Getting Data In (GDI) with OTel and UF

Deploy Complex Environments and Capture Metrics

4 min

Objective: Learn how to efficiently deploy complex infrastructure components such as Kafka and MongoDB to demonstrate metrics collection with Splunk O11y IM integrations

Duration: 15 Minutes

Scenario

A prospect uses Kafka and MongoDB in their environment. Since there are integrations for these services, you’d like to demonstrate this to the prospect. What is a quick and efficient way to set up a live environment with these services and have metrics collected?

1. Where can I find helm charts?

2. Review Apache Kafka packaged by Bitnami

We will deploy the helm chart with these options enabled:

3. Review MongoDB(R) packaged by Bitnami

We will deploy the helm chart with these options enabled:

4. Install Kafka and MongoDB with helm charts

bash 
helm repo add bitnami https://charts.bitnami.com/bitnami
bash 
helm install kafka --set replicaCount=3 --set metrics.jmx.enabled=true --set metrics.kafka.enabled=true  --set deleteTopicEnable=true bitnami/kafka
bash 
helm install mongodb --set metrics.enabled=true bitnami/mongodb --set global.namespaceOverride=default --set auth.rootUser=root --set auth.rootPassword=splunk --set auth.enabled=false --version 12.1.31

Verify the helm chart installation

bash 
helm list

Verify the helm chart installation

bash 
kubectl get pods

Use information for each Helm chart and Splunk O11y Data Setup to generate values.yaml for capturing metrics from Kafka and MongoDB.

Note

values.yaml for the different services will be passed to the Splunk Helm Chart at installation time. These will configure the OTEL collector to capture metrics from these services.

4.1 Example kafka.values.yaml

yaml 
otelAgent:
config:
    receivers:
    receiver_creator:
        receivers:
        smartagent/kafka:
            rule: type == "pod" && name matches "kafka"
            config:
                    #endpoint: '`endpoint`:5555'
            port: 5555
            type: collectd/kafka
            clusterName: sl-kafka
otelK8sClusterReceiver:
k8sEventsEnabled: true
config:
    receivers:
    kafkametrics:
        brokers: kafka:9092
        protocol_version: 2.0.0
        scrapers:
        - brokers
        - topics
        - consumers
    service:
    pipelines:
        metrics:
        receivers:
                #- prometheus
        - k8s_cluster
        - kafkametrics

4.2 Example mongodb.values.yaml

yaml 
otelAgent:
    config:
        receivers:
        receiver_creator:
            receivers:
            smartagent/mongodb:
                rule: type == "pod" && name matches "mongo"
                config:
                type: collectd/mongodb
                host: mongodb.default.svc.cluster.local
                port: 27017
                databases: ["admin", "O11y", "local", "config"]
                sendCollectionMetrics: true
                sendCollectionTopMetrics: true

4.3 Example zookeeper.values.yaml

yaml 
otelAgent:
    config:
        receivers:
        receiver_creator:
            receivers:
            smartagent/zookeeper:
                rule: type == "pod" && name matches "kafka-zookeeper"
                config:
                type: collectd/zookeeper
                host: kafka-zookeeper
                port: 2181

5. Install the Splunk OTEL helm chart

bash 
cd /home/splunk/realtime_enrichment/otel_yamls/ 

helm repo add splunk-otel-collector-chart https://signalfx.github.io/splunk-otel-collector-chart

helm repo update
bash 
helm install --set provider=' ' --set distro=' ' --set splunkObservability.accessToken=$ACCESS_TOKEN --set clusterName=$clusterName --set splunkObservability.realm=$REALM --set otelCollector.enabled='false' --set gateway.enabled='false' --values kafka.values.yaml --values mongodb.values.yaml --values zookeeper.values.yaml --values alwayson.values.yaml --values k3slogs.yaml --generate-name splunk-otel-collector-chart/splunk-otel-collector

6. Verify installation

Verify that the Kafka, MongoDB and Splunk OTEL Collector helm charts are installed, note that names may differ.

bash 
helm list
bash 
kubectl get pods

7. Verify dashboards

Verify that out of the box dashboards for Kafka, MongoDB and Zookeeper are populated in the Infrastructure Monitor landing page. Drill down into each component to view granular details for each service.

Tip: You can use the filter k8s.cluster.name with your cluster name to find your instance.

IM-landing-page

k8-navigator

mongodb-dash

kafka-dash

Last Modified ·