✨(backend) prefix for service index name

Service opensearch index is now defined by the property 'index_name' and
prefixed by the new setting OPENSEARCH_INDEX_PREFIX (default: 'find').
Fix parallel execution of tests.

Signed-off-by: Fabre Florian <ffabre@hybird.org>

CHANGELOG.md

@@ -22,3 +22,6 @@ and this project adheres to
               list of services
 - 🔧(compose) rename docker network 'lasuite-net' as 'lasuite-network'
 - ✨(backend) add demo service for Drive.
+- 🐛(backend) Fix parallel test execution issues
+- ✨(backend) Add OPENSEARCH_INDEX_PREFIX setting to prevent naming overlaping
+              issues if the opensearch database is shared between apps.

src/backend/core/management/commands/reindex_with_embedding.py

@@ -9,6 +9,7 @@ from django.core.management.base import BaseCommand, CommandError
 
 from opensearchpy.exceptions import NotFoundError
 
+from core.models import get_opensearch_index_name
 from core.services.opensearch import (
     check_hybrid_search_enabled,
     embed_text,
@@ -31,7 +32,7 @@ class Command(BaseCommand):
     def handle(self, *args, **options):
         """Launch the reindexing with embedding."""
 
-        index_name = options["index_name"]
+        index_name = get_opensearch_index_name(options["index_name"])
 
         if not check_hybrid_search_enabled():
             raise CommandError("Hybrid search is not enabled or properly configured.")

src/backend/core/models.py

@@ -3,9 +3,11 @@
 import secrets
 import string
 
+from django.conf import settings
 from django.contrib.auth.models import AbstractUser
 from django.db import models
 from django.db.models.functions import Length
+from django.utils.functional import cached_property
 from django.utils.text import slugify
 from django.utils.translation import gettext_lazy as _
 
@@ -13,6 +15,11 @@ models.CharField.register_lookup(Length)
 TOKEN_LENGTH = 50
 
 
+def get_opensearch_index_name(name: str):
+    """Returns the opensearch index for a service name"""
+    return f"{settings.OPENSEARCH_INDEX_PREFIX}-{name}"
+
+
 class User(AbstractUser):
     """User for the find application"""
 
@@ -61,3 +68,8 @@ class Service(models.Model):
         )
         token = "".join(secrets.choice(characters) for _ in range(TOKEN_LENGTH))
         return token
+
+    @cached_property
+    def index_name(self):
+        """Returns the opensearch index for the service"""
+        return get_opensearch_index_name(self.name)

src/backend/core/tests/commands/test_reindex_with_embedding.py

@@ -15,12 +15,12 @@ from core.management.commands.reindex_with_embedding import (
 from core.management.commands.reindex_with_embedding import (
     reindex_with_embedding,
 )
+from core.models import get_opensearch_index_name
 from core.services.opensearch import check_hybrid_search_enabled, opensearch_client
 from core.tests.mock import albert_embedding_response
 from core.tests.utils import (
     bulk_create_documents,
     delete_search_pipeline,
-    delete_test_indices,
     enable_hybrid_search,
     prepare_index,
 )
@@ -43,7 +43,6 @@ def clear_caches():
     # is different and must be cleared separately
     check_hybrid_search_enabled_command.cache_clear()
     delete_search_pipeline()
-    delete_test_indices()
 
 
 @responses.activate
@@ -58,11 +57,12 @@ def test_reindex_with_embedding_command(settings):
             {"title": "cat", "content": "cats are curious and independent pets"},
         ]
     )
-    prepare_index(SERVICE_NAME, documents)
+    index_name = get_opensearch_index_name(SERVICE_NAME)
+    prepare_index(index_name, documents)
 
     # the index has not been embedded in the initial state
     initial_index = opensearch_client_.search(
-        index=SERVICE_NAME, size=3, body={"query": {"match_all": {}}}
+        index=index_name, size=3, body={"query": {"match_all": {}}}
     )
     assert len(initial_index["hits"]["hits"]) == 3
     for embedded_hit in initial_index["hits"]["hits"]:
@@ -81,9 +81,9 @@ def test_reindex_with_embedding_command(settings):
 
     call_command("reindex_with_embedding", SERVICE_NAME)
 
-    opensearch_client_.indices.refresh(index=SERVICE_NAME)
+    opensearch_client_.indices.refresh(index=index_name)
     embedded_index = opensearch_client_.search(
-        index=SERVICE_NAME, size=3, body={"query": {"match_all": {}}}
+        index=index_name, size=3, body={"query": {"match_all": {}}}
     )
 
     # the source index has been replaced with embedding version
@@ -121,7 +121,8 @@ def test_reindex_can_fail_and_restart(settings):
             {"title": "cat", "content": "cats are curious and independent pets"},
         ]
     )
-    prepare_index(SERVICE_NAME, documents)
+    index_name = get_opensearch_index_name(SERVICE_NAME)
+    prepare_index(index_name, documents)
 
