimization policy ID.

• dscp: Indicates the DSCP code for upstream packets.

• flow_first_pkt_rx_time_ms: Indicates the timestamp of when the first packet was detected during theTraffic Optimization.

• flow_last_pkt_rx_time_ms: Indicates the timestamp of when the last packet was detected during theTraffic Optimization.

• flow_cumulative_rx_bytes: Indicates the number of bytes transferred by this flow.

Large TODR

The following is a sample output of a Large TODR.

2,1,2606:ae00:c663:b66f:0000:0058:be03:ae01,0172:0020:0224:0059:2200:0000:0000:0033,0,0,1489572442058,1489572444759,5051223,1489572444755,3900,3900,0,0,0,11,0,0,11,1,1,1489572444755,1950,4,0,0,1,1489572446761,0,2010,0,0,1,1489572448771,0,2007,0,0,1,1489572450778,0,2008,0,0,1,1489572452786,0,2005,0,0,1,1489572454791,0,2003,0,0,1,1489572456794,0,2005,0,0,1,1489572458799,0,2003,0,0,1,1489572460802,0,1848,0,0,0,1489572462650,0,2002,0,0,1,1489572464652,0,107,0,0Where:

• instance_id: Instance ID.

• flow_type: Large flow (1)

• srcIP: Indicates the source IP address.

• dstIP: Indicates the destination IP address.

• policy_id: Indicates the Traffic Optimization policy ID.

• dscp: Indicates the DSCP code for upstream packets.

• flow_first_pkt_rx_time_ms: Indicates the timestamp of when the first packet was detected during theTraffic Optimization.



• flow_last_pkt_rx_time_ms: Indicates the timestamp of when the last packet was detected during theTraffic Optimization.

• flow_cumulative_rx_bytes: Indicates the number of bytes transferred by this flow.

• large_detection_time_ms: Indicates the timestamp of when the flow was detected as Large.

• avg_burst_rate_kbps: Indicates the average rate in Kbps of all the measured bursts.

• avg_eff_rate_kbps: Indicates the average effective rate in Kbps.

• final_link_peak_kbps: Indicates the highest detected link peak over the life of the Large flow.

• recovered_capacity_bytes: Indicates the recovered capacity in Kbps for this Large flow.

• recovered_capacity_ms: Indicates the timestamp of recovered capacity for this Large flow.

• phase_count: Indicates the Large flow phase count.

• min_gbr_kbps: Indicates the Minimum Guaranteed Bit Rate (GBR) in Kbps.

• max_gbr_kbps: Indicates the Maximum Guaranteed Bit Rate (MBR) in Kbps.

• phase_count_record: Indicates the number of phases present in this record.

• end_of_phases: 0 (not end of phases) or 1 (end of phases).

• Large flow phase attributes:

◦phase_type: Indicates the type of the phase

◦phase_start_time_ms: Indicates the timestamp for the start time of the phase.

◦burst_bytes: Indicates the burst size in bytes.

◦burst_duration_ms: Indicates the burst duration in milliseconds.

◦link_peak_kbps: Indicates the peak rate for the flow during its life.

◦flow_control_rate_kbps: Indicates the rate at which flow control was attempted (or 0 if non-flowcontrol phase).

LicensingThe CUTO is a licensed Cisco solution. Contact your Cisco account representative for detailed informationon specific licensing requirements. For information on installing and verifying licenses, refer to theManagingLicense Keys section of the Software Management Operations chapter in the System Administration Guide.

Limitations and RestrictionsThe values which the P-GW chooses to send to the CUTO engine are the values associated from the bearerGBR and bearer MBR.

For this release, only downlink GBR and MBR are sent to the engine for Traffic Optimization.



Configuring CUTOThis section provides information on enabling support for the CUTO solution.

Loading Traffic OptimizationUse the following command under the Global Configuration Mode to load Traffic Optimization as a solution:

configurerequire active-charging traffic-optimizationexit

Enabling or disabling the Traffic Optimization can be done through the Service-scheme framework.Important

Enabling CUTO Configuration ProfileUse the following command under ACS Configuration Mode to enable the Traffic Optimization profile:

configureactive-charging service service_name

traffic-optimization-profileend

Notes:

• The above CLI command enables config-traffic-optim, a new configuration mode.

