RiaB Logging and Monitoring
This tutorial explains how to run RiaB with logging and monitoring enabled.
For logging, RiaB enables the collection of logs using fluent-bit, storage in elasticsearch DB, and presentation of metrics in grafana (kibana can also be used). Logs are collected from all pods in the RiaB k8s cluster. For monitoring, RiaB enables a colletions of pods and node metrics using prometheus exporters and grafana.
Enabling logging/monitoring in the sdran-in-a-box-values yaml file
To enable logging/monitoring in RiaB, in the sdran-in-a-box-values.yaml (available for the latest version) file remove the comments of the lines below:
# Monitoring/Logging
fluent-bit:
enabled: true
opendistro-es:
enabled: true
prometheus-stack:
enabled: true
Associated with the monitoring of sdran components is the onos-exporter, the exporter for ONOS SD-RAN (µONOS Architecture) to scrape, format, and export onos KPIs to TSDB databases (e.g., Prometheus). Currently the implementation supports Prometheus. In order to enable onos-exporter, as shown below, make sure the prometheus-stack is enabled too.
prometheus-stack:
enabled: true
onos-exporter:
enabled: true
Accessing grafana and looking for logs and metrics
After modified the values file, then run the make command to instantiate RiaB. After deployed, the services and pods related to logging and monitoring will be shown as:
$ kubectl -n riab get svc
...
alertmanager-operated ClusterIP None <none> 9093/TCP,9094/TCP,9094/UDP 90s
prometheus-operated ClusterIP None <none> 9090/TCP 90s
sd-ran-fluent-bit ClusterIP 192.168.205.134 <none> 2020/TCP 90s
sd-ran-grafana ClusterIP 192.168.209.213 <none> 80/TCP 90s
sd-ran-kube-prometheus-sta-alertmanager ClusterIP 192.168.166.174 <none> 9093/TCP 90s
sd-ran-kube-prometheus-sta-operator ClusterIP 192.168.152.79 <none> 443/TCP 90s
sd-ran-kube-prometheus-sta-prometheus ClusterIP 192.168.199.115 <none> 9090/TCP 90s
sd-ran-kube-state-metrics ClusterIP 192.168.155.231 <none> 8080/TCP 90s
sd-ran-opendistro-es-client-service ClusterIP 192.168.183.47 <none> 9200/TCP,9300/TCP,9600/TCP,9650/TCP 90s
sd-ran-opendistro-es-data-svc ClusterIP None <none> 9300/TCP,9200/TCP,9600/TCP,9650/TCP 90s
sd-ran-opendistro-es-discovery ClusterIP None <none> 9300/TCP 90s
sd-ran-opendistro-es-kibana-svc ClusterIP 192.168.129.238 <none> 5601/TCP 90s
sd-ran-prometheus-node-exporter ClusterIP 192.168.137.224 <none> 9100/TCP 90s
$ kubectl -n riab get pods
...
alertmanager-sd-ran-kube-prometheus-sta-alertmanager-0 2/2 Running 0 43s
sd-ran-fluent-bit-x75mt 1/1 Running 0 43s
sd-ran-grafana-584fbb69cf-cvzcn 2/2 Running 0 43s
sd-ran-kube-prometheus-sta-operator-5f47f669dd-xkm7k 1/1 Running 0 43s
sd-ran-kube-state-metrics-6f675bf9cf-kzkdc 1/1 Running 0 43s
sd-ran-opendistro-es-client-78649698c8-tfh47 1/1 Running 0 42s
sd-ran-opendistro-es-data-0 1/1 Running 0 43s
sd-ran-opendistro-es-kibana-7cff67c748-vp4g8 1/1 Running 0 43s
sd-ran-opendistro-es-master-0 1/1 Running 0 43s
sd-ran-prometheus-node-exporter-grt5k 1/1 Running 0 43s
Port forward grafana service
Make a port-forward rule to the grafana service on port 3000.
kubectl -n riab port-forward svc/sd-ran-grafana 3000:80
Access the grafana dashboards
Open a browser and access localhost:3000
. The credentials to access grafana are: username: admin
and password: prom-operator
.
To look at the grafana dashboard for the sdran component logs, check in the left menu of grafana the option dashboards and selec the submenu Manage (or just access in the browser the address http://localhost:3000/dashboards
).
