Share via

Azure AI Index Search not working

D'Ambrosio Marlon 0 Reputation points
2025-11-18T11:24:46.0466667+00:00

Hello,

I have no idea why and how this happened.

Basically from one day to the next my Azure AI Index and the searching within is broken.
I dont have an indexer and use my own application to write into the index.

User's image

Like there are documents in there, but they seemed to not being able to be searched. I developed with Azure AI Search for about two months now and it seemed like from one day to the next I cannot search it anymore. I did not change anything in the code or reindexed information as the application was already in testing. So just somewhat the last week, it broke.
I even changed the index, changed the resource and what worked before, did not work now.

I changed the API Version to ingest multiple times and nothing changed. I created new Resources in different regions it does not change the reuslts.I always have documents in the index and cannot search them. Neither a simple search nor a normal semnatic search. What am I doing wrong?

The Azure Endpoint I am calling does always return succesfull as well. The content I am uploading is not empty either it is filled with information and emebddings like it should be.

It is not the scheme, because I did not change anything there and it was working before right. But just to be complete this here is the scheme I use.

Was there an update , did something change or have I missed something that just breaks it?

{
  "@odata.etag": "\"0x8DE268DCF3309F5\"",
  "name": "legal-documents-index-testing",
  "permissionFilterOption": "enabled",
  "fields": [
    {
      "name": "id",
      "type": "Edm.String",
      "searchable": false,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": true,
      "facetable": true,
      "key": true,
      "synonymMaps": []
    },
    {
      "name": "content",
      "type": "Edm.String",
      "searchable": true,
      "filterable": false,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": false,
      "key": false,
      "analyzer": "en.lucene",
      "synonymMaps": []
    },
    {
      "name": "content_vector",
      "type": "Collection(Edm.Single)",
      "searchable": true,
      "filterable": false,
      "retrievable": false,
      "stored": true,
      "sortable": false,
      "facetable": false,
      "key": false,
      "dimensions": 1536,
      "vectorSearchProfile": "default-vector-profile",
      "synonymMaps": []
    },
    {
      "name": "hierarchy_path",
      "type": "Edm.String",
      "searchable": true,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": true,
      "facetable": true,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "section_titles",
      "type": "Collection(Edm.String)",
      "searchable": true,
      "filterable": false,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": false,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "document_tag",
      "type": "Edm.String",
      "searchable": false,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": true,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "document_id",
      "type": "Edm.String",
      "searchable": false,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": true,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "document_title",
      "type": "Edm.String",
      "searchable": true,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": false,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "metadata_storage_path",
      "type": "Edm.String",
      "searchable": false,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": false,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "package_id",
      "type": "Edm.String",
      "searchable": true,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": true,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "keywords",
      "type": "Collection(Edm.String)",
      "searchable": true,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": true,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "source_document",
      "type": "Edm.String",
      "searchable": false,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": true,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "page",
      "type": "Collection(Edm.Int32)",
      "searchable": false,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": false,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "word_count",
      "type": "Edm.Int32",
      "searchable": false,
      "filterable": false,
      "retrievable": true,
      "stored": true,
      "sortable": true,
      "facetable": false,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "metadata_storage_name",
      "type": "Edm.String",
      "searchable": false,
      "filterable": false,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": false,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "metadata_storage_last_modified",
      "type": "Edm.DateTimeOffset",
      "searchable": false,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": true,
      "facetable": false,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "metadata_author",
      "type": "Edm.String",
      "searchable": true,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": true,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "user_ids",
      "type": "Collection(Edm.String)",
      "searchable": false,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": true,
      "key": false,
      "permissionFilter": "userIds",
      "synonymMaps": []
    },
    {
      "name": "group_ids",
      "type": "Collection(Edm.String)",
      "searchable": false,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": true,
      "key": false,
      "permissionFilter": "groupIds",
      "synonymMaps": []
    },
    {
      "name": "rbac_scope",
      "type": "Edm.String",
      "searchable": false,
      "filterable": true,
      "retrievable": false,
      "stored": true,
      "sortable": true,
      "facetable": true,
      "key": false,
      "permissionFilter": "rbacScope",
      "synonymMaps": []
    },
    {
      "name": "images",
      "type": "Collection(Edm.ComplexType)",
      "fields": [
        {
          "name": "image_id",
          "type": "Edm.String",
          "searchable": false,
          "filterable": true,
          "retrievable": true,
          "stored": true,
          "sortable": false,
          "facetable": false,
          "key": false,
          "synonymMaps": []
        },
        {
          "name": "page",
          "type": "Edm.Int32",
          "searchable": false,
          "filterable": true,
          "retrievable": true,
          "stored": true,
          "sortable": false,
          "facetable": false,
          "key": false,
          "synonymMaps": []
        },
        {
          "name": "format",
          "type": "Edm.String",
          "searchable": false,
          "filterable": true,
          "retrievable": true,
          "stored": true,
          "sortable": false,
          "facetable": true,
          "key": false,
          "synonymMaps": []
        },
        {
          "name": "description",
          "type": "Edm.String",
          "searchable": true,
          "filterable": false,
          "retrievable": true,
          "stored": true,
          "sortable": false,
          "facetable": false,
          "key": false,
          "synonymMaps": []
        },
        {
          "name": "bbox",
          "type": "Edm.String",
          "searchable": false,
          "filterable": false,
          "retrievable": true,
          "stored": true,
          "sortable": false,
          "facetable": false,
          "key": false,
          "synonymMaps": []
        }
      ]
    }
  ],
  "scoringProfiles": [],
  "suggesters": [],
  "analyzers": [],
  "normalizers": [],
  "tokenizers": [],
  "tokenFilters": [],
  "charFilters": [],
  "similarity": {
    "@odata.type": "#Microsoft.Azure.Search.BM25Similarity"
  },
  "semantic": {
    "configurations": [
      {
        "name": "legal-semantic-config",
        "flightingOptIn": false,
        "rankingOrder": "BoostedRerankerScore",
        "prioritizedFields": {
          "titleField": {
            "fieldName": "document_title"
          },
          "prioritizedContentFields": [
            {
              "fieldName": "content"
            }
          ],
          "prioritizedKeywordsFields": [
            {
              "fieldName": "section_titles"
            }
          ]
        }
      }
    ]
  },
  "vectorSearch": {
    "algorithms": [
      {
        "name": "default-hnsw",
        "kind": "hnsw",
        "hnswParameters": {
          "metric": "cosine",
          "m": 4,
          "efConstruction": 400,
          "efSearch": 500
        }
      }
    ],
    "profiles": [
      {
        "name": "default-vector-profile",
        "algorithm": "default-hnsw"
      }
    ],
    "vectorizers": [],
    "compressions": []
  }
}