Configuring the Operating ModeUse the following command to configure the operating mode under Traffic Optimization Profile Mode forthe CUTO Library:


traffic-optimization-profilemode [ active | passive ]end

Notes:

• mode: Sets the mode of operation for Traffic Optimization.

• active: Active mode where both Traffic Optimization and flow monitoring is done on the packet.

• passive: Passive mode where no flow-control is performed but monitoring is done on the packet.

Configuring Threshold ValueUse the following command to configure the threshold value for the TCP flow to be considered for the CUTO:


traffic-optimization-profile


P-GW Changes in Release 21.2Configuring CUTO

heavy-session detection-threshold bytesend

Notes:

• detection-threshold bytes: Specifies the Detection Threshold (in bytes), beyond which it is consideredas heavy session.

bytes must be an integer from 1 to 4294967295.

For optimum Traffic Optimization benefits, it is recommended to set the threshold above 3 MB.

Enabling CUTO Configuration Profile Using Service-scheme FrameworkThe service-scheme framework is used to enable TCP Optimization at APN, Rulebase, QCI and Rule level.There are two main constructs for the service-scheme framework:

• Subscriber-base – This helps in associating subscribers with service-scheme based on the subs-classconfiguration.

◦subs-class – The conditions defined under subs-class enables in classifying the subscribers basedon rule-base, APN, v-APN name. The conditions can also be defined in combination, and both ORas well as AND operators are supported while evaluating them.

• Service-scheme – This helps in associating actions based on trigger conditions which can be triggeredeither at call-setup time, Bearer-creation time, or flow-creation time.

◦trigger-condition – For any trigger, the trigger-action application is based on conditions definedunder the trigger-condition.

◦trigger-actions – Defines the actions to be taken on the classified flow. These actions can beTraffic Optimization, throttle-suppress, and so on.

Session Setup Trigger

The any-match = TRUE, a wildcard configuration, is the only supported condition for this trigger and sothis is applicable to all the flows of the subscriber.

Following is a sample configuration:

configactive-charging service service_name

trigger-action trigger_action_nametraffic-optimizationexit

trigger-condition trigger_condition_name1any-match = TRUEexit

service-scheme service_scheme_nametrigger sess-setup

priority priority_value trigger-condition trigger_condition_name1 trigger-actiontrigger_action_name

exitsubs-class sub_class_name

apn = apn_nameexit



subscriber-base subscriber_base_namepriority priority_value subs-class sub_class_name bind service-scheme service_scheme_name

end

Bearer Creation Trigger

The trigger conditions related to QCI can be used for this trigger, and so this is applicable to all the flows ofspecific bearers.

The following is a sample configuration:




trigger-condition trigger_condition_name2qci = qci_valueexit

service-scheme service_scheme_nametrigger bearer-creation


exitexit

subs-class sub_class_nameapn = apn_nameexit


end

Flow Creation Trigger

The trigger conditions related to rule-name and QCI can be used here, and so this is related to specific flow.

The following is a sample configuration:




trigger-condition trigger_condition_name2qci = qci_valueexit

trigger-condition trigger_condition_name3rule-name = rule_nameexit



service-scheme service_scheme_nametrigger bearer-creation


exitexit

subs-class sub_class_nameapn = apn_nameexit


endNotes:

• trigger_condition_name3 can have only rules, only QCI, both rule and QCI, or either of rule and QCI.

The following table illustrates the different levels of Traffic Optimization and their corresponding SubscriberClass configuration and Triggers.

Subscriber Class configuration and TriggersTraffic Optimization Levels

subs-class sc1any-match = TRUEexit

Sessetup trigger condition is any-match = TRUE

Applicable to all the calls or flows

subs-class sc1rulebase = prepaidexit


Applicable to all calls or flows of arulebase

subs-class sc1apn = cisco.comexit


Applicable to all calls or flows of anAPN

trigger-condition TC1qci = 1exit

Bearer creation trigger condition is TC1

Applicable to all flows of a Bearer

trigger-condition TC1qci = 1rule-name = tcpmulti-line-or all-linesexit

Flow creation trigger condition is TC1

Applicable to a particular flow

In case of LTE to eHRPD handover, since QCI is not valid for eHRPD, it is recommended to configurerule-name as the trigger-condition under service-scheme.

