Skip to content

Latest commit

 

History

History
117 lines (96 loc) · 6.8 KB

prometheus_setup.md

File metadata and controls

117 lines (96 loc) · 6.8 KB

Task: Setup Prometheus monitoring

We are going to setup Prometheus <-> ClickHouse-operator integration in k8s environment. This document assumes k8s cluster already setup and kubectl has access to it.

We may have two starting points:

  1. Prometheus installation is available and we just need to gather clickhouse-operator metrics
  2. Need to install Prometheus at first and integrate it with clickhouse-operator afterwards

Prometheus already available

In case we have Prometheus already setup, what we need is to specify clickhouse-operator's HTTP endpoint where Prometheus can gather metrics from clickhouse-operator. Endpoint is a Service of type ClusterIP named as clickhouse-operator-metrics It is located in the same namespace as clickhouse-operator:

kubectl get service clickhouse-operator-metrics -n kube-system
NAME                          TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)    AGE
clickhouse-operator-metrics   ClusterIP   10.105.15.167   <none>        8888/TCP   95m

For debug purposes, we can port-forward it into our local OS as

kubectl --namespace=kube-system port-forward service/clickhouse-operator-metrics 8888

and access raw data with browser by navigating to http://localhost:8888/metrics

So, what we need in general, is to point Prometheus to gather data from:

http://<service/clickhouse-operator-metrics>:8888/metrics

Setup Prometheus and integrate it with clickhouse-operator

In case we do not have Prometheus available, we can setup it directly into k8s and integrate with clickhouse-operator

We are going to setup and manage Prometheus with Prometheus Operator

We already have manifests available. We can either run create-prometheus.sh or setup the whole process by hands, in case we need to edit configuration.

  • We'd like to setup all Prometheus-related components into dedicated namespace. Let's create it
kubectl create namespace prometheus
  • Create CRD for prometheus-operator like kind:Prometheus, kind:PrometheusRule etc.
kubectl --namespace=prometheus apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/master/example/prometheus-operator-crd/monitoring.coreos.com_prometheuses.yaml
kubectl --namespace=prometheus apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/master/example/prometheus-operator-crd/monitoring.coreos.com_prometheusrules.yaml
kubectl --namespace=prometheus apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/master/example/prometheus-operator-crd/monitoring.coreos.com_alertmanagers.yaml
kubectl --namespace=prometheus apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/master/example/prometheus-operator-crd/monitoring.coreos.com_podmonitors.yaml
kubectl --namespace=prometheus apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/master/example/prometheus-operator-crd/monitoring.coreos.com_servicemonitors.yaml
kubectl --namespace=prometheus apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/master/example/prometheus-operator-crd/monitoring.coreos.com_thanosrulers.yaml
  • Create RBAC settings for prometheus-operator
kubectl --namespace=prometheus apply --server-side -f <( \
  wget -qO- https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/master/example/rbac/prometheus/prometheus-cluster-role-binding.yaml | \
  sed "s/namespace: default/namespace: prometheus/" \
)
kubectl --namespace=prometheus apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/master/example/rbac/prometheus/prometheus-cluster-role.yaml
kubectl --namespace=prometheus apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/master/example/rbac/prometheus/prometheus-service-account.yaml
  • Setup prometheus-operator into dedicated namespace
  kubectl --namespace=prometheus apply --server-side -f  <( \
      wget -qO- https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/master/bundle.yaml | \
      sed "s/namespace: default/namespace: prometheus/" \
  )
  • Setup prometheus into dedicated namespace. prometheus-operator would be used to create prometheus instance
kubectl apply --namespace=prometheus --server-side -f <(wget -qO- https://raw.githubusercontent.com/Altinity/clickhouse-operator/master/deploy/prometheus/prometheus-template.yaml | PROMETHEUS_NAMESPACE=prometheus envsubst)
export SLACK_WEBHOOK_URL=https://hooks.slack.com/services/XXXX/YYYYY/ZZZZZ
export SLACK_CHANNEL="#alerts-channel-name"
export PROMETHEUS_NAMESPACE=prometheus
export ALERT_MANAGER_EXTERNAL_URL=https://your.external-domain.for-alertmanger/    
kubectl apply --namespace=prometheus -f <( \
  wget -qO- https://raw.githubusercontent.com/Altinity/clickhouse-operator/master/deploy/prometheus/prometheus-alertmanager-template.yaml | \
  envsubst \
)
  • Setup clickhouse-operator alert rules for prometheus
kubectl apply --namespace=prometheus -f https://raw.githubusercontent.com/Altinity/clickhouse-operator/master/deploy/prometheus/prometheus-alert-rules.yaml

At this point Prometheus and AlertManager is up and running. Also, all kubernetes pods which contains meta.annotations prometheus.io/scrape: 'true', promethus.io/port: NNNN, prometheus.io/path: '/metrics', prometheus.io/scheme: http, will discover via prometheus kubernetes_sd_config job kubernetes-pods.

Now we should have Prometheus gathering metrics from clickhouse-operator. Let's check it out. To get access to Prometheus. Port-forward Prometheus to localhost as:

kubectl --namespace=prometheus port-forward service/prometheus 9090

and navigate browser to http://localhost:9090 Prometheus should appear.

To get access to Alert Manager. Port-forward Prometheus to localhost as:

kubectl --namespace=prometheus port-forward service/alertmanager 9093

and navigate browser to http://localhost:9093/alerts Alert Manager should appear.

We can check whether clickhouse-operator is available at http://localhost:9090/targets

More Prometheus docs