Why am i getting 'Unknown field 'embedding' in vector field list.' exception?

[PanClifton]

hello

i am using kernel memory to implement RAG with Azure AI Search, when i am calling KernelMemory.AskAsync(...) the follwoing exception is thrown. Could you kindly advise why exception is thrown?

Azure.RequestFailedException: 'Unknown field 'embedding' in vector field list.
Status: 400 (Bad Request)
ErrorCode: InvalidRequestParameter

this is my setup
project file

<Project Sdk="Microsoft.NET.Sdk">

	<PropertyGroup>
		<OutputType>Exe</OutputType>
		<TargetFramework>net8.0</TargetFramework>
		<ImplicitUsings>enable</ImplicitUsings>
		<Nullable>enable</Nullable>
	</PropertyGroup>

	<ItemGroup>
		<PackageReference Include="Azure.Identity" Version="1.13.1" />
		<PackageReference Include="Microsoft.Extensions.DependencyInjection" Version="8.0.1" />
		<PackageReference Include="Microsoft.Extensions.Hosting.Abstractions" Version="8.0.1" />
		<PackageReference Include="Microsoft.KernelMemory.Abstractions" Version="0.92.241112.1" />
		<PackageReference Include="Microsoft.KernelMemory.Core" Version="0.92.241112.1" />
		<PackageReference Include="Microsoft.KernelMemory.MemoryDb.AzureAISearch" Version="0.92.241112.1" />
		<PackageReference Include="Microsoft.KernelMemory.Service.AspNetCore" Version="0.92.241112.1" />
	</ItemGroup>

	<ItemGroup>
	  <ProjectReference Include="..\HeiDi.Sandbox.SemanticKernelPoc.Utils\HeiDi.Sandbox.SemanticKernelPoc.Utils.csproj" />
	</ItemGroup>

</Project>

program.cs

using HeiDi.Sandbox.SemanticKernelPoc.Utils.Providers;
using Microsoft.Extensions.DependencyInjection;
using Microsoft.Extensions.Logging;
using Microsoft.KernelMemory;

var configurationProvider = ConfigurationsProvider.CreateNew();
var aiSearchConfigration = configurationProvider.AiSearchConfiguration;
var openAiConfigration = configurationProvider.OpenAiConfigration;

var azureOpenAITextConfig = new AzureOpenAIConfig()
{
    Auth = AzureOpenAIConfig.AuthTypes.APIKey,
    Endpoint = openAiConfigration.Endpoint,
    APIType = AzureOpenAIConfig.APITypes.ChatCompletion,
    APIKey = openAiConfigration.ApiKey,
    Deployment = openAiConfigration.ModelDelpoymentName
};
var azureOpenAIEmbeddingConfig = new AzureOpenAIConfig()
{
    Auth = AzureOpenAIConfig.AuthTypes.APIKey,
    Endpoint = openAiConfigration.Endpoint,
    APIKey = openAiConfigration.ApiKey,
    APIType = AzureOpenAIConfig.APITypes.EmbeddingGeneration,
    Deployment = openAiConfigration.TextEmbeddedModelDelpoymentName
};

var azureAISearchConfig = new AzureAISearchConfig()
{
    APIKey = aiSearchConfigration.ApiKey,
    Endpoint = aiSearchConfigration.Endpoint,
    Auth = AzureAISearchConfig.AuthTypes.APIKey,
    UseHybridSearch = true,
};

var memoryBuilder =
    new KernelMemoryBuilder()
        .WithAzureOpenAITextGeneration(azureOpenAITextConfig)
        .WithAzureOpenAITextEmbeddingGeneration(azureOpenAIEmbeddingConfig)
        .WithAzureAISearchMemoryDb(azureAISearchConfig)
        .WithSearchClientConfig(new SearchClientConfig { MaxMatchesCount = 2, Temperature = 0, TopP = 0 });

memoryBuilder.Services.AddLogging(services => services.AddConsole().SetMinimumLevel(LogLevel.Trace));

var memory = memoryBuilder.Build();

try
{
    var answer = await memory.AskAsync("Could you describe the change request process including all steps?", "contracts-index");
}
catch (Exception ex)
{
    /*
       
        exception message

        Unknown field 'embedding' in vector field list.
        Status: 400 (Bad Request)
        ErrorCode: InvalidRequestParameter

        Content:
        {"error":{"code":"InvalidRequestParameter","message":"Unknown field 'embedding' in vector field list.","details":[{"code":"UnknownField","message":"Unknown field 'embedding' in vector field list."}]}}

        Headers:
        Cache-Control: no-cache,no-store
        Pragma: no-cache
        Server: Microsoft-IIS/10.0
        client-request-id: 8a3f39a7-255c-4fa4-a147-b3f1cab2b38b
        x-ms-client-request-id: 8a3f39a7-255c-4fa4-a147-b3f1cab2b38b
        request-id: 8a3f39a7-255c-4fa4-a147-b3f1cab2b38b
        elapsed-time: 67
        Strict-Transport-Security: REDACTED
        Date: Mon, 18 Nov 2024 08:08:48 GMT
        Content-Length: 200
        Content-Type: application/json; charset=utf-8
        Content-Language: REDACTED
        Expires: -1

    */
}

// tags, payload

Console.WriteLine("End of the line....");

[dluc]

hi @PanClifton - did you import some data or are you working with an empty db? any chance you might be reusing an index created by some other application?

[PanClifton]

i am working on an index created by other application.

[PanClifton]

i have some files on blob storageand i want create an index on them.
what should i do?
@dluc

[dluc]

Hi @PanClifton,

Kernel Memory automatically manages indexes with a specific schema and set of fields, so it’s likely that the other application’s index uses a different, incompatible schema.

To import files from blob storage, you can use the ImportDocument API and upload the files directly.

[PanClifton]

thanks @dluc