Important



Generating TODRUse the following command under ACS Configuration Mode to enable the Traffic Optimization Data Record(TODR) generation:


traffic-optimization-profiledata-recordend

Notes:

• If previously configured, use the no data-record command to disable generating TODR.

Monitoring and TroubleshootingThis section provides information regarding commands available to monitor and troubleshoot the CUTO onthe P-GW.

CUTO Show Commands and/or OutputsThis section provides information regarding show commands and their outputs in support of CUTO solution.

show active-charging traffic-optimization counters

The show active-charging traffic-optimization counters sessmgr { all | instance number } is a newlyintroduced command where:

• counters – Displays aggregate flow counters/statistics from CUTO engine.

This CLI command is license dependent and visible only if the license is loaded.Important

The following is a sample output:

show active-charging traffic-optimization counters sessmgr allTraffic Optimization Flows:

Active Normal Flow Count : 0Active Large Flow Count : 0Active Managed Large Flow Count : 0Active Unmanaged Large Flow Count : 0Total Normal Flow Count : 0Total Large Flow Count : 0Total Managed Large Flow Count : 0Total Unmanaged Large Flow Count : 0Total IO Bytes : 0Total Large Flow Bytes : 0Total Recovered Capacity Bytes : 0Total Recovered Capacity ms : 0

show active-charging traffic-optimization info

This is a newly introduced show command in Exec Mode where:


P-GW Changes in Release 21.2Monitoring and Troubleshooting

• traffic-optimization – Displays all Traffic Optimization options.

• info – Displays CUTO engine information.

The output of this CLI command displays the version, mode, and configuration values.

The following is a sample output:

show active-charging traffic-optimization infoVersion : 2.0.0Mode : ActiveConfiguration:

Threshold Bytes : 5000000Lower Bandwidth : 0Upper Bandwidth : 0Min Session Time : 0Min Session Size : 0

Civic Address Display Format Changed in Monitor Protocol andshow subscribers pgw-only full all Command Output

Feature Information

Summary Data









P-GW Administration GuideRelated Documentation

Revision History



P-GW Changes in Release 21.2Civic Address Display Format Changed in Monitor Protocol and show subscribers pgw-only full all Command Output



Feature ChangesThe civic address displayed in the show subscribers pgw-only full all command output and the civic addressdisplayed in the monitor protocol of the Gy and Gz interfaces is required to be reformatted for better clarity.The current release addresses this change.

Previous Behavior:

The civic address in the show subscribers pgw-only full all command output and the civic address in themonitor protocol of the Gy and Gz interfaces used to be displayed as follows (CIVIC Address : IN234grna).

Gy Monitor Protocol Output:Civic-Addr: IN234grnaGz Monitor Protocol Output:tWANUserLocationInformationcivicAddress IN234grnaCommand Output:show sub pgw-only full allTWAN User Location Information:

SSID : blizzardBSSID : AA:A0:C9:00:C8:29CIVIC Address : IN234grna

New Behavior:

In the show subscribers pgw-only full all command output, the civic address for the Gy and Gz interfacesis now displayed as follows:

Gy Monitor Protocol Output:Civic-Addr: CC:IN CAType:23 CALen:4 CAVal:grna,Gz Monitor Protocol Output:tWANUserLocationInformation

civicAddress CC:IN CAType:23 CALen:4 CAVal:grnaCommand Output:show sub pgw-only fullTWAN User Location Information

CIVIC Address : CC:IN CAType:23 CALen:4 CAVal:grna,

CLI Configuration for APN String

Feature Summary

Summary Data












Revision History

Revision history details are not provided for features introduced before Release 21.2.Important



Feature ChangesWhen an APN name is configured, if an extra period is added in the APN name, it duplicates the APN nameleading to assert in the egtpinmgr. The current release resolves this issue.

Previous Behavior:When an APN name was configured with an extra period, the APN string was acceptedwith a NULL token. For example, apn internet.com and apn internet..com was allowed.

New Behavior:When an APN name is configured with an extra period, an error is displayed therebydisallowing the use of apn internet.com and apn internet..com strings.