In the menu that shows, look for the dashboard named Kubernetes / Logs / Pod
. It is possible to access this dashboard directly via the address http://localhost:3000/d/e2QUYvPMk/kubernetes-logs-pod?orgId=1&refresh=10s
.
In the top menu, the dropdown menus allow the selection of the Namespace riab
and one of its Pods. It is also possible to type a string to be found in the logs of a particular pod using the field String.
Similarly, other dashboards can be found in the left menu of grafana, showing for instance each pod workload in the dashboad Kubernetes / Compute Resources / Workload
.
Accessing kibana and looking for logs
The helm chart repository used by sdran for the kibana and elasticsearch instantiation had its settings modified to disable security modules. However to disable kibana security modules a different approach must be followed so it can be accessible via browser (see https://opendistro.github.io/for-elasticsearch-docs/docs/security/configuration/disable/).
The simples way to do that, is creating a new docker image for kibana. Thus, create a file named Dockerfile
with the following content.
FROM amazon/opendistro-for-elasticsearch-kibana:1.13.0
RUN /usr/share/kibana/bin/kibana-plugin remove opendistroSecurityKibana
Then in the folder where the Dockerfile is located run the following command.
docker build --tag=amazon/opendistro-for-elasticsearch-kibana-no-security:1.13.0 .
Add the following lines to the end of the file sdran-in-a-box-values.yaml.
opendistro-es:
kibana:
image: amazon/opendistro-for-elasticsearch-kibana-no-security
imagePullPolicy: IfNotPresent
Then execute RiaB make command and after finished, run the port-forward to the kibana service (as stated below) and access in the browser the address localhost:5601
. The kibana portal will be accessible without login to browse the logs of the RiaB components.
kubectl -n riab port-forward svc/sd-ran-opendistro-es-kibana-svc 5601
About the onos-exporter metrics
The onos-exporter scrapes the following KPIs from the onos components:
onos-e2t:
onos_e2t_connections
Description: The number of e2t connections.
Value: float64.
Dimensions/Labels: connection_type, id, plmnid, remote_ip, remote_port.
Example: onos_e2t_connections{connection_type=”G_NB”,id=”00000000003020f9:0”,plmnid=”1279014”,remote_ip=”192.168.84.46”,remote_port=”35823”,sdran=”e2t”} 1
onos-e2sub:
onos_e2sub_subscriptions
Description: The number of e2sub subscriptions.
Value: float64.
Dimensions/Labels: appid, e2nodeid, id, lifecycle_status, revision, service_model_name, service_model_version.
Example: onos_e2sub_subscriptions{appid=”onos-kpimon-v2”,e2nodeid=”00000000003020f9:0”,id=”2a0e7586-b8ac-11eb-b363-6f6e6f732d6b”,lifecycle_status=”ACTIVE”,revision=”17”,sdran=”e2sub”,service_model_name=”oran-e2sm-kpm”,service_model_version=”v2”} 1
onos-kpimon-v2:
onos_xappkpimon_rrc_conn_avg
Description: RRCConnAvg the mean number of users in RRC connected mode during each granularity period.
Value: float64.
Dimensions/Labels: cellid, egnbid, plmnid.
Example: onos_xappkpimon_rrc_conn_avg{cellid=”343332707639553”,egnbid=”5153”,plmnid=”1279014”,sdran=”xappkpimon”} 5
onos_xappkpimon_rrc_conn_max
Description: RRCConnMax the max number of users in RRC connected mode during each granularity period.
Value: float64.
Dimensions/Labels: cellid, egnbid, plmnid.
Example: onos_xappkpimon_rrc_conn_max{cellid=”343332707639553”,egnbid=”5153”,plmnid=”1279014”,sdran=”xappkpimon”} 5
onos_xappkpimon_rrc_connestabatt_tot
Description: RRCConnEstabAttTot total number of RRC connection establishment attempts.
Value: float64.
Dimensions/Labels: cellid, egnbid, plmnid.
Example: onos_xappkpimon_rrc_connestabatt_tot{cellid=”343332707639553”,egnbid=”5153”,plmnid=”1279014”,sdran=”xappkpimon”} 5
onos_xappkpimon_rrc_connestabsucc_tot
Description: RRCConnEstabSuccTot total number of successful RRC Connection establishments.
Value: float64.
Dimensions/Labels: cellid, egnbid, plmnid.
