1. Cloud Native Security Edge Protection Proxy (SEPP) User's Guide
  2. Metrics, KPIs and Alerts

3 Metrics, KPIs and Alerts

SEPP Metrics

This section provides information about the SEPP metrics.

Microservice Name : Consumer N32c
SL.No. Prometheus Stat Metrics Name Metrics Description(including pegging condition) Dimensions
1. sepp_n32c_handshake_failure If N32c Handshake Procedure fails, this metrics will be pegged and corresponding alarm will be raised.
  1. app(consumer, producer)
  2. peer_domain
  3. peer_fqdn
  4. n32c_procedure
  5. vendor
  6. nfInstanceId
  7. kubernetes_namespace
2. sepp_cn32c_handshake_requests_total

Total number of requests send over n32c for handshake procedure.

Condition:: When sepp initiates any handshake procedure requests towards peer sepp.
  1. peer_domain
  2. peer_fqdn
  3. n32c_procedure
  4. vendor
  5. nfInstanceId
  6. kubernetes_namespace
  7. direction=egress
3. sepp_cn32c_handshake_response_total

Total number of responses received over n32c for handshake procedure.

Condition:: When sepp receives any handshake procedure response from peer sepp. It can be successful or failure based on responseCode
  1. peer_domain
  2. peer_fqdn
  3. n32c_procedure
  4. responseCode(2xx,4xx,5xx)
  5. vendor
  6. nfInstanceId
  7. kubernetes_namespace
  8. direction=ingress
4. sepp_cn32c_handshake_initiation_req_total

Total number of Handshake Initiation requests received from config-mgr.

Condition:: When handshake initiation requests received from config-mgr.
  1. peer_domain
  2. peer_fqdn
  3. vendor
  4. nfInstanceId
  5. kubernetes_namespace
5. sepp_cn32c_handshake_delete_req_total

Total number of Handshake context delete requests received from config-mgr.

Condition:: When handshake context delete requests received from config-mgr.
  1. peer-domain
  2. peer_fqdn
  3. vendor
  4. nfInstanceId
  5. kubernetes_namespace
Microservice Name : Producer N32c
S.No. Prometheus Stat Metrics Name Metrics Description(including pegging condition) Dimensions
1. sepp_pn32c_handshake_requests_total

Total number of requests received over n32c for handshake procedure.

Condition:: When any handshake procedure request received from peer sepp.
  1. peer_domain
  2. peer_fqdn
  3. n32c_procedure
  4. vendor
  5. nfInstanceId
  6. kubernetes_namespace
  7. direction=ingress
2. sepp_pn32c_handshake_response_total

Total number of responses sent over n32c for handshake procedure.

Condition:: When sepp sends response to handshake procedure received. It can be a success response or failure response based on success code.
  1. peer_domain
  2. peer_fqdn
  3. n32c_procedure
  4. statusCode(2xx,4xx,5xx)
  5. vendor
  6. nfInstanceId
  7. kubernetes_namespace
  8. direction=egress
Microservice Name : Consumer N32f
S.No. Prometheus Stat Metrics Name Metrics Description(including pegging condition) Dimensions
1. sepp_cn32f_requests_failure

Total number of requests failed to be sent from cn32f to remote SEPP.

Condition:: When any error or exception occurs on cn32f side because of which request is not sent to pn32f.
  1. peer_domain
  2. peer_id
  3. peer_fqdn
  4. vendor
  5. nfInstanceId
  6. kubernetes_namespace
  7. statuscode
  8. error_msg
2. sepp_cn32f_response_failure

Total number of response failed to be sent from cn32f pod to NF.

Condition:: When any error or exception occurs on cn32f side because of which request is not sent to NF.
  1. peer_domain
  2. peer_id
  3. peer_fqdn
  4. vendor
  5. nfInstanceId
  6. kubernetes_namespace
  7. statuscode
  8. error_msg
3. sepp_cn32f_requests

Total number of requests received from NF.

Condition:: When a request is received on InboundInterface of cn32f.
  1. peer_domain
  2. peer_id
  3. peer_fqdn
  4. vendor
  5. nfInstanceId
  6. kubernetes_namespace
4. sepp_cn32f_response

Total number of response received from remote SEPP.

Condition:: When a response is received on OutboundInterface of cn32f.
  1. peer_domain
  2. peer_id
  3. peer_fqdn
  4. vendor
  5. nfInstanceId
  6. kubernetes_namespace

Microservice Name: pN32f

S.No Prometheus Stat Metric Name Metric Description(including pegging condition) Dimensions
1 sepp_pn32f_requests_rx

Number of requests received from Peer Sepp

Condition: When a request reaches pn32f from peer Sepp.
  1. vendor
  2. nfInstanceId
  3. kubernetes_namespace
  4. kubernetes_podname
  5. peer_fqdn
2 sepp_pn32f_requests_tx Number of requests transmitted to NF.Condition: When a request transmits a message to NF
  1. vendor
  2. nfInstanceId
  3. kubernetes_namespace
  4. kubernetes_podname
  5. peer_fqdn
3 sepp_pn32f_responses_rx

Number of responses received from EGW

Condition: When a response reaches pn32f from EGW.
  1. vendor
  2. nfInstanceId
  3. kubernetes_namespace
  4. kubernetes_podname
  5. statusCode(2xx,4xx,5xx)
  6. error-reason(in case of EGW failure)
  7. peer_fqdn
4 sepp_pn32f_responses_tx Number of responses transmitted to cSepp.Condition: When a responses transmits a message to cSepp
  1. vendor
  2. nfInstanceId
  3. kubernetes_namespace
  4. kubernetes_podname
  5. statusCode(2xx,4xx,5xx)
  6. peer_fqdn

Note:

The templates for dashboard creations not shared in the deliverables. So dashboard needs to be created manually based on needs

SEPP KPIs

This section provides information about the SEPP KPIs.

