Goal

This walk-through guides you to setup an in-cluster Elasticsearch and Kibana suite, with cluster-level logging data gathered by Fluent Bit. You may access and search logs from every pod in cluster, as long as the workload in pod writes log to stdout or stderr.

Prerequisites

• Kubernetes 1.11 or higher (minikube not working)
• Predefined storage class called hdd-ssd (you may change it in eck.yaml)

Deployment Steps

Clone this repo to get necessary yaml files.

Elasticsearch and Kibana

Elastic Cloud on Kubernetes(ECK) is now generally available. ECK makes it easy to deploy Elasticsearch and Kibana on k8s with various topology.

• Deploy ECK

1

kubectl apply -f https://download.elastic.co/downloads/eck/1.0.1/all-in-one.yaml

• Create Namespace logging

1

kubectl create -f ./namespace.yml

• Deploy Elasticsearch and Kibana

1

kubectl create -f ./eck.yml

Fluent Bit

FluentBit runs as DaemonSet on every node in cluster, gathering logs from every workload. FluentBit attach metadata like pod name and label to logs delivered to Elasticsearch.

Well-structured log(in JSON) can be searched/filtered by term in Elasticsearch.

1
2
3
4
5

kubectl create -f fluent-bit-service-account.yaml
kubectl create -f fluent-bit-role.yaml
kubectl create -f fluent-bit-role-binding.yaml
kubectl create -f fluent-bit-configmap.yaml
kubectl create -f fluent-bit-ds.yaml

And off you go.

Reference

• Elastic Cloud on Kubernetes
• Kubernetes Logging with Fluent Bit
• Fluent Bit Manual
• Enabling Native Realms on ECK
• Fluent Bit: Elasticsearch output should probably not use a type (flb_type)