Skip to content

analytics.mdx

Latest commit

 

History

History
101 lines (68 loc) · 6.35 KB

analytics.mdx

File metadata and controls

101 lines (68 loc) · 6.35 KB
title Analytics
description How to create and use charts and dashboards in OpenOps Analytics
icon chart-line

import JoinCommunity from '/snippets/join-community.mdx' import { NarrowImage } from '/snippets/narrow-image.jsx'

You can visualize data that you collect with OpenOps tables and display the visualizations in the Analytics view in OpenOps.

In addition to OpenOps tables, OpenOps Analytics can connect to and visualize your existing data sources, such as CSV files, Google Sheets, Postgres databases, and more.

The analytics service is optional. You can disable it by setting `OPS_ANALYTICS_ENABLED` variable to `false` in your `.env` file if you don't need analytics functionality.

Analytics admin portal

To configure data visualizations, click Analytics on the main navigation bar in OpenOps. When the Analytics view opens, click Admin Panel to proceed to the Analytics admin portal that is separate from the main OpenOps application.

To sign in to the Analytics portal, use admin as the username and the password defined with the OPS_ANALYTICS_ADMIN_PASSWORD variable in the .env file in your OpenOps installation folder. (If you haven’t changed the password, the default is please-change-this-password-1.)

If you don't have access to the OpenOps installation folder: * Contact your OpenOps instance administrator if you're using a self-managed OpenOps instance. * [Contact OpenOps support](mailto:support@openops.com) if you're using an OpenOps-hosted instance.

Once you log in, you can see a built-in dashboard called FinOps. Modifying this dashboard will affect the Analytics tab in OpenOps.

Built-in dashboards in the analytics portal

Behind every OpenOps table, there is a Postgres database view that can be used to build charts and queries.

When a new table is created in OpenOps tables, the system automatically creates a new database view with the naming convention table name _ table id _ userfriendly. You can use this view to create dashboards in OpenOps, as well as connect it to other BI systems.

Building new charts

To configure a new chart based on an OpenOps table to display in the Analytics view in OpenOps, do the following:

  1. Navigate to Datasets and click + Dataset: Creating a dataset
  2. Select openops_tables_connection for Database and public for Schema. In the Table dropdown, select a Postgres DB view corresponding to the OpenOps table you want to source data from. For example, here is what a view may be named for the Opportunities table:
  3. Click Create dataset and create chart.
  4. In the Create a new chart view, select a chart type, then click Create new chart: Selecting a chart type
  5. Configure your chart. Refer to this Superset documentation page if you need help. To preview the configured chart, click Create chart in the right pane. Chart preview
  6. If you make further changes to the chart, click Update chart to refresh the preview.
  7. When you're done, click Save. In the Save chart dialog, select FinOps in the Add to dashboard dropdown:
  8. Click Save. When you go back to the Analytics view in OpenOps, you can see that the chart is now displayed there: Chart in OpenOps Analytics

For more information on how to create charts, see the Superset documentation.

Creating virtual (SQL) datasets

In the Analytics admin portal, a dataset can be created from:

  • A physical table or view (select it from the Table dropdown).
  • A virtual dataset, defined by a SQL query.

Create a virtual dataset in the Analytics admin portal by using the portal’s dataset creation flow and providing a SQL query.

Tips and best practices

  • Enter a non-empty SQL query. Creating a virtual dataset requires SQL that is not empty or whitespace.

Troubleshooting

  • Dataset creation fails with an “SQL query cannot be empty or whitespace” error: Edit the dataset and provide a SQL query with at least one non-whitespace character.

When a dataset already exists

When creating datasets programmatically (for example, via internal services that call the analytics API), dataset creation can be configured to either:

  • Reuse an existing dataset when the dataset already exists.
  • Recreate an existing dataset by deleting it first and then creating a new one (a delete-and-recreate flow).

Connecting to external data sources

Here's what you can do to connect OpenOps Analytics to external data sources such as CSV files:

  1. In the Analytics admin portal, in the top right corner, click Settings, then Database Connections.
  2. Click + Database to connect to an external data source, such as your own Postgres instance or Google Sheets.
  3. Once you have created a database connection, if you’d like to upload CSVs to it, click the Edit icon for the database connection: Editing a database connection
  4. In the popup that displays, go to Advanced > Security, scroll down, and enable Allow file uploads to database.
  5. Click Finish.

To upload CSV files into your database, click + in the top right corner, then select Data > Upload CSV to database:

You can now build charts based on the data from your external data source, similar to how you do it with OpenOps tables. When you add charts to the FinOps dashboard, they become available in the Analytics view in OpenOps.