     # enable hybrid search after first indexing
     enable_hybrid_search(settings)
@@ -147,16 +148,16 @@ def test_reindex_can_fail_and_restart(settings):
         status=200,
     )
 
-    result = reindex_with_embedding(SERVICE_NAME)
+    result = reindex_with_embedding(index_name)
 
     # assert results reflect 2 successes and 1 failure
     assert result["nb_success_embedding"] == 2
     assert result["nb_failed_embedding"] == 1
 
     # assert the index state
-    opensearch_client_.indices.refresh(index=SERVICE_NAME)
+    opensearch_client_.indices.refresh(index=index_name)
     embedded_index = opensearch_client_.search(
-        index=SERVICE_NAME, size=3, body={"query": {"match_all": {}}}
+        index=index_name, size=3, body={"query": {"match_all": {}}}
     )
     # Should have 2 documents with embeddings, 1 without due to error
     embedded_count = 0
@@ -180,16 +181,16 @@ def test_reindex_can_fail_and_restart(settings):
         json=albert_embedding_response.response,
         status=200,
     )
-    result = reindex_with_embedding(SERVICE_NAME)
+    result = reindex_with_embedding(index_name)
 
     # assert results
     assert result["nb_success_embedding"] == 1
     assert result["nb_failed_embedding"] == 0
 
     # assert there is now 1 more success and 0 failures
-    opensearch_client_.indices.refresh(index=SERVICE_NAME)
+    opensearch_client_.indices.refresh(index=index_name)
     embedded_index = opensearch_client_.search(
-        index=SERVICE_NAME, size=3, body={"query": {"match_all": {}}}
+        index=index_name, size=3, body={"query": {"match_all": {}}}
     )
     for hit in embedded_index["hits"]["hits"]:
         assert (
@@ -220,7 +221,8 @@ def test_reindex_preserves_concurrent_updates(settings):
             {"title": "dog", "content": "dogs are loyal domestic animals"},
         ]
     )
-    prepare_index(SERVICE_NAME, documents)
+    index_name = get_opensearch_index_name(SERVICE_NAME)
+    prepare_index(index_name, documents)
     enable_hybrid_search(settings)
 
     updated_title = "updated dog"
@@ -231,7 +233,7 @@ def test_reindex_preserves_concurrent_updates(settings):
     patch(
         "core.services.opensearch.opensearch_client_.search",
         side_effect=opensearch_client_.update(
-            index=SERVICE_NAME,
+            index=index_name,
             id=documents[1]["id"],
             body={
                 "doc": {
@@ -249,13 +251,13 @@ def test_reindex_preserves_concurrent_updates(settings):
         json=albert_embedding_response.response,
         status=200,
     )
-    result = reindex_with_embedding(SERVICE_NAME)
+    result = reindex_with_embedding(index_name)
     assert result["nb_success_embedding"] == 2
     assert result["nb_failed_embedding"] == 0
 
-    opensearch_client_.indices.refresh(index=SERVICE_NAME)
+    opensearch_client_.indices.refresh(index=index_name)
     embedded_index = opensearch_client_.search(
-        index=SERVICE_NAME, size=2, body={"query": {"match_all": {}}}
+        index=index_name, size=2, body={"query": {"match_all": {}}}
     )
     # Check that the latest update is preserved
     dog_doc = [
@@ -281,7 +283,9 @@ def test_reindex_command_but_index_does_not_exist(settings):
     wrong_index = "wrong-index-name"
     enable_hybrid_search(settings)
 
+    wrong_index_name = get_opensearch_index_name(wrong_index)
+
     with pytest.raises(CommandError) as err:
         call_command("reindex_with_embedding", wrong_index)
 
-    assert str(err.value) == f"Index {wrong_index} does not exist."
+    assert str(err.value) == f"Index {wrong_index_name} does not exist."

src/backend/core/tests/conftest.py

@@ -1,10 +1,15 @@
 """Fixtures for tests in the find core application"""
 
 import pytest
+from faker import Faker
 from lasuite.oidc_resource_server.authentication import (
     get_resource_server_backend,
 )
 
+from core.services import opensearch
+
+fake = Faker()
+
 
 @pytest.fixture(name="jwt_rs_backend")
 def jwt_resource_server_backend_fixture(settings):
@@ -20,3 +25,27 @@ def jwt_resource_server_backend_fixture(settings):
 
     settings.OIDC_RS_BACKEND_CLASS = _original_backend
     get_resource_server_backend.cache_clear()
+
+
+@pytest.fixture(autouse=True)
+def cleanup_test_index(settings):
+    """
+    Fixture to set a randomized prefix for all service indexes within the tests
+    and remove them on tear down.
+    """
+    _original_prefix = settings.OPENSEARCH_INDEX_PREFIX
+    prefix = "".join(fake.random_letters(5)).lower()
+    settings.OPENSEARCH_INDEX_PREFIX = prefix
+
+    # Create client here to prevent "teardown" issues when the opensearch settings are
+    # removed for error tests.
+    client = opensearch.opensearch_client()
+
+    yield
+
+    settings.OPENSEARCH_INDEX_PREFIX = _original_prefix
+
+    try:
+        client.indices.delete(index=f"{prefix}-*")
+    except opensearch.NotFoundError:
+        pass

