Install Datalore Enterprise using Docker or Kubernetes (Helm)

The instructions in this article describe both installation methods: Docker and Kubernetes. We recommend that you choose the Docker-based installation if your primary goal is to get familiar with the product and its key features. This is the easier and quicker procedure of the two.

Prerequisites

Before installation, make sure that you have the following:

Docker
Docker Compose V2

To check the requirements, execute docker compose version.

Expected result:

docker compose version
Docker Compose version v2.2.3

K8s cluster (tested with v1.22 version of Kubernetes)
Kubectl on your machine pointed to this cluster
Helm (tested with the v3.2.4 version)

Hardware requirements

Datalore server machine: 4GB of RAM (the number of CPU is irrelevant if the load is not high)
For every additional computational machine (pod or container) per notebook: minimum 2GB of RAM required, 4GB of RAM recommended

Basic Datalore installation

Follow the instruction to install Datalore using the selected method.

Install Datalore

Depending on the installation type, do one of the following:
- (For Docker) Clone or download the content of this repository.
- (For Helm) Add the Datalore helm repository:
  $ helm repo add datalore https://jetbrains.github.io/datalore-configs/charts
Do the following to set up your database:
Open docker-compose.yaml in the [repository_folder]/docker-compose folder in any text editor and replace the values of DB_PASSWORD and POSTGRES_PASSWORD properties with any random string (both properties must have the same value). This string will be used as your database password. Make sure you keep it secret.
Create a datalore.values.yaml file and add there your volume configuration and dbRootPassword (to be used as your database password, random string advised).
For example:
volumes: - name: storage emptyDir: { } - name: postgresql-data emptyDir: { } dbRootPassword: "super_secret_password"
This configuration uses emptyDir volumes. This means that after each Datalore restart, all data is wiped. Replace it by available in your cluster persistent volumes.
Run the following command and wait for Datalore to start up:
docker compose up
helm install -f datalore.values.yaml <release_name> datalore/datalore --version 0.2.4
<release_name> can be anything, for example, datalore.
You can run kubectl port-forward svc/datalore 8080 to test if Datalore can start up. However, to make it accessible, make sure you configure ingress.
Go to http://127.0.0.1:8080/ and create the first user. The first created user will automatically receive admin rights.
Click your avatar in the upper-right corner, select Admin panel and provide your license key.

Email, plan, and gift code options

This chapter explains how to configure email, plan, and gift code options for both types of installation.

Enable emails

By default, emails are not enabled and email verification for users is disabled. To enable the email service, specify the following parameters in the docker-compose file (for Docker) or values under the dataloreEnv key in the datalore.values.yaml file (for Helm):

MAIL_SENDER_EMAIL: sender's email
MAIL_SENDER_NAME: sender's name
MAIL_SENDER_USERNAME: username of SMTP user
MAIL_SENDER_PASSWORD password of SMTP user
MAIL_SMTP_SERVER: SMTP server host
MAIL_SMTP_PORT: SMTP server port

If you want to disable email verification while having enabled emails, you can explicitly set the FORCE_EMAIL_VERIFICATION property to false.

Example (Docker)

services:
  datalore:
    ...
    environment:
      ...
      MAIL_SMTP_SERVER: "email-smtp.your_domain.com"
      MAIL_SMTP_PORT: "465"
      MAIL_SENDER_USERNAME: "email_user"
      MAIL_SENDER_PASSWORD: "pa$$w0rd"
      MAIL_SENDER_EMAIL: "no_reply.datalore@you_domain.com"
      MAIL_SENDER_NAME: "Datalore Team"
      FORCE_EMAIL_VERIFICATION: "false"

Enable plans

Use the plans feature to limit resources that are available for users. To enable the feature, do the following:

For Docker, mount the configuration file into the Datalore container and put the path to it into the DATALORE_PLANS_CONFIGURATION variable.
services: datalore: ... environment: ... DATALORE_PLANS_CONFIGURATION: "file:/opt/datalore/configs/plans.yaml"
For Helm, specify plan parameters under the plansConfig key in values.yaml.

Parameters for plan configurations are the same for both installation types. Each plan has a unique planId. Make sure that one of them is marked as default. When adding or configuring a plan, specify the following parameters:

Option	Value example	Description
diskUsageLimit	10 Gb	Disk space in persistence storage allowed per user
numRunningInstancesLimit	10	Number of parallel running agents allowed per user
parallelInstancesQuota	PT100500H	Deprecated
instanceDurationQuotaMap	basic-agent: "PT100H" large-agent: "PT10H"	Computation time per month available for each agent type; more than one agent type allowed, details in the tip below

Example (Docker):

                        - planId: "Student"
  default: true
  name: "Student plan"
  instanceDurationQuotaMap:
    docker-base: "PT100H"
    docker-swarm: "PT100H"
    teamcity-basic: "PT100500H"
  diskUsageLimit: "5.00 GB"
  numRunningInstancesLimit: 3
  parallelInstancesQuota: "PT120H"
- planId: "Teacher"
  name: "Teacher plan"
  instanceDurationQuotaMap:
    docker-base: "PT100H"
    docker-swarm: "PT50H"
  diskUsageLimit: "25.00 GB"
  numRunningInstancesLimit: 10

Enable gift codes

Turn on the option by setting GIFT_CODES_ENABLED to true in the docker-compose file (for Docker) or values under the dataloreEnv key in the datalore (for Helm).

Example (Docker):

services:
  datalore:
    ...
    environment:
      ...
      GIFT_CODES_ENABLED: "true"

Refer to Admin user actions to find information on how to use gift codes.

Last modified: 24 October 2022