Go-Live Preparation

This document assumes that you have set up SPARQUE. You have configured your workspace in a way that fits your project. The test data is connected, the strategies and APIs work to your satisfaction.

Now, the next step is to prepare the SPARQUE production instance. This is a one-time process.

Once the production instance exists, you connect the APIs to your live shop.

Preconditions

  • A SPARQUE workspace exists.
  • You have set up and/or adapted all data pipelines, strategies and APIs to your project requirements (with test data).

Process Steps

For go-live preparation, follow the steps below:

Step 1: Request New Preproduction Configuration (Optional Step)

The production instance will be a clone of an existing configuration. Since the default configuration usually uses test data, you can get an additional configuration where you connect production data.

Exception: In case of projects where the amount of data is relatively small, you can also use production data in the default configuration. In this case, the production instance can be a clone of the default configuration, and this step here is obsolete.

For further information on configurations, see Configurations.

Create a service ticket: "Create preproduction configuration for SPARQUE workspace <myworkspacename>".

Step 2: Make One Configuration Production-Ready and Test

The production instance will be a clone of an existing configuration, so this configuration must reflect the exact target state. The production instance cannot be modified, i.e., the file selection cannot be changed later on. Therefore, ensure that the connected data is correct.

If you requested a preproduction configuration, perform the following steps there. Else, continue in the default configuration.

Data:

  1. Upload all production data to Azure Blob Storage creating a folder for each dataset. Place the files in the folders.
  2. Switch all datasets and pipelines to Fetch from URL and link them to blob objects/files.
  3. Load/index all datasets and ensure they are working with the blob data.
  4. Create a list of all datasets (by name).*

APIs:

  1. Update all APIs with the new data and verify that they are working as expected. If you use the API Wrapper, please test also with the Wrapper APIs to validate that your workspace is still compatible. See Setup for Using the SPARQUE API Wrapper for details.
  2. Special case: The health_check API is intended as a simple validation that checks if your workspace is healthy. Please adapt this API according to your requirements. Keep the check simple. Please note that the result is not evaluated, only the HTTP status code and the response time are relevant for the health check. For more details, see Health Check.
  3. Create a list of all APIs in SPARQUE Desk by name and indicate whether authentication is enabled or disabled.*
    Find more information on authentication here: Authentication Guide for Server Applications.

The information marked with * is needed in the next step.

Step 3: Request Creation of Production Instance

Create a service ticket: "Create production instance for SPARQUE workspace <myworkspacename>".

Provide the following information:

  • Workspace name
  • Configuration to be cloned (default or preproduction)
  • Desired time (UTC) for daily index rebuild (you need to ensure that the new data is available in the storage account before this time)
  • CORS configuration, if needed
  • List of all datasets (by name)
  • List of all APIs (by name) and if authentication is enabled/disabled

Step 4: Use the Production APIs in Your Production PWA

Once the production instance exists, you can use the production APIs.

If you use the API Wrapper, please refer to the API documentation linked on this page: SPARQUE API Wrapper. The API Wrapper automatically connects to the production APIs.

The search index is rebuilt daily. New data created during the day appears in the shop search the next day.