Skip to Content
DocsQuick StartPost Install

Post Install Setup Guide

The base install is Workbench, Server, Runtime, Notebooks, and optional Grafana. Scoring and learning happen on the Runtime, not Grafana.

After login, open the Runtime Dashboard (Runtime 8091, or 8099 for local debug). Invoke a campaign, then Accept an offer (POST /response). Hosted demo: https://medemo.ecosystem.ai .

Base Install

  • Workbench: Operator UI for projects and deployments.
  • Server: Modeling and push to Runtime.
  • Runtime: POST /invocations and POST /response (params is a JSON string).
  • Notebooks: Processing / generative workers.
  • Grafana: Optional infra charts on port 3000 — not the Runtime tester UI.

Step 1. Log in to the App Workbench

The default username and password is admin@ecosystem.ai and password. Please change the password immediately after install.

Step 2. Log in to Notebooks

You need to create a new user account for the Notebooks. The default username is admin@ecosystem.ai. You can assign any password during the registration process.

Step 3. Setup key and login to Grafana

The default username and password is admin@ecosystem.ai and password. Please change the password immediately after install.

to access Grafana, navigate to:

http://127.0.0.1:3000

OR

http://localhost:3000

You need to generate a key in your Workbench in order to connect to the ecosystem.Ai server for viewing your data. In the Workbench navigate to the Management menu and click on Generate API Keys. Click on the Create New button, assign a name and description and time duration for valid validation. Copy the key. You can use this key to connect to the ecosystem.Ai server inside Grafana.

In Grafana navigate to the Connections menu and click on Data Sources. Click on ecosystem-server and scroll down to HTTP headers. Click on Reset and the field to the left will open up. Type the word ```Bearer ` and then paste your key after the space:

Bearer <your key>

Save & Test might produce an error, but you can ignore this. You can now create dashboards and panels in Grafana. Navigate to the Dashboards.

Step 4. Score a campaign on the Runtime

Grafana is not the scoring UI. Open the Runtime Dashboard against port 8091 (or 8099 for local debug):

  1. Invoke an ACTIVE campaign (POST /invocations; params is a JSON string).
  2. Accept a ranked offer (POST /response with that row’s uuid and offer_name).
  3. Optional: MCP → Ontology Explorer; Testing → campaign BDD.

Demo host: https://medemo.ecosystem.ai .

Additional Components Install

The additional components package can be obtained from the ecosystem.Ai marketplace. The package contains the following components:

  • Presto: A distributed SQL query engine for running queries on large datasets.
  • Superset: A data visualization tool for creating dashboards.
  • Postgres: A database for storing relational data.

Login to Superset

The default username and password is admin@ecosystem.ai and password. Please change the password immediately after install.

Last updated on
OpenShiftIntro