Impact on Customer: Configuring null token APN string is not allowed. The advantage in this release isthat an error is displayed in the egtpinmgr when a null token APN string is configured.



Error Message Provides Additional Information for TDF RuledefValidation

Feature Information

Summary Data




P-GW, S-GW, SAEGW, GGSNApplicable Product(s)





Release Change ReferenceRelated Documentation

Revision History




Feature ChangesDynamic TDF ruledefs are shown as errors in the show config errors command output. This issue is seenwhen the dynamic TDF application-based ruledefs are configured under ECS (Active Charging Service) andnot included in the rulebase. The error message in the log is very concise and limited in information. Thecurrent release addresses this issue.

Previous Behavior:


P-GW Changes in Release 21.2Error Message Provides Additional Information for TDF Ruledef Validation

Dynamic TDF ruledefs configured in the Active Charging Service that do not have any reference to the localrulebases are displayed as errors in the show configuration errors command output with limited information:Error : Ruledef <7-7-FOTA-ONSTAR-DYNAMIC-CNT-01> is not used in anyrulebase or group-of-ruledefs.

Error : group-of-ruledefs <group> is not used in any rulebase.New Behavior:

When dynamic TDF ruledefs are configured in the Active Charging Service and are not included in the localrulebase, the show configuration errors command output continues to display them as errors. These errorsnow provide additional information thereby indicating the possibility that the configured ruledef can be ADC:Error : Ruledef <7-7-FOTA-ONSTAR-DYNAMIC-CNT-01> is not used in anyrulebase or group-of-ruledefs. It can be a dynamically enabled ADCruledef and could be ignored.

Error : group-of-ruledefs <group> is not used in any rulebase. Itcan be a dynamically enabled ADC group and could be ignored.

FAPA Disabled Due to Bandwidth Policy Usage

Feature Information

Summary Data







CSCvd34551Related CDETS ID(s)



Revision History



P-GW Changes in Release 21.2FAPA Disabled Due to Bandwidth Policy Usage



Feature ChangesIn the existing FAPA behavior, it was observed that if traffic hits a charging action that has a "flowlimit-for-bandwidth id" configured, those flows would not engage the FAPA. The current release resolvesthis issue.

Previous Behavior:

When traffic hit a charging action with "flow limit-for-bandwidth id" configured, it did not engage into FAPAeven when all other eligibility conditions of engagement were met.

New Behavior:

When traffic hits a charging action with "flow limit-for-bandwidth id" configured, it now engages into FAPAwhen all other eligibility conditions of engagement are met.

Customer Impact:

Performance improvement can be observed in throughput due to FAPA engagement.

Inclusion of APN AMBR in the Create Session Response

Feature Information

Summary Data

New FunctionalityStatus











P-GW System Administration Guide

Statistics and Counters Reference


Revision History




Feature ChangesCurrently, APN AMBR is included in the Create Session Response even if it is not modified by the PCRF.This feature has been implemented to comply to the 3GPP specifications. With this feature, now APNAMBRwill be included on the S5/S8, S4/S11, and S2a/S2b interfaces if the received APN-AMBR has been modifiedby the PCRF.

Previous Behavior: APN AMBR was included in Create Session Response even if it was not modified bythe PCRF.

New Behavior: Now, APN AMBR will be included on the S5/S8, S4/S11, and S2a/S2b interfaces if thereceivedAPN-AMBRhas beenmodified by the PCRF. A newCLI command has been added to enable/disableinclusion of the APN-AMBR value in the Create Session Response for the modified value received by thePCRF.

P-GW supports APN-AMBR value upto 4294967 kbps.Important

Command Changes

egtpTwo new keywords have been added to the egtp command to enable/disable inclusion of APN-AMBR valuein Create Session Response for modified value received by PCRF.

configurecontext context_name

pgw service service_name[ default | no ] egtp create-session-rsp apn-ambr-always-includeend



Notes:

• default:APNAMBR is not included in the CS Response if APNAMBR is not received from the PCRF.P-GW supports APN-AMBR value upto 4294967 kbps.

• no: Does not include APN AMBR in the CS Response. This is the default behavior.