Example: onos_xappkpimon_rrc_connestabsucc_tot{cellid=”343332707639553”,egnbid=”5153”,plmnid=”1279014”,sdran=”xappkpimon”} 5
onos_xappkpimon_rrc_connreestabatt_hofail
Description: RRCConnReEstabAttHOFail total number of RRC connection re-establishment attempts due to Handover failure.
Value: float64.
Dimensions/Labels: cellid, egnbid, plmnid.
Example: onos_xappkpimon_rrc_connreestabatt_hofail{cellid=”343332707639553”,egnbid=”5153”,plmnid=”1279014”,sdran=”xappkpimon”} 5
onos_xappkpimon_rrc_connreestabatt_other
Description: RRCConnReEstabAttOther total number of RRC connection re-establishment attempts due to Other reasons.
Value: float64.
Dimensions/Labels: cellid, egnbid, plmnid.
Example: onos_xappkpimon_rrc_connreestabatt_other{cellid=”343332707639553”,egnbid=”5153”,plmnid=”1279014”,sdran=”xappkpimon”} 5
onos_xappkpimon_rrc_connreestabatt_reconfigfail
Description: RRCConnReEstabAttreconfigFail total number of RRC connection re-establishment attempts due to reconfiguration failure.
Value: float64.
Dimensions/Labels: cellid, egnbid, plmnid.
Example: onos_xappkpimon_rrc_connreestabatt_reconfigfail{cellid=”343332707639553”,egnbid=”5153”,plmnid=”1279014”,sdran=”xappkpimon”} 5
onos_xappkpimon_rrc_connreestabatt_tot
Description: RRCConnReEstabAttTot total number of RRC connection re-establishment attempts.
Value: float64.
Dimensions/Labels: cellid, egnbid, plmnid.
Example: onos_xappkpimon_rrc_connreestabatt_tot{cellid=”343332707639553”,egnbid=”5153”,plmnid=”1279014”,sdran=”xappkpimon”} 5
onos-pci:
onos_xapppci_conflicts
Description: The number of pci conflicts.
Value: float64.
Dimensions/Labels: cellid.
Example: onos_xapppci_conflicts{cellid=”343332707639809”,sdran=”xapppci”} 9
Accessing the onos-exporter metrics
To look at the onos-exporter metrics, it’s possible to access the onos-exporter directly or visualize the metrics in grafana.
To access the metrics directly have a port-forward kubectl command for onos-exporter service:
kubectl -n riab port-forward svc/onos-exporter 9861
Then access the address localhost:9861/metrics
in the browser. The exporter shows golang related metrics too.
To access the metrics using grafana, proceed with the access to grafana. After accessing grafana go to the Explore
item on the left menu, on the openned window select the Prometheus data source, and type the name of the metrics to see its visualization and click on the Run query
button.
Enabling custom sdran logging parser
In fluentbit there is the possibility to declare custom parsers for particular kubernetes pods, check this to learn more. It can be configured via annotations to a deployment pod. The sdran components have a particular log format, as they use the logging package of onos-lib-go. In the sdran logs presented by grafana, there is no strict parsing, i.e., the logs are presented in a raw format.
A fluentbit logging parser for the sdran components is defined in the sdran chart, check its structure definition as shown below:
customParsers: |
[PARSER]
Name sdran
Format regex
Regex ^(?<timestamp>\d{4}-\d{2}-\d{2}.\d{2}:\d{2}:\d{2}.\d{3}).\s+(?<logLevel>\S+)\s+(?<process>\S+)\s+(?<file>\S+):(?<lineNo>\d+)\s+(?<log>.*)$
Time_Key timestamp
Time_Format %Y-%m-%dT%H:%M:%S.%L
In order to enable this logging parser, there is the need to define annotations in the template deployments of the helm charts for those components. This annotation is defined as the value below:
spec:
template:
metadata:
annotations:
fluentbit.io/parser: sdran
Given the application of the fluentbit logging parser to a Kubernetes pod deployment via the annotations above, the sdran log format in grafana for such component will be parsed, as the Regex patter of the custom sdran parser, with the fields: timestamp, logLevel, process, file, lineNo, log. The field log will be the last one presenting the message in the logs shown in the grafana SDRAN logging dashboard.
Cleaning
As logging and monitoring are enabled via the same sdran umbrella helm chart, when running the commands to reset the test and clean the environment, logging and monitoring k8s components will also be removed in RiaB.