BeamMonitoring

This folder contains resources required to deploy the Beam metrics stack.

There are two types of metrics in Beam:

• Community metrics. The stack includes:

  • Python scripts for ingesting data from sources (Jenkins and GitHub)
  • Postgres analytics database

• Test Results, i.e. metrics published by tests (IO Performance tests, Load tests and Nexmark tests). Beam uses InfluxDB time series database to store test metrics.

Both types of metrics are presented in Grafana dashboard available here.

Both stacks can be deployed on your local machine.

All components run within Docker containers. These are composed together via docker-compose for local hosting, and Kubernetes for the production instance on GCP.

Local setup

Docker Compose is used to host the full metrics stack on your local machine.

• install docker
  • https://docs.docker.com/install/#supported-platforms
• install docker-compose
  • https://docs.docker.com/compose/install/#install-compose

# Remove old docker
sudo apt-get remove docker docker-engine docker.io

# Install docker
sudo apt-get update
sudo apt-get install \
     apt-transport-https \
     ca-certificates \
     curl \
     gnupg2 \
     software-properties-common
curl -fsSL https://download.docker.com/linux/debian/gpg | sudo apt-key add -
sudo apt-key fingerprint 0EBFCD88
sudo add-apt-repository \
   "deb [arch=amd64] https://download.docker.com/linux/debian \
   $(lsb_release -cs) \
   stable"
sudo apt-get update
sudo apt-get install docker-ce

###################################################
# Commands below here required to spin up docker  #
# containers locally. Can be omitted for kuberctl #
# setup.                                          #
###################################################

# Install docker-compose
sudo curl -L https://github.com/docker/compose/releases/download/1.22.0/docker-compose-$(uname -s)-$(uname -m) -o /usr/local/bin/docker-compose
sudo chmod +x /usr/local/bin/docker-compose


# start docker service if it is not running already
sudo service docker start

# Build images required for spinning up docker-compose
# related containers.
docker-compose build

# Spinup docker-compose related containers.
docker-compose up

After running these commands, you can access the services running on your local machine:

• Grafana: http://localhost:3000
• Postgres DB: localhost:5432
• InfluxDB: http://localhost:8086

If you're deploying for the first time on your machine, follow the wiki instructions on how to manually configure Grafana.

Grafana, Postgres and InfluxDB containers persist data to Docker volumes, which will be restored on subsequent runs. To start from a clean state, you must also wipe out these volumes. (List volumes via docker volume ls)

Kubernetes setup

Cluster Specification

• Name: metrics
• GKE Version: 1.22.6-gke.300
• Node Pool Version: 1.22.6-gke.300
• Runtime: Container-Optimized OS with containerd (cos_containerd)
• Authentication method: OAuth Client Certificate.

Kubernetes deployment instructions are maintained in the wiki:

• Community metrics
• Test Results Monitoring

Note: Basic Auth is not supported on BEAM Clusters as of March 4th 2022

Prior to v1.19 GKE allowed to log into a cluster using basic authentication methods, which relies on a username and password, and now it is deprecated.

Currently, OAuth is the standard authentication method, previous usernames and passwords were removed so only the certificate remains as master auth.

Some Beam Performance tests need to be logged into the cluster so they can create its own resources, the correct method to do so is by setting up the right Kubeconfig inside the worker and execute get credentials within GCP.

In the future if you need to create an automatic process that need to have access to the cluster, use OAuth inside your script or job.