• create-session-rsp: Provides an option to include APN-AMBR in the Create Session Response

• apn-ambr-always-include: Always includes APN-AMBR IE in Create Session Response.

Performance Indicator Changes

show configThis command has been modified to display the following output:plmn id mcc 404 mnc 00plmn id mcc 404 mnc 005 primaryplmn id mcc 404 mnc 090fqdn host abc.net realm 3gpp.orgdns-client context ISP1associate ggsn-service GGSNassociate egtp-service PGW21EGTPegtp create-session-rsp apn-ambr-always-include

show pgw-service nameThis command has been modified to display the following output:EGTP SGW Restoration Handling: DisabledSession Hold Timer: n/aTimeout: n/aEGTP Modify bearer cmd negotiate qos: DisabledEGTP GnGp Modify bearer res with APN-AMBR: DisabledEGTP Modify bearer res with CHARGING-ID: DisabledEGTP Modify bearer res with CHARGING-FQDN or CHARGING-GW-ADDRESS: DisabledEGTP Modify bearer res with MSISDN: DisabledEGTP Modify Bearer Response with Context Not Found cause if IMEI/IMEISV mismatch: DisabledEGTP Bit Rate in Rounded Down Kbps: DisabledEGTP Suppress Update Bearer Request (no bitrate change): DisabledEGTP Create Session Response with APN-AMBR IE: Enabled

Incorrect Cause Code in Create Session Response by P-GW

Feature Information

Summary Data




P-GW Changes in Release 21.2Performance Indicator Changes


P-GW, SAEGWApplicable Product(s)





P-GW Administration Guide.

SAEGW Administration Guide.


Revision History




Feature ChangesPrevious Behavior: "Create Session Request" with reserved QCI was rejected by the P-GW with the causecode "EGTP_CAUSE_SERVICE_NOT_SUPPORTED" in the "Create Session Response" message.

New Behavior: Now, "Create Session Request" with reserved QCI is rejected by the P-GW with the causecode "EGTP_CAUSE_MANDATORY_IE_INCORRECT" in the "Create Session Response" message.

Impact on Customer: Cause code change in Create Session Response message will be seen.

Increase in ARP Support

Feature Information

Summary Data






P-GW, SAEGW, S-GWApplicable Product(s)





Command Line Interface ReferenceRelated Documentation

Revision History



April 27, 201721.2Modified in this release.

Feature ChangesEarlier, the maximum number of three ARPs that were supported on eMPS profile. This feature extendsconfiguration support to increase maximum of eight ARPs.

This support has been extended to other CLI commands as well. Following are the features that can beconfigured for maximum 8 ARPs:

1 allow-on-congestion CLI command

2 eMPS profile

3 Overload profile and self-protection CLI

4 GTPC RLF Throttling CLI

After upgrade/downgrade, existing sessions will continue its eMPS characteristics until the next transactionon the session. It will fetch the new configuration during the first transaction after upgrade/downgrade.

Important



Configuring the ARPsThe following section provides the different configuration commands that support configuration of ARPs.

allow-on-congestionThis CLI command has been modified to support configuration of max 8 ARP's for the "allow-on-congestion"congestion CLI.

egtp-service:

configurecontext <context_name>

egtp-service <egtp-service_name>[no] gtpc allow-on-congestion arp <arp-value>end

Notes:

• arp: Configures the gateway to allow calls for this ARP even under congestion. Enter ARP PriorityLevel, ranging from 1...15 - integer 1..15.

emps-profileThis CLI command has been modified to extend the support from 3 eMPS ARPs to 8 eMPS ARPs.

configure[no] emps-profile <emps-profile_name>earp {[1..15] {[1.15] {[1..15]}}}end

Notes:

• earp: Configures eARP values ranging from 1...15 - integer 1..15. Configures maximum of 8 eARPpriority level values so that sessions with configured eARP priority values can be marked as eMPSsessions.

gtpc-overload-control-profileCurrently, P-GW supports 3 ARPs under Overload Profile and self-protection CLI's. This CLI command hasbeen modified to extend the support to 8 eMPS ARPs.

configuregtpc-overload-control-profile<gtpc-overload-control-profile_name>

[no] throttling-behavior {earp {1|2|3|4|5|6|7|8|9|10|11|12|13|14|15}*} {exclude}[no] self-protection-behavior {earp {1|2|3|4|5|6|7|8|9|10|11|12|13|14|15}*} {exclude}end

Notes:

• earp: Configures maximum of 8 EARP priority level values so that incoming request messages forconfigured earp priority values are not rejected even if system is under self-protection mode.


