microsoft/autogen · error · ValueError

Index '{self.search_config.index_name}' not found.

Error message

Index '{self.search_config.index_name}' not found.

What it means

Raised by AzureAISearchTool when the underlying azure.search.documents client throws an exception whose message contains 'not found', most commonly because the index named in AzureAISearchConfig.index_name does not exist on the search service. The original HttpResponseError is chained via `from e`. It is wrapped in a plain ValueError so callers see a config-oriented message instead of an SDK HTTP error.

Source

Thrown at python/packages/autogen-ext/src/autogen_ext/tools/azure/_ai_search.py:579

                except Exception as e:
                    logger.warning(f"Error processing search document: {e}")
                    continue

            if self.search_config.enable_caching:
                self._cache[cache_key] = {"results": results, "timestamp": time.time()}

            return SearchResults(results=results)

        except asyncio.CancelledError:
            raise
        except Exception as e:
            error_msg = str(e)
            if isinstance(e, HttpResponseError):
                if hasattr(e, "message") and e.message:
                    error_msg = e.message

            if "not found" in error_msg.lower():
                raise ValueError(f"Index '{self.search_config.index_name}' not found.") from e
            elif "unauthorized" in error_msg.lower() or "401" in error_msg:
                raise ValueError(f"Authentication failed: {error_msg}") from e
            else:
                raise ValueError(f"Error from Azure AI Search: {error_msg}") from e

    def _to_config(self) -> AzureAISearchConfig:
        """Convert the current instance to a configuration object."""
        return self.search_config

    @property
    def schema(self) -> ToolSchema:
        """Return the schema for the tool."""
        return {
            "name": self.name,
            "description": self.description,
            "parameters": {
                "type": "object",
                "properties": {"query": {"type": "string", "description": "Search query text"}},

View on GitHub (pinned to 027ecf0a37)

Solutions

1. Verify the index exists: az search index list --service-name <service> --resource-group <rg> and compare with the exact index_name passed to the tool.
2. Print/inspect search_config.index_name right before use to catch typos, casing, and stray whitespace (e.g. f'[{tool._to_config().index_name}]').
3. Confirm the endpoint belongs to the search service that actually hosts the index.
4. If the index was deleted or not yet created, (re)create and populate it before running the tool.
5. In multi-tenant setups, drive index_name from configuration per environment instead of hard-coding it.

Example fix

// before
AzureAISearchTool(name='search', endpoint=endpoint, index_name='hotels', credential=AzureKeyCredential(key))
// after (match the real index name on the service)
AzureAISearchTool(name='search', endpoint=endpoint, index_name='hotels-idx', credential=AzureKeyCredential(key))

from azure.search.documents.indexes.aio import SearchIndexClient
from azure.core.credentials import AzureKeyCredential

async def index_exists(endpoint: str, index_name: str, credential) -> bool:
    client = SearchIndexClient(endpoint=endpoint, credential=credential)
    try:
        names = [i async for i in client.list_index_names()]
        return index_name in names
    finally:
        await client.close()

def is_valid_index_name(name: str) -> bool:
    return bool(name) and name == name.strip() and ' ' not in name

try:
    results = await tool.run(args)
except ValueError as e:
    if 'not found' in str(e) and tool._to_config().index_name in str(e):
        # bad index name / wrong service — fix config, do not retry blindly
        raise
    raise

When it happens

Trigger: Calling the tool's run/execution path (e.g. agent invokes the Azure AI Search tool) where SearchClient.search() fails with a 404/ResourceNotFound response; passing index_name='hotels' when the service only has 'hotels-idx'; using an index name with typos, wrong casing, or one that was deleted; querying a different (wrong) search service whose endpoint hosts no such index.

Common situations: Index name copied from another environment (dev index name used against prod service), index deleted or never created before running the agent, endpoint pointing at a different Azure AI Search resource, or a trailing-space/typo in index_name in the constructor or config dict.

Related errors

• Invalid server URL configuration
• Error from Azure AI Search: {error_msg}
• Invalid configuration: {str(e)}
• vector_fields must contain at least one field name for vecto
• vector_fields must contain at least one field name for hybri

AI-assisted analysis of microsoft/autogen@027ecf0a37 (2026-08-15). Data as JSON: /api/errors/d314e879e6c15693. Report an issue: GitHub.