src/backend/core/tests/test_api_documents_index_bulk.py

@@ -43,7 +43,7 @@ def test_api_documents_index_bulk_invalid_token():
 
 def test_api_documents_index_bulk_success():
     """A registered service should be able to index documents in bulk with a valid token."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(3)
 
     response = APIClient().post(
@@ -61,14 +61,11 @@ def test_api_documents_index_bulk_success():
 def test_api_documents_index_bulk_ensure_index():
     """A registered service should be create the opensearch index if need."""
     opensearch_client_ = opensearch.opensearch_client()
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(3)
 
-    # Delete the index
-    opensearch_client_.indices.delete(index="*test*")
-
     with pytest.raises(opensearch.NotFoundError):
-        opensearch_client_.indices.get(index="test-service")
+        opensearch_client_.indices.get(index=service.index_name)
 
     response = APIClient().post(
         "/api/v1.0/documents/index/",
@@ -83,7 +80,7 @@ def test_api_documents_index_bulk_ensure_index():
     assert [d["status"] for d in responses] == ["success"] * 3
 
     # The index has been rebuilt
-    opensearch_client_.indices.get(index="test-service")
+    opensearch_client_.indices.get(index=service.index_name)
 
 
 @pytest.mark.parametrize(
@@ -202,7 +199,7 @@ def test_api_documents_index_bulk_invalid_document(
     field, invalid_value, error_type, error_message
 ):
     """Test bulk document indexing with various invalid fields."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(3)
 
     # Modify the first document with the invalid value for the specified field
@@ -242,7 +239,7 @@ def test_api_documents_index_bulk_invalid_document(
 )
 def test_api_documents_index_bulk_required(field):
     """Test bulk document indexing with a required field missing."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(3)
 
     del documents[0][field]
@@ -274,7 +271,7 @@ def test_api_documents_index_bulk_required(field):
 )
 def test_api_documents_index_bulk_default(field, default_value):
     """Test bulk document indexing while removing optional fields that have default values."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(3)
 
     del documents[0][field]
@@ -291,14 +288,14 @@ def test_api_documents_index_bulk_default(field, default_value):
     assert [d["status"] for d in responses] == ["success"] * 3
 
     indexed_document = opensearch.opensearch_client().get(
-        index=service.name, id=responses[0]["_id"]
+        index=service.index_name, id=responses[0]["_id"]
     )["_source"]
     assert indexed_document[field] == default_value
 
 
 def test_api_documents_index_bulk_updated_at_before_created_at():
     """Test bulk document indexing with updated_at before created_at."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(3)
 
     documents[0]["updated_at"] = documents[0]["created_at"] - datetime.timedelta(
@@ -332,7 +329,7 @@ def test_api_documents_index_bulk_updated_at_before_created_at():
 )
 def test_api_documents_index_bulk_datetime_future(field):
     """Test bulk document indexing with datetimes in the future."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(3)
 
     now = timezone.now()
@@ -360,7 +357,7 @@ def test_api_documents_index_bulk_datetime_future(field):
 
 def test_api_documents_index_empty_content_check():
     """Test bulk document indexing with both empty title & content."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(3)
 
     documents[0]["content"] = ""
@@ -388,7 +385,7 @@ def test_api_documents_index_empty_content_check():
 
 def test_api_documents_index_opensearch_errors():
     """Test bulk document indexing errors"""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(3)
 
     with mock.patch.object(opensearch.opensearch_client(), "bulk") as mock_bulk:

src/backend/core/tests/test_api_documents_index_single.py

@@ -11,7 +11,7 @@ from rest_framework.test import APIClient
 from core import factories
 from core.services import opensearch
 from core.tests.mock import albert_embedding_response
-from core.tests.utils import delete_test_indices, enable_hybrid_search
+from core.tests.utils import enable_hybrid_search
 
 pytestmark = pytest.mark.django_db
 
@@ -56,7 +56,7 @@ def test_api_documents_index_single_hybrid_enabled_success(settings):
     If hybrid search is enabled, the indexing should have embedding of
     dimension settings.EMBEDDING_DIMENSION.
     """
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     enable_hybrid_search(settings)
     responses.add(
         responses.POST,
@@ -64,6 +64,7 @@ def test_api_documents_index_single_hybrid_enabled_success(settings):
         json=albert_embedding_response.response,
         status=200,
     )
+
     document = factories.DocumentSchemaFactory.build()
 
     response = APIClient().post(
@@ -77,7 +78,7 @@ def test_api_documents_index_single_hybrid_enabled_success(settings):
     assert response.json()["_id"] == str(document["id"])
 
     new_indexed_document = opensearch.opensearch_client().get(
-        index=service.name, id=str(document["id"])
+        index=service.index_name, id=str(document["id"])
     )
     assert new_indexed_document["_version"] == 1
     assert new_indexed_document["_source"]["title"] == document["title"].strip().lower()
@@ -94,7 +95,7 @@ def test_api_documents_index_single_hybrid_enabled_success(settings):
 
 def test_api_documents_index_single_hybrid_disabled_success():
     """If hybrid search is not enabled, the indexing should have an embedding equal to None."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     document = factories.DocumentSchemaFactory.build()
     opensearch.check_hybrid_search_enabled.cache_clear()
 
