ElasticSearch

Install BotKube Backend in Kubernetes cluster

Using helm

• We will be using helm to install BotKube in Kubernetes. Follow this guide to install helm if you don’t have it installed already
• Add infracloudio chart repository

  $ helm repo add infracloudio https://infracloudio.github.io/charts
  $ helm repo update

• Deploy BotKube backend using helm install in your cluster.

  $ helm install --version v0.12.4 botkube --namespace botkube \
  --set communications.elasticsearch.enabled=true \
  --set communications.elasticsearch.server=<ELASTICSEARCH_ADDRESS> \
  --set communications.elasticsearch.username=<ELASTICSEARCH_USERNAME> \
  --set communications.elasticsearch.password=<ELASTICSEARCH_PASSWORD> \
  --set communications.elasticsearch.index.name=<ELASTICSEARCH_INDEX_NAME> \
  --set communications.elasticsearch.index.type=<ELASTICSEARCH_INDEX_TYPE> \
  --set communications.elasticsearch.index.shards=<ELASTICSEARCH_INDEX_SHARDS> \
  --set communications.elasticsearch.index.replicas=<ELASTICSEARCH_INDEX_REPLICAS> \
  --set config.settings.clustername=<CLUSTER_NAME> \
  --set image.repository=infracloudio/botkube \
  --set image.tag=v0.12.4 \
  infracloudio/botkube

where,
- ELASTICSEARCH_ADDRESS is an address on which ElasticSearch server is reachable e.g https://example.com:9243
- ELASTICSEARCH_USERNAME is the username for authentication to Els server
- ELASTICSEARCH_PASSWORD is a password for the username to authenticate with Els server
- ELASTICSEARCH_INDEX_NAME (optional) is an index name on which BotKube events will be stored (default: botkube)
- ELASTICSEARCH_INDEX_TYPE (optional) contains type for the BotKube index (default: botkube-event)
- ELASTICSEARCH_INDEX_SHARDS (optional) denotes number of shards for BotKube index (default: 1)
- ELASTICSEARCH_INDEX_REPLICAS (optional) is the number of replicas of the shards (default: 0)
- CLUSTER_NAME is the cluster name set in the incoming messages

Configuration syntax is explained here. Complete list of helm options is documented here.

• Create new file config.yaml and add resource configuration as described on the configuration page.

  (You can refer sample config from https://raw.githubusercontent.com/infracloudio/botkube/v0.12.4/helm/botkube/sample-res-config.yaml)

  config:
    ## Resources you want to watch
    resources:
    - name: v1/pods        # Name of the resource. Resource name must be in 
                           # group/version/resource (G/V/R) format
                           # resource name should be plural
                           # (e.g apps/v1/deployments, v1/pods)
      namespaces:          # List of namespaces, "all" will watch all the namespaces
        include:
        - all
        ignore:            # List of namespaces to be ignored, used only with include: all
        - kube-system      # example : include [all], ignore [x,y,z]
      events:              # List of lifecycle events you want to receive,
                           # e.g create, update, delete, error OR all
      - create
      - delete
      - error
    - name: batch/v1/jobs
      namespaces:
        include:
        - ns1
        - ns2
      events:
      - create
      - update
      - delete
      - error
      updateSetting:
        includeDiff: true
        fields:
        - spec.template.spec.containers[*].image
        - status.conditions[*].type

• Pass the yaml file as a flag to helm install command. e.g

  $ helm install --version v0.12.4 --name botkube --namespace botkube -f /path/to/config.yaml --set=...other args..
  

Alternatively, you can also update the configuration at runtime as documented here

Using kubectl

• Make sure that you have kubectl cli installed and have access to Kubernetes cluster
• Download deployment specs yaml

$ wget -q https://raw.githubusercontent.com/infracloudio/botkube/v0.12.4/deploy-all-in-one.yaml

• Open downloaded deploy-all-in-one.yaml and update the configuration.
  

Set ELASTICSEARCH_ENABLED=true, ELASTICSEARCH_USERNAME, ELASTICSEARCH_PASSWORD, clustername, index settings and update the resource events configuration you want to receive notifications for in the configmap.

where,
- ELASTICSEARCH_ADDRESS is an address on which ElasticSearch server is reachable e.g https://example.com:9243
- ELASTICSEARCH_USERNAME is the username for authentication to Els server
- ELASTICSEARCH_PASSWORD is a password for the username to authenticate with Els server

• Create botkube namespace and deploy resources

$ kubectl create -f deploy-all-in-one.yaml

• Check pod status in botkube namespace.

Remove BotKube

Using helm

If you have installed BotKube backend using helm, execute following command to completely remove BotKube and related resources from your cluster

$ helm delete --purge botkube

Using kubectl

$ kubectl delete -f https://raw.githubusercontent.com/infracloudio/botkube/v0.12.4/deploy-all-in-one.yaml