Monitoring

Self-Hosting Documentation Access

This section requires a password to access. Interested in self-hosting? Contact sales to learn more.

Password:

Access Documentation

Self-Hosting Documentation Access Granted Logout

LlamaCloud services expose metrics, which are collected by Prometheus and visualized in Grafana.

Prerequisites

To monitor your LlamaCloud deployment, you’ll need:

• Prometheus - For metrics collection and storage
• Grafana - For metrics visualization
• AlertManager - For alert management

These services can be deployed using the kube-prometheus-stack Helm chart, which provides a complete monitoring solution that includes:

• Prometheus server for metrics collection
• Grafana for visualization with pre-configured dashboards
• AlertManager for handling alerts
• Node exporter for hardware and OS metrics
• kube-state-metrics for Kubernetes object metrics
• Prometheus Operator for managing Prometheus instances

Use this file as a starting point for a basic installation:

kube-prometheus-stack.yaml

prometheus:
  enabled: true
grafana:
  enabled: true
alertmanager:
  enabled: true

helm repo add prometheus-community https://prometheus-community.github.io/helm-charts
helm repo update
helm install kube-prometheus-stack prometheus-community/kube-prometheus-stack \
    -f kube-prometheus-stack.yaml

For more information about the kube-prometheus-stack Helm chart, please refer to the kube-prometheus-stack README.

Service Monitoring

After installing the kube-prometheus-stack Helm chart, you should create ServiceMonitor objects. These objects tell Prometheus to scrape the metric endpoints exposed by the LlamaCloud services.

LlamaCloud

apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  # 👇 The name of the service you want to monitor.
  name: llamacloud
  # 👇 The namespace where the ServiceMonitor object lives; typically the namespace where you installed the chart.
  namespace: <kube-prometheus-stack-namespace>
  labels:
    # 👇 Typically this is the same namespace as above.
    release: kube-prometheus-stack
spec:
  namespaceSelector:
    matchNames:
      # 👇 The namespace you installed Llamacloud into.
      - <llamacloud-namespace>
  selector:
    matchLabels:
      app.kubernetes.io/instance: llamacloud
      app.kubernetes.io/name: llamacloud
      app.kubernetes.io/managed-by: Helm
  endpoints:
    - port: http
      path: /metrics
      interval: 30s
      scrapeTimeout: 10s

JobsService

apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  # 👇 The name of the service you want to monitor.
  name: llamacloud-operator
  # 👇 The namespace where the ServiceMonitor object lives; typically the namespace where you installed the chart.
  namespace: <kube-prometheus-stack-namespace>
  labels:
    # 👇 Typically this is the same namespace as above.
    release: kube-prometheus-stack
spec:
  namespaceSelector:
    matchNames:
      # 👇 The namespace you installed Llamacloud into.
      - <llamacloud-namespace>
  selector:
    matchLabels:
      app.kubernetes.io/instance: llamacloud
      app.kubernetes.io/name: llamacloud-operator
      app.kubernetes.io/managed-by: Helm
  endpoints:
    - port: http
      path: /metrics
      interval: 30s
      scrapeTimeout: 10s

JobsWorker

apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  # 👇 The name of the service you want to monitor.
  name: llamacloud-worker
  # 👇 The namespace where the ServiceMonitor object lives; typically the namespace where you installed the chart.
  namespace: <kube-prometheus-stack-namespace>
  labels:
    # 👇 Typically this is the same namespace as above.
    release: kube-prometheus-stack
spec:
  namespaceSelector:
    matchNames:
      # 👇 The namespace you installed Llamacloud into.
      - <llamacloud-namespace>
  selector:
    matchLabels:
      app.kubernetes.io/instance: llamacloud
      app.kubernetes.io/name: llamacloud-worker
      app.kubernetes.io/managed-by: Helm
  endpoints:
    - port: http
      path: /metrics
      interval: 30s
      scrapeTimeout: 10s

LlamaParse

apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  # 👇 The name of the service you want to monitor.
  name: llamacloud-parse
  # 👇 The namespace where the ServiceMonitor object lives; typically the namespace where you installed the chart.
  namespace: <kube-prometheus-stack-namespace>
  labels:
    # 👇 Typically this is the same namespace as above.
    release: kube-prometheus-stack
spec:
  namespaceSelector:
    matchNames:
     # 👇 The namespace you installed Llamacloud into.
      - <llamacloud-namespace>
  selector:
    matchLabels:
      app.kubernetes.io/instance: llamacloud
      app.kubernetes.io/name: llamacloud-parse
      app.kubernetes.io/managed-by: Helm
  endpoints:
    - port: http
      path: /metrics
      interval: 30s
      scrapeTimeout: 10s

LlamaParseOcr

apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  # 👇 The name of the service you want to monitor.
  name: llamacloud-ocr
  # 👇 The namespace where the ServiceMonitor object lives; typically the namespace where you installed the chart.
  namespace: <kube-prometheus-stack-namespace>
  labels:
    # 👇 Typically this is the same namespace as above.
    release: kube-prometheus-stack
spec:
  namespaceSelector:
    matchNames:
      # 👇 The namespace you installed Llamacloud into.
      - <llamacloud-namespace>
  selector:
    matchLabels:
      app.kubernetes.io/instance: llamacloud
      app.kubernetes.io/name: llamacloud-ocr
      app.kubernetes.io/managed-by: Helm
  endpoints:
    - port: http
      path: /metrics
      interval: 30s
      scrapeTimeout: 10s

Grafana

Once you have installed the kube-prometheus-stack and installed the ServiceMonitor objects into your Kubernetes cluster, you will be able to view the metrics in Grafana.

First, get the Grafana admin user’s password:

kubectl --namespace <llamacloud-namespace> get secrets kube-prometheus-stack-grafana -o jsonpath="{.data.admin-password}" | base64 -d ; echo

Next, create a port-forward to Grafana:

export GRAFANA_POD_NAME=$(kubectl --namespace <llamacloud-namespace> \
    get pod -l "app.kubernetes.io/name=grafana,app.kubernetes.io/instance=kube-prometheus-stack" -oname)
kubectl --namespace <llamacloud-namespace>  port-forward $GRAFANA_POD_NAME 3000

Now open your browswer to localhost:3000 and log into the Grafana console. You should be able to explore all the LlamaCloud metrics in the Explore tab.