Now this index has only 2 documents but it had about 2000. I dont know what happened if this is a broken resource, if there was an update or some configurations that changed.

Thank you for all the help, this i extremly frustrating

Azure AI Search
Azure AI Search
An Azure search service with built-in artificial intelligence capabilities that enrich information to help identify and explore relevant content at scale.

2 answers

Sort by: Most helpful
Most helpful Newest Oldest
  1. Vimal Lalani 2,400 Reputation points Microsoft External Staff Moderator
    2025-11-18T12:58:19.9766667+00:00

    Hi @D'Ambrosio Marlon

    Thank you for posting your question on the Microsoft Q&A forum.

    Based on the details you shared, the search feature in your Azure AI Search resource was working earlier but has recently stopped returning results, even though the index contains documents and the API calls succeed. Creating new indexes, new resources, and changing API versions has also not helped.

    Here are a few common checks that may help identify the issue:

    • If documents share the same key, earlier documents may be overwritten. Please verify that each document has a unique key in your data source.
    • Check if any index updates, field changes, or deletions occurred recently, as these can affect search results.
    • Diagnostic logs can show ingestion failures, throttling, or query issues that may not surface in the API response.
    • If you are using indexers, review the execution history for errors or warnings that may explain the missing search results. Where can you find specific indexer errors

    You can refer to for more guidance on troubleshooting indexing issues

    Please feel free to post back with any additional details or error messages,

    0 comments
  2. Sina Salam 26,661 Reputation points Volunteer Moderator
    2025-11-20T08:46:27.7833333+00:00

    Hello D'Ambrosio Marlon,

    Welcome to the Microsoft Q&A and thank you for posting your questions here.

    I understand that your Azure AI Index Search not working.

    To be specific, narrow down by running through ingestion monitoring, logs checking, REST tests, and API migrations, you can pinpoint whether it’s a key/permission overwrite, vector mismatch, API route change, or permission filter failure and to fix it follow the below:

    1. To validate API version and permissions query endpoint: GET /indexes/{indexName}/api-version=2025-05-01-preview if it fails or unknown properties appear, push a migration to 2025-09-01-stable or 2025-11-01-preview. - https://learn.microsoft.com/en-us/azure/search/search-api-migration and check your application is not continuing to use 2023-07-01-preview (initial vector support, now deprecated). - https://learn.microsoft.com/en-us/azure/search/search-api-migration and make sure your vector push matches index schema (e.g., "content_vector": [...], with 1536 Float32 or Float16 compatible with Collection(Edm.Single)).
    2. On permission filter check, check that permissionFilterOption: enabled means ACLs are enforced and confirm: 
         1.  Pushed metadata match the indexable field structure.
   
   
   2.  Correct identity included via `x-ms-query-source-authorization` token or API key with viewer role. - https://github.com/MicrosoftDocs/azure-ai-docs/blob/main/articles/search/search-query-access-control-rbac-enforcement.md
   
   
   3.  Test `search=*` with API key or curl command to bypass filter temporarily.
   ```1. For monitoring ingested data: 

   - Use REST: `GET /indexers/{index}/status?api-version=2024-07-01` even for push.
   
   - Confirm status: `status=="success"` and `itemsIndexed > 0`. Loop until success or log fail. - https://stackoverflow.com/questions/79554938/azure-ai-search-documents-not-available-immediately-after-upload-how-to-check
    3. Use Search Explorer / REST to Test Fields by Run: 
         GET https://<service>.search.windows.net/indexes/<index>/docs?search=*&$count=true&api-version=<latest>
    4. Then, inspect diagnostic logs thoroughly via Azure Portal (or Log Analytics), search the SearchService logs for:
    5. For vector-specific validation:
      • Test fallback to text query only: content: "keyword" without embeddings. If that returns results, the problem is in vector search.
      • If vector search broken while text works, ensure incompatibility isn’t due to new syntax (enableRescoring). - https://docs.azure.cn/en-us/search/whats-new

    I hope this is helpful! Do not hesitate to let me know if you have any other questions or clarifications.

    Please don't forget to close up the thread here by upvoting and accept it as an answer if it is helpful.

    0 comments

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.