@@ -109,7 +110,7 @@ def test_api_documents_index_single_hybrid_disabled_success():
     assert response.json()["_id"] == str(document["id"])
 
     new_indexed_document = opensearch.opensearch_client().get(
-        index=service.name, id=str(document["id"])
+        index=service.index_name, id=str(document["id"])
     )
     assert new_indexed_document["_version"] == 1
     assert new_indexed_document["_source"]["title"] == document["title"].strip().lower()
@@ -117,15 +118,14 @@ def test_api_documents_index_single_hybrid_disabled_success():
     assert new_indexed_document["_source"]["embedding"] is None
 
 
-def test_api_documents_index_bulk_ensure_index(settings):
+def test_api_documents_index_single_ensure_index(settings):
     """A registered service should be create the opensearch index if need."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     document = factories.DocumentSchemaFactory.build()
     opensearch_client_ = opensearch.opensearch_client()
-    delete_test_indices()
 
     with pytest.raises(opensearch.NotFoundError):
-        opensearch_client_.indices.get(index="test-service")
+        opensearch_client_.indices.get(index=service.index_name)
 
     response = APIClient().post(
         "/api/v1.0/documents/index/",
@@ -138,9 +138,9 @@ def test_api_documents_index_bulk_ensure_index(settings):
     assert response.json()["_id"] == str(document["id"])
 
     # The index has been rebuilt
-    data = opensearch_client_.indices.get(index="test-service")
+    data = opensearch_client_.indices.get(index=service.index_name)
 
-    assert data["test-service"]["mappings"] == {
+    assert data[service.index_name]["mappings"] == {
         "dynamic": "strict",
         "properties": {
             "id": {"type": "keyword"},
@@ -295,7 +295,7 @@ def test_api_documents_index_single_invalid_document(
     field, invalid_value, error_type, error_message
 ):
     """Test document indexing with various invalid fields."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     document = factories.DocumentSchemaFactory.build()
 
     # Modify the document with the invalid value for the specified field
@@ -330,7 +330,7 @@ def test_api_documents_index_single_invalid_document(
 )
 def test_api_documents_index_single_required(field):
     """Test document indexing with a required field missing."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     document = factories.DocumentSchemaFactory.build()
 
     del document[field]
@@ -357,7 +357,7 @@ def test_api_documents_index_single_required(field):
 )
 def test_api_documents_index_single_default(field, default_value):
     """Test document indexing while removing optional fields that have default values."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     document = factories.DocumentSchemaFactory.build()
 
     del document[field]
@@ -373,14 +373,14 @@ def test_api_documents_index_single_default(field, default_value):
     assert response.json()["_id"] == str(document["id"])
 
     indexed_document = opensearch.opensearch_client().get(
-        index=service.name, id=str(document["id"])
+        index=service.index_name, id=str(document["id"])
     )["_source"]
     assert indexed_document[field] == default_value
 
 
 def test_api_documents_index_single_udpated_at_before_created():
     """Test document indexing with updated_at before created_at."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     document = factories.DocumentSchemaFactory.build()
 
     document["updated_at"] = document["created_at"] - datetime.timedelta(seconds=1)
@@ -406,7 +406,7 @@ def test_api_documents_index_single_udpated_at_before_created():
 )
 def test_api_documents_index_single_datetime_future(field):
     """Test document indexing with datetimes in the future."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     document = factories.DocumentSchemaFactory.build()
 
     now = timezone.now()
@@ -426,7 +426,7 @@ def test_api_documents_index_single_datetime_future(field):
 
 def test_api_documents_index_empty_content_check():
     """Test document indexing with both empty title & content."""
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     document = factories.DocumentSchemaFactory.build()
 
     document["content"] = ""

src/backend/core/tests/test_api_documents_search.py

@@ -49,8 +49,8 @@ def test_api_documents_search_auth_invalid_parameters(settings):
     settings.OIDC_RS_CLIENT_ID = None
     settings.OIDC_RS_CLIENT_SECRET = None
 
