Thursday, May 4, 2023

How to Include @search.score in Azure Cognitive Search Suggest Response

 You can return the search score (@search.score) in the response along with the suggested search terms using Azure Cognitive Search Suggest API. Here's how you can do it:

  1. In the suggest query, add "@search.score" to the "select" parameter to include the search score in the response.

For example:

https://[service name].search.windows.net/indexes/[index name]/docs/suggest?api-version=[api-version]&suggesterName=[suggester name]&search=[user input]&$select=searchText,@search.score

  1. In the Suggester definition, add "@search.score" to the "sourceFields" parameter to enable scoring of the suggested search terms.

For example:

{ "name": "[suggester name]", "searchMode": "analyzingInfixMatching", "sourceFields": ["[field name 1]", "[field name 2]", "@search.score"] }


How to Create a Static Website in Azure Accessible Only on Company VPN with Custom Domain

 It is possible to have a static website with a custom domain that is fully locked down to just a company's VPN in Azure.

Here are the steps you can follow:

  1. Create a storage account and enable static website hosting.

  2. Upload your static website content to the $web container in the storage account.

  3. Create a private endpoint for the storage account.

  4. Configure the private endpoint to allow traffic only from the company's VPN.

  5. Create a custom domain and add a CNAME record pointing to the Azure CDN endpoint.

  6. Create a CDN profile and a CDN endpoint.

  7. Configure the CDN endpoint to use the storage account as the origin.

  8. Configure the CDN endpoint to use HTTPS and a custom domain.

  9. Lock down the CDN endpoint to allow traffic only from the company's VPN.

By following these steps, you can have a static website with a custom domain that is fully locked down to just a company's VPN in Azure. The CDN endpoint will serve the static website content from the storage account, and access to the CDN endpoint will be restricted to only the company's VPN.

Tuesday, May 2, 2023

Detecting Changes in Azure Data Factory Triggers with KQL Queries

 To detect changes in Azure Data Factory (ADF) triggers using Kusto Query Language (KQL), you can use the AzureActivity table in Log Analytics. You can use the following KQL query to identify trigger changes:


AzureActivity | where Category == "DataFactoryPipelineRun" | where OperationName == "Microsoft.DataFactory/factories/pipelines/create" | where ResourceProvider == "MICROSOFT.DATAFACTORY" | where ActivityStatusValue == "Succeeded" | where Details contains "New-AzDataFactoryPipeline"


This query looks for successful pipeline creation operations in ADF and specifically checks if the New-AzDataFactoryPipeline command was used, indicating a new pipeline was created. You can adjust the query to filter for specific triggers or time ranges by adding additional where clauses.

Note that if ADF auditing is not enabled, or if the logs are not sent to Log Analytics, the AzureActivity table may not contain the necessary information.

How cache can be enabled for embeded text as well for search query results in Azure AI ?

 Great question, Rahul! Caching in the context of Azure AI (especially when using **RAG pipelines with Azure OpenAI + Azure AI Search**) can...