KPI Details SEPP Microservice Metrics Used Service Operation Response Code
cn32c Handshake Success rate cn32c (sum(ocsepp_cn32c_handshake_response_total)/sum(ocsepp_cn32c_handshake_requests_total))*100 n32c Handshake 200
cn32f Routing Success Rate cn32f (sum(ocsepp_cn32f_response_total)/sum(ocsepp_cn32f_requests_total))*100 n32f message forward All
cn32f Requests Rate Per Remote SEPP cn32f sum(irate(ocsepp_cn32f_requests_total[2m]))by(PEER_DOMAIN, PEER_FQDN, PLMN_ID) n32f message forward All
cn32f Response Rate Per Remote SEPP cn32f sum(irate(ocsepp_cn32f_response_total[2m])) by(PEER_DOMAIN, PEER_FQDN, PLMN_ID) n32f message forward All
cn32c Handshake Failure Per Remote SEPP cn32c sum(ocsepp_n32c_handshake_failure)by(peer_domain, peer_fqdn, peer_plmn_id) n32c Handshake 4xx and 5xx
pn32c Handshake Success rate pn32c (sum(ocsepp_pn32c_handshake_response_total)/sum(ocsepp_pn32c_handshake_requests_total))*100 n32c Handshake 200
pn32f Routing Success Rate pn32f (sum(ocsepp_pn32f_response_total)/sum(ocsepp_pn32f_requests_total))*100 n32f message forward All
pn32f Requests Rate pn32f sum(irate(ocsepp_pn32f_requests_rx_total[2m])) n32f message forward All
pn32f Response Rate pn32f sum(irate(ocsepp_pn32f_response_rx_total[2m])) n32f message forward All
pn32c Handshake Failure Per Remote SEPP pn32c sum(ocsepp_n32c_handshake_failure)by(peer_domain, peer_fqdn, peer_plmn_id) n32c Handshake 4xx and 5xx

SEPP Alerts

This section provides information about the SEPP alerts and their configuration.

Table 3-1 SEPP Alerts

SL.no. Alert Severity oid Dimensions Description Remarks
1 SEPPCn32cHandshakeFailureAlert Major 1.3.6.1.4.1.323.5.3.46.1.2.2001 reason Handshake procedure has failed on Consumer sepp SEPP-1.4
2 SEPPPn32cHandshakeFailureAlert Major 1.3.6.1.4.1.323.5.3.46.1.2.3001 reason Handshake procedure has failed on Producer sepp SEPP-1.4
3 SEPPN32fRoutingFailure warning 1.3.6.1.4.1.323.5.3.46.1.2.4001 error_msg N32f service not able to forward message SEPP-1.4
4 SEPPPodMemoryUsageAlert warning 1.3.6.1.4.1.323.5.3.46.1.2.4003   Pod memory usage is above threshold ( 70% ) SEPP-1.4
5 SEPPPodCpuUsageAlert warning 1.3.6.1.4.1.323.5.3.46.1.2.4002   Pod CPU usage is above threshold ( 70% ) SEPP-1.4

SEPP Alert Configuration

This section describes the Measurement based Alert rules configuration for SEPP. The Alert Manager uses the Prometheus measurements values as reported by microservices in conditions under alert rules to trigger alerts.

Steps for SEPP Alert Configuration in Prometheus

_Namespace_ - Prometheus NameSpace
  1. Find the config map to configure alerts in prometheus server by executing the following command:
    kubectl get configmap -n <Namespace>

    where, <Namespace> is the prometheus server namespace used in helm install command.

  2. Take Backup of current config map of prometheus server by executing the following command:
    kubectl get configmaps _NAME_-server -o yaml -n _Namespace_ > /tmp/tempConfig.yaml
    where, <Namespace> is the prometheus server namespace used in helm install command.
    For Example, assuming chart name is "prometheus-alert", so "_NAME_-server" becomes "prometheus-alert-server", execute the following command to find the config map:
    kubectl get configmaps prometheus-alert-server -o yaml -n prometheus-alert2 > /tmp/tempConfig.yaml
  3. Check if alertssepp is present in the t_mapConfig.yaml file by executing the following command:
    cat /tmp/t_mapConfig.yaml  | grep alertssepp
  4. If alertssepp is present, delete the alertssepp entry from the t_mapConfig.yaml file, by executing the following command:
    sed -i '/etc\/config\/alertssepp/d' /tmp/t_mapConfig.yaml
  5. If alertssepp is not present, add the alertssepp entry in the t_mapConfig.yaml file by executing the following command:
    sed -i '/rule_files:/a\    \- /etc/config/alertssepp'  /tmp/t_mapConfig.yaml
  6. Reload the config map with the modifed file by executing the following command:
    kubectl replace configmap <Name> -f /tmp/t_mapConfig.yaml
  7. Add seppAlertRules.yaml file into prometheus config map under filename of SEPP alert file by executing the following command :
    kubectl patch configmap <Name> -n <Namespace> --type merge --patch
"$(cat <PATH>/seppAlertRules.yaml)"
  8. Restart prometheus-server pod.
  9. Verify the alerts in prometheus GUI.

Note:

Prometheus server takes updated configmap reloaded after sometime automatically (~20 sec)