-    service = factories.ServiceFactory(name="test-service")
-    prepare_index(service.name, [])
+    service = factories.ServiceFactory()
+    prepare_index(service.index_name, [])
 
     response = APIClient().post(
         "/api/v1.0/documents/search/",
@@ -70,7 +70,7 @@ def test_api_documents_search_opensearch_env_variables_not_set(settings):
     result in a 500 internal server error
     """
     setup_oicd_resource_server(responses, settings, sub="user_sub")
-    factories.ServiceFactory(name="test-service")
+    factories.ServiceFactory()
 
     del settings.OPENSEARCH_HOST  # Remove required settings
     del settings.OPENSEARCH_PASSWORD
@@ -100,8 +100,8 @@ def test_api_documents_search_query_unknown_user(settings):
 
     token = build_authorization_bearer()
 
-    service = factories.ServiceFactory(name="test-service")
-    prepare_index(service.name, [])
+    service = factories.ServiceFactory()
+    prepare_index(service.index_name, [])
 
     response = APIClient().post(
         "/api/v1.0/documents/search/",
@@ -124,7 +124,7 @@ def test_api_documents_search_services_invalid_parameters(settings):
         json=albert_embedding_response.response,
         status=200,
     )
-    factories.ServiceFactory(name="test-service")
+    factories.ServiceFactory()
 
     response = APIClient().post(
         "/api/v1.0/documents/search/",
@@ -154,7 +154,7 @@ def test_api_documents_search_reached_docs_invalid_parameters(settings):
         json=albert_embedding_response.response,
         status=200,
     )
-    factories.ServiceFactory(name="test-service")
+    factories.ServiceFactory()
 
     response = APIClient().post(
         "/api/v1.0/documents/search/",
@@ -186,11 +186,11 @@ def test_api_documents_search_match_all(settings):
         status=200,
     )
     nb_documents = 12
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(
         nb_documents, reach=random.choice(["public", "authenticated"])
     )
-    prepare_index(service.name, documents)
+    prepare_index(service.index_name, documents)
 
     response = APIClient().post(
         "/api/v1.0/documents/search/",
@@ -209,7 +209,7 @@ def test_api_documents_search_match_all(settings):
 def test_api_documents_full_text_search_query_title(settings):
     """Searching a document by its title should work as expected"""
     setup_oicd_resource_server(responses, settings, sub="user_sub")
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
 
     documents = bulk_create_documents(
         [
@@ -218,7 +218,8 @@ def test_api_documents_full_text_search_query_title(settings):
             {"title": "The brown goat", "content": "the wolf"},
         ]
     )
-    prepare_index(service.name, documents)
+
+    prepare_index(service.index_name, documents)
 
     response = APIClient().post(
         "/api/v1.0/documents/search/",
@@ -278,7 +279,7 @@ def test_api_documents_full_text_search(settings):
     setup_oicd_resource_server(responses, settings, sub="user_sub")
     token = build_authorization_bearer()
 
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = bulk_create_documents(
         [
             {"title": "The quick brown fox", "content": "the wolf"},
@@ -286,7 +287,7 @@ def test_api_documents_full_text_search(settings):
             {"title": "The brown goat", "content": "the wolf"},
         ]
     )
-    prepare_index(service.name, documents)
+    prepare_index(service.index_name, documents)
 
     response = APIClient().post(
         "/api/v1.0/documents/search/",
@@ -356,7 +357,7 @@ def test_api_documents_hybrid_search(settings):
         status=200,
     )  # mock embedding API
 
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = bulk_create_documents(
         [
             {"title": "The quick brown fox", "content": "the wolf"},
@@ -364,7 +365,7 @@ def test_api_documents_hybrid_search(settings):
             {"title": "The brown goat", "content": "the wolf"},
         ]
     )
-    prepare_index(service.name, documents)
+    prepare_index(service.index_name, documents)
 
     response = APIClient().post(
         "/api/v1.0/documents/search/",
@@ -458,11 +459,11 @@ def test_api_documents_search_ordering_by_fields(settings):
         json=albert_embedding_response.response,
         status=200,
     )
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(
         4, reach=random.choice(["public", "authenticated"])
     )
-    prepare_index(service.name, documents)
+    prepare_index(service.index_name, documents)
 
     parameters = [
         (enums.TITLE, "asc"),
@@ -511,11 +512,11 @@ def test_api_documents_search_ordering_by_relevance(settings):
         json=albert_embedding_response.response,
         status=200,
     )
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(
         4, reach=random.choice(["public", "authenticated"])
     )
-    prepare_index(service.name, documents)
+    prepare_index(service.index_name, documents)
 
     for direction in ["asc", "desc"]:
         response = APIClient().post(
@@ -552,11 +553,11 @@ def test_api_documents_search_ordering_by_unknown_field(settings):
         status=200,
     )
     # Setup: Initialize the service and documents only once
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(
         2, reach=random.choice(["public", "authenticated"])
     )
-    prepare_index(service.name, documents)
+    prepare_index(service.index_name, documents)
 