P-GW Changes in Release 21.2Configuring the ARPs

throttling-behaviour already supports 8 ARP values. No change has been made to this command.Important

RLF already supports 8 ARP values. No change has been made to this command.Important

NEMO-LMA Heartbeat

Feature Information

Summary Data

New FeatureStatus






CSCuw08182Related CDETS ID(s)

NEMO PMIPv6 Heartbeat on LMA (SAEGW)Related Changes in This Release



SAEGW Administration Guide


Revision History





P-GW Changes in Release 21.2NEMO-LMA Heartbeat

Feature DescriptionPath management mechanism through Heartbeat messages between the Mobile Router (MR) and NetworkMobility-Local Mobility Anchor (NEMO-LMA) is important to know the availability of the peers, to detectfailures, quickly inform peers if recovery from the node fails and allow a peer to take appropriate action.

How It WorksThe MR and the LMA exchange Heartbeat messages at regular intervals to detect the status of availabilitybetween them. The NEMO-LMA initiates a Heartbeat exchange with theMR, by sending a Heartbeat Requestmessage, to check if theMR is reachable. The NEMO-LMA records the sequence number of the last HeartbeatRequest message and is used to match the corresponding Heartbeat Response. The NEMO-LMA responds toa Heartbeat Request message with a Heartbeat Response message, irrespective of whether there is PMIPv6session with the corresponding peer.

Binding Error

When the Binding Error message, with status set to 2, is received in response to a Heartbeat Request message,the NEMO-LMA does not use Heartbeat messages further with the corresponding peer.

Failure Detection

When the LMA node does not receive the Heartbeat response for the configurable parametermax-heartbeat-retransmission <value> exceed-action drop-session, the NEMO-LMA concludes that thepeer is not reachable. As such, the Heartbeat request to the peer is stopped and clears the NEMO-LMA sessionwithout any traps.

Restart Detection

If the restart counter value is different from the previous received value, then it assumes that the peer hadcrashed and recovered. And so, the existing NEMO-LMA sessions are cleared.

License RequirementsUse of NEMO requires that a valid license key be installed. Contact your Cisco account or Supportrepresentative for information on how to obtain a license.

Configuring NEMO-LMA HeartbeatUse the following commands under LMAService ConfigurationMode to clear NEMO-LMA sessions withoutgenerating traps:

configurecontext context_name

lma-service service_nameheartbeat retransmission max number [ exceed-action drop-session ]end

Notes:


P-GW Changes in Release 21.2Feature Description

• retransmission max: The maximum number of heartbeat retransmissions allowed. The number mustbe an integer from 0 to 15. Default: 3

• exceed-action: Specifies the action to be taken after the maximum number of Heartbeat retransmissionsis reached.

• drop-session: Used for dropping the session when path failure is detected.

• The exceed-action and drop-session keywords are valid only for NEMO-LMA sessions and takeseffect if the Heartbeat feature is enabled.

Monitoring and Troubleshooting the NEMO-LMA HeartbeatThe following sections describe commands available to monitor and troubleshoot the feature.

monitor protocolWhen using themonitor protocol command, enable option 48 to monitor the Heartbeat Request/Replymessages.

show lma-service statisticsUse this command to see the statistics related to Heartbeat messages. The output generated appears similarto the following:Total Disconnects: 1

Lifetime expiry: 0 Deregistrations: 0Admin Drops: 0 Path Failure Drops: 1Other Reasons: 0

show session disconnect-reasonsUse this command to see the call disconnected due to heartbeat path failure. The output generated appearssimilar to the following:mme-guti_realloc_failed-detach(615) 0 0.00000mme-pcscf-rest-detach(616) 0 0.00000Reject-ho-old-tun-path-failure(617) 0 0.00000mip-path-failure(618) 0 0.00000

Bulk Statistics

LMA Schema

The following new bulk statistics variable is added to the LMA schema in support of this feature:

• lma-pathfailsessionscleared – If any path failures/restarted counter value changes for the NEMO-LMAHeartbeat feature, the sessions disconnection counter is incremented.