     # Define the parameters manually
     directions = ["asc", "desc"]
@@ -599,11 +600,11 @@ def test_api_documents_search_ordering_by_unknown_direction(settings):
         json=albert_embedding_response.response,
         status=200,
     )
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(
         2, reach=random.choice(["public", "authenticated"])
     )
-    prepare_index(service.name, documents)
+    prepare_index(service.index_name, documents)
 
     for field in enums.ORDER_BY_OPTIONS:
         response = APIClient().post(
@@ -639,11 +640,11 @@ def test_api_documents_search_filtering_by_reach(settings):
         json=albert_embedding_response.response,
         status=200,
     )
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(
         4, reach=random.choice(["public", "authenticated"])
     )
-    prepare_index(service.name, documents)
+    prepare_index(service.index_name, documents)
 
     for reach in enums.ReachEnum:
         response = APIClient().post(
@@ -675,12 +676,12 @@ def test_api_documents_search_with_nb_results(settings):
         json=albert_embedding_response.response,
         status=200,
     )
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(
         9, reach=random.choice(["public", "authenticated"])
     )
     ids = [str(doc["id"]) for doc in documents]
-    prepare_index(service.name, documents)
+    prepare_index(service.index_name, documents)
 
     nb_results = 3
     response = APIClient().post(
@@ -741,11 +742,11 @@ def test_api_documents_search_nb_results_invalid_parameters(settings):
         json=albert_embedding_response.response,
         status=200,
     )
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = factories.DocumentSchemaFactory.build_batch(
         4, reach=random.choice(["public", "authenticated"])
     )
-    prepare_index(service.name, documents)
+    prepare_index(service.index_name, documents)
 
     parameters = [
         (
@@ -782,13 +783,13 @@ def test_api_documents_search_nb_results_with_filtering(settings):
         json=albert_embedding_response.response,
         status=200,
     )
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     public_documents = factories.DocumentSchemaFactory.build_batch(3, reach="public")
     public_ids = [str(doc["id"]) for doc in public_documents]
     private_documents = factories.DocumentSchemaFactory.build_batch(
         2, reach="authenticated"
     )
-    prepare_index(service.name, public_documents + private_documents)
+    prepare_index(service.index_name, public_documents + private_documents)
 
     nb_results = 3
     response = APIClient().post(

src/backend/core/tests/test_api_documents_search_access_control.py

@@ -15,7 +15,6 @@ from core.services.opensearch import opensearch_client
 from .mock import albert_embedding_response
 from .utils import (
     build_authorization_bearer,
-    delete_test_indices,
     prepare_index,
     setup_oicd_resource_server,
 )
@@ -32,11 +31,11 @@ def test_api_documents_search_access_control_anonymous(settings):
         json=albert_embedding_response.response,
         status=200,
     )
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents = []
     for reach in enums.ReachEnum:
         documents.extend(factories.DocumentSchemaFactory.build_batch(3, reach=reach))
-    prepare_index(service.name, documents)
+    prepare_index(service.index_name, documents)
 
     response = APIClient().post("/api/v1.0/documents/search/?q=*")
 
@@ -61,7 +60,7 @@ def test_api_documents_search_access_control(settings):
     setup_oicd_resource_server(responses, settings, sub="user_sub")
     token = build_authorization_bearer()
 
-    service = factories.ServiceFactory(name="test-service")
+    service = factories.ServiceFactory()
     documents_reach = factories.DocumentSchemaFactory.build_batch(6)
     documents_open = [
         doc for doc in documents_reach if doc["reach"] in ["authenticated", "public"]
@@ -71,7 +70,7 @@ def test_api_documents_search_access_control(settings):
     )
     expected_ids = [doc["id"] for doc in documents_open + documents_user]
 
-    prepare_index(service.name, documents_user + documents_reach)
+    prepare_index(service.index_name, documents_user + documents_reach)
 
     response = APIClient().post(
         "/api/v1.0/documents/search/",
@@ -114,7 +113,7 @@ def test_api_documents_search_access__only_visited_public(
     setup_oicd_resource_server(responses, settings, sub="user_sub", audience="docs")
     token = build_authorization_bearer()
 
-    service = factories.ServiceFactory(name="test-service", client_id="docs")
+    service = factories.ServiceFactory(client_id="docs")
 
     docs = [
         factories.DocumentSchemaFactory(
@@ -123,7 +122,7 @@ def test_api_documents_search_access__only_visited_public(
         for doc_id in doc_ids
     ]
 
-    prepare_index(service.name, docs)
+    prepare_index(service.index_name, docs)
 
     response = APIClient().post(
         "/api/v1.0/documents/search/",
@@ -151,7 +150,7 @@ def test_api_documents_search_access__any_owner_public(settings):
     setup_oicd_resource_server(responses, settings, sub="user_sub", audience="docs")
     token = build_authorization_bearer()
 
-    service = factories.ServiceFactory(name="test-service", client_id="docs")
+    service = factories.ServiceFactory(client_id="docs")
 
     docs = factories.DocumentSchemaFactory.build_batch(
         6,
@@ -165,7 +164,7 @@ def test_api_documents_search_access__any_owner_public(settings):
         users=["other_sub"],
     )
 
-    prepare_index(service.name, docs + other_docs)
+    prepare_index(service.index_name, docs + other_docs)
 
     expected = [d["id"] for d in docs]
 
@@ -195,8 +194,8 @@ def test_api_documents_search_access__services(settings):
     setup_oicd_resource_server(responses, settings, sub="user_sub", audience="a-client")
     token = build_authorization_bearer()
 
-    service_a = factories.ServiceFactory(name="test-index-a", client_id="a-client")
-    service_b = factories.ServiceFactory(name="test-index-b", client_id="b-client")
+    service_a = factories.ServiceFactory(client_id="a-client")
+    service_b = factories.ServiceFactory(client_id="b-client")
 
     service_a_docs = factories.DocumentSchemaFactory.build_batch(
         3, reach=enums.ReachEnum.AUTHENTICATED, users=["user_sub"]
@@ -207,8 +206,8 @@ def test_api_documents_search_access__services(settings):
 
     expected_ids = [doc["id"] for doc in service_a_docs]
 
-    prepare_index(service_a.name, service_a_docs)
-    prepare_index(service_b.name, service_b_docs, cleanup=False)
+    prepare_index(service_a.index_name, service_a_docs)
+    prepare_index(service_b.index_name, service_b_docs)
 
     response = APIClient().post(
         "/api/v1.0/documents/search/",
@@ -236,7 +235,6 @@ def test_api_documents_search_access__missing_index(settings):
     token = build_authorization_bearer()
     factories.ServiceFactory(name="test-index-a", client_id="a-client")
 
-    delete_test_indices()
     opensearch_client.cache_clear()
 
     # a-client has no index. ignore it.
@@ -284,9 +282,9 @@ def test_api_documents_search_access__related_services(settings):
 
     expected_ids = [doc["id"] for doc in service_a_docs + service_c_docs]
 
-    prepare_index(service_a.name, service_a_docs)
-    prepare_index(service_b.name, service_b_docs, cleanup=False)
-    prepare_index(service_c.name, service_c_docs, cleanup=False)
+    prepare_index(service_a.index_name, service_a_docs)
+    prepare_index(service_b.index_name, service_b_docs)
+    prepare_index(service_c.index_name, service_c_docs)
 
     response = APIClient().post(
         "/api/v1.0/documents/search/",
@@ -328,8 +326,8 @@ def test_api_documents_search_access__related_missing_index(settings):
 
     expected_ids = [doc["id"] for doc in service_c_docs]
 
-    prepare_index(service_b.name, service_b_docs)
-    prepare_index(service_c.name, service_c_docs, cleanup=False)
+    prepare_index(service_b.index_name, service_b_docs)
+    prepare_index(service_c.index_name, service_c_docs)
 
     # a-client has no index. ignore it.
     response = APIClient().post(
@@ -375,9 +373,9 @@ def test_api_documents_search_access__request_services(settings):
 
     expected_ids = [doc["id"] for doc in service_c_docs]
 
-    prepare_index(service_a.name, service_a_docs)
-    prepare_index(service_b.name, service_b_docs, cleanup=False)
-    prepare_index(service_c.name, service_c_docs, cleanup=False)
+    prepare_index(service_a.index_name, service_a_docs)
+    prepare_index(service_b.index_name, service_b_docs)
+    prepare_index(service_c.index_name, service_c_docs)
 
     response = APIClient().post(
         "/api/v1.0/documents/search/",
@@ -453,7 +451,7 @@ def test_api_documents_search_access__authenticated(settings):
     setup_oicd_resource_server(responses, settings, sub="user_sub", audience="docs")
     token = build_authorization_bearer()
 
-    service = factories.ServiceFactory(name="test-service", client_id="docs")
+    service = factories.ServiceFactory(client_id="docs")
 
     documents_open = factories.DocumentSchemaFactory.build_batch(
         2, reach=enums.ReachEnum.PUBLIC
@@ -470,7 +468,9 @@ def test_api_documents_search_access__authenticated(settings):
     )
     documents = documents_user + documents_open + documents_restricted
 
-    prepare_index(service.name, documents_user + documents_open + documents_restricted)
+    prepare_index(
+        service.index_name, documents_user + documents_open + documents_restricted
+    )
 
     # Only owned documents (reach is ignored)
     response = APIClient().post(

src/backend/core/tests/test_models_services.py

@@ -17,10 +17,11 @@ def test_models_services_name_unique():
         factories.ServiceFactory(name=service.name)
 