P-GW Changes in Release 21.2Monitoring and Troubleshooting the NEMO-LMA Heartbeat

Piggyback Support on S2b Interface

Feature Information

Summary Data

New FunctionalityStatus



ePDG, P-GWApplicable Product(s)



CSCvc97504, CSCvd09958Related CDETS ID(s)

ePDG S2b Piggybacking SupportRelated Changes in This Release

ePDG Administration Guide



Revision History




Feature DescriptionThis feature supports piggybacking of "Create Session Response" and "Create Bearer Request" messages onePDG and P-GWover the S2b interface. If piggybacking flag is set by the ePDG in the Create Session Request,P-GW can now send Create Session Response and Create Bearer Request together to the ePDG and eliminatethe possibility of reordering of these messages.


P-GW Changes in Release 21.2Piggyback Support on S2b Interface

21-1-Change-Reference_chapter_01101.pdf#unique_17

Spare Protocol Types Supported in RAN-NAS IE for S2b Interface

Feature Information

Summary Data












Revision History




Feature ChangesJust after starting the call model, P-GW receives a large number of negative CBResponse and UBResponsemessages. The RAN-NAS IE coming in the messages have spare protocol types (6 to 15) that are not supported.Protocol types (6 to 15) are sent by ePDG to report internal failures. The current release resolves this issue.

Previous Behavior:

For S2b interface, spare protocol types in RAN/NAS IE were not supported.


P-GW Changes in Release 21.2Spare Protocol Types Supported in RAN-NAS IE for S2b Interface

New Behavior:

For RAN/NAS IE, the spare protocol types are now supported using a CLI and the length for cause value forthese spare IE can also be configured. This support is available only for the S2b interface.

Spare protocol types are supported only for Failed Create Bearer Response/Failed Update Bearer Responsemessages.

The cause value length depicts the length of cause value to be decoded; it can be 1 octet or 2 octets. The defaultvalue is 2 octets. Mismatch between length of cause value IE and configured CLI value leads to IE beingignored.

Customer Impact:

CLI needs to be added for allowing spare RAN/NAS protocol types.

Command Changesran-nas decode proto-type-spare cause-value-length

This is a new command added to the eGTP Service Configuration mode. Run this command to enable spareprotocol types for the S2b interface:configcontext <context_name>egtp-service <service_name>ran-nas decode proto-type-spare cause-value-length <1 | 2>

Update Bearer Request Sends Exact Same CRI in CCA-U forDynamic Rule Already Installed

Feature Information

Summary Data










P-GW Changes in Release 21.2Command Changes


Revision History




Feature ChangesAfter dynamic rule is installed, CCR-U is sent with SUCCESSFUL_RESOURCE_ALLOCATION. Howeveron sending the same CRI as that of the dynamic rule in CCA-U, the Update Bearer Request (UBR) is sent bythe P-GW for this rule even if the CRI is exactly the same. This causes the CCR-U and UBR to continouslyloop. The current release resolves this issue.

Ideally, if invalid values of pre-emption-capability or pre-emption-vulnerability AVP is sent from the PCRF,IMSA validates them. If it is invalid, it converts them to default values, that is, pre-emption-capability: 1(Default Value) and pre-emption-vulnerability: 0 (Default Value).

Previous Behavior:

While installing dynamic rules, if PCRF sends pre-emption-capability or pre-emption-vulnerability (or both)AVP values other than 0 (Enabled) or 1 (Disabled), the same values were propagated to the P-GW.

New Behavior:

While installing dynamic rules, if PCRF sends pre-emption-capability or pre-emption-vulnerability (or both)AVP values other than 0 (Enabled) or 1 (Disabled), on receiving these, the IMSA converts them to the defaultvalues (as per 3GPP TS 29.212 V13.2.0 Section: 5.3.46/47) and propagates them to the P-GW.





Documents

P-GW Changes in Release 21 · • NEMO-LMAHeartbeat,page29 ... Release Change Reference, StarOS Release 21.2 10 P-GW Changes in Release 21.2 How CUTO Works. List of Attributes and