 
-def test_models_services_name_slugified():
+def test_models_services_name_slugified(settings):
     """The name field should be slugified."""
     service = factories.ServiceFactory(name="My service name")
     assert service.name == "my-service-name"
+    assert service.index_name == f"{settings.OPENSEARCH_INDEX_PREFIX}-my-service-name"
 
 
 def test_models_services_token_50_characters_exact():

src/backend/core/tests/test_opensearch.py

@@ -11,6 +11,7 @@ import responses
 
 from core.services import opensearch
 
+from .. import factories
 from ..services.opensearch import (
     check_hybrid_search_enabled,
     embed_text,
@@ -170,7 +171,8 @@ def test_fall_back_on_full_text_search_if_hybrid_search_disabled(settings, caplo
             {"title": "cat", "content": "cats are curious and independent pets"},
         ]
     )
-    prepare_index(SERVICE_NAME, documents)
+    service = factories.ServiceFactory(name=SERVICE_NAME)
+    prepare_index(service.index_name, documents)
 
     q = "wolf"
     with caplog.at_level(logging.INFO):
@@ -207,7 +209,8 @@ def test_fall_back_on_full_text_search_if_embedding_api_fails(settings, caplog):
             {"title": "cat", "content": "cats are curious and independent pets"},
         ]
     )
-    prepare_index(SERVICE_NAME, documents)
+    service = factories.ServiceFactory(name=SERVICE_NAME)
+    prepare_index(service.index_name, documents)
 
     q = "wolf"
     with caplog.at_level(logging.INFO):
@@ -238,7 +241,8 @@ def test_fall_back_on_full_text_search_if_variable_are_missing(settings, caplog)
             {"title": "cat", "content": "cats are curious and independent pets"},
         ]
     )
-    prepare_index(SERVICE_NAME, documents)
+    service = factories.ServiceFactory(name=SERVICE_NAME)
+    prepare_index(service.index_name, documents)
 
     q = "wolf"
     with caplog.at_level(logging.INFO):
@@ -274,7 +278,8 @@ def test_match_all(settings, caplog):
             {"title": "cat", "content": "cats are curious and independent pets"},
         ]
     )
-    prepare_index(SERVICE_NAME, documents)
+    service = factories.ServiceFactory(name=SERVICE_NAME)
+    prepare_index(service.index_name, documents)
 
     q = "*"
     with caplog.at_level(logging.INFO):
@@ -304,7 +309,8 @@ def test_search_ordering_by_relevance(settings, caplog):
         ]
     )
     q = "canine pet"
-    prepare_index(SERVICE_NAME, documents)
+    service = factories.ServiceFactory(name=SERVICE_NAME)
+    prepare_index(service.index_name, documents)
 
     for direction in ["asc", "desc"]:
         with caplog.at_level(logging.INFO):
@@ -338,7 +344,8 @@ def test_hybrid_search_number_of_matches(settings):
             {"title": "cat", "content": "cats are curious and independent pets"},
         ]
     )
-    prepare_index(SERVICE_NAME, documents)
+    service = factories.ServiceFactory(name=SERVICE_NAME)
+    prepare_index(service.index_name, documents)
 
     q = "pony"  # full-text matches 0 document
     for nb_results in [1, 2, 3]:  # semantic should match k documents

src/backend/core/tests/utils.py

@@ -58,16 +58,8 @@ def delete_search_pipeline():
         logger.info("Search pipeline not found, nothing to delete.")
 
 
-def delete_test_indices():
-    """Drop all search index containing the 'test' word"""
-    opensearch.opensearch_client().indices.delete(index="*test*")
-
-
-def prepare_index(index_name, documents: List, cleanup=True):
+def prepare_index(index_name, documents: List):
     """Prepare the search index before testing a query on it."""
-    if cleanup:
-        delete_test_indices()
-
     opensearch.ensure_index_exists(index_name)
 
     # Index new documents

src/backend/core/views.py

@@ -13,7 +13,7 @@ from rest_framework.response import Response
 
 from . import schemas
 from .authentication import ServiceTokenAuthentication
-from .models import Service
+from .models import Service, get_opensearch_index_name
 from .permissions import IsAuthAuthenticated
 from .services.opensearch import (
     check_hybrid_search_enabled,
@@ -86,7 +86,7 @@ class IndexDocumentView(views.APIView):
             - Returns a list of results for all documents, with details of success and indexing
               errors.
         """
-        index_name = request.auth.name
+        index_name = request.auth.index_name
         opensearch_client_ = opensearch_client()
 
         if isinstance(request.data, list):
@@ -191,7 +191,7 @@ class SearchDocumentView(ResourceServerMixin, views.APIView):
             if len(available) < len(services):
                 raise SuspiciousOperation("Some requested services are not available")
 
-        return allowed_services
+        return [get_opensearch_index_name(name) for name in allowed_services]
 
     def post(self, request, *args, **kwargs):
         """

src/backend/find/settings.py

@@ -243,6 +243,9 @@ class Base(Configuration):
     OPENSEARCH_USE_SSL = values.BooleanValue(
         default=True, environ_name="OPENSEARCH_USE_SSL", environ_prefix=None
     )
+    OPENSEARCH_INDEX_PREFIX = values.Value(
+        default="find", environ_name="OPENSEARCH_INDEX_PREFIX", environ_prefix=None
+    )
 
     SPECTACULAR_SETTINGS = {
         "TITLE": "Find API",