perf(api): avoid N+1 query loading finding resource tags (#11426)

Co-authored-by: Davidm4r <david.copo@gmail.com>
Co-authored-by: Pepe Fagoaga <pepe@prowler.com>

.env

@@ -145,7 +145,7 @@ SENTRY_RELEASE=local
 NEXT_PUBLIC_SENTRY_ENVIRONMENT=${SENTRY_ENVIRONMENT}
 
 #### Prowler release version ####
-NEXT_PUBLIC_PROWLER_RELEASE_VERSION=v5.29.0
+NEXT_PUBLIC_PROWLER_RELEASE_VERSION=v5.29.1
 
 # Social login credentials
 SOCIAL_GOOGLE_OAUTH_CALLBACK_URL="${AUTH_URL}/api/auth/callback/google"

api/CHANGELOG.md

@@ -2,6 +2,15 @@
 
 All notable changes to the **Prowler API** are documented in this file.
 
+## [1.30.1] (Prowler v5.29.1)
+
+### 🐞 Fixed
+
+- `GET /api/v1/findings` N+1 query loading `resources__tags` when listing findings [(#11420)](https://github.com/prowler-cloud/prowler/pull/11420)
+- Clean up the scan tmp output directory when `scan-report` fails so partial files do not accumulate and fill the worker disk (`No space left on device`) [(#11421)](https://github.com/prowler-cloud/prowler/pull/11421)
+
+---
+
 ## [1.30.0] (Prowler v5.29.0)
 
 ### 🔄 Changed

api/pyproject.toml

@@ -43,7 +43,7 @@ dependencies = [
   "defusedxml==0.7.1",
   "gunicorn==23.0.0",
   "lxml==6.1.0",
-  "prowler @ git+https://github.com/prowler-cloud/prowler.git@master",
+  "prowler @ git+https://github.com/prowler-cloud/prowler.git@v5.29",
   "psycopg2-binary==2.9.9",
   "pytest-celery[redis] (==1.3.0)",
   "sentry-sdk[django] (==2.56.0)",
@@ -68,7 +68,7 @@ name = "prowler-api"
 package-mode = false
 # Needed for the SDK compatibility
 requires-python = ">=3.11,<3.13"
-version = "1.30.0"
+version = "1.30.1"
 
 [tool.uv]
 # Transitive pins matching master to avoid silent drift; bump deliberately.

api/src/backend/api/specs/v1.yaml

@@ -1,7 +1,7 @@
 openapi: 3.0.3
 info:
   title: Prowler API
-  version: 1.30.0
+  version: 1.30.1
   description: |-
     Prowler API specification.
 

api/src/backend/api/tests/test_views.py

@@ -24,9 +24,11 @@ from conftest import (
     today_after_n_days,
 )
 from django.conf import settings
+from django.db import connection
 from django.db.models import Count
 from django.http import JsonResponse
 from django.test import RequestFactory
+from django.test.utils import CaptureQueriesContext
 from django.urls import reverse
 from django_celery_results.models import TaskResult
 from rest_framework import status
@@ -64,6 +66,7 @@ from api.models import (
     ProviderSecret,
     Resource,
     ResourceFindingMapping,
+    ResourceTag,
     Role,
     RoleProviderGroupRelationship,
     SAMLConfiguration,
@@ -3856,16 +3859,20 @@ class TestScanViewSet:
         scan.output_location = "dummy"
         scan.save()
 
-        dummy_task = Task.objects.create(tenant_id=scan.tenant_id)
-        dummy_task.id = "dummy-task-id"
-        dummy_task_data = {"id": dummy_task.id, "state": StateChoices.EXECUTING}
+        task_result = TaskResult.objects.create(
+            task_id=str(uuid4()),
+            task_name="scan-report",
+            task_kwargs={"scan_id": str(scan.id)},
+        )
+        task = Task.objects.create(
+            tenant_id=scan.tenant_id,
+            task_runner_task=task_result,
+        )
+        dummy_task_data = {"id": str(task.id), "state": StateChoices.EXECUTING}
 
-        with (
-            patch("api.v1.views.Task.objects.get", return_value=dummy_task),
-            patch(
-                "api.v1.views.TaskSerializer",
-                return_value=type("DummySerializer", (), {"data": dummy_task_data}),
-            ),
+        with patch(
+            "api.v1.views.TaskSerializer",
+            return_value=type("DummySerializer", (), {"data": dummy_task_data}),
         ):
             url = reverse("scan-report", kwargs={"pk": scan.id})
             response = authenticated_client.get(url)
@@ -4186,6 +4193,88 @@ class TestScanViewSet:
         assert resp.status_code == status.HTTP_302_FOUND
         assert resp["Location"] == presigned_url
 
+    def test_compliance_s3_returns_latest_match(
+        self, authenticated_client, scans_fixture, monkeypatch
+    ):
+        """When several files match, the most recently modified one is served."""
+        scan = scans_fixture[0]
+        bucket = "bucket"
+        scan.output_location = f"s3://{bucket}/path/scan.zip"
+        scan.state = StateChoices.COMPLETED
+        scan.save()
+
+        monkeypatch.setattr(
+            "api.v1.views.env",
+            type("env", (), {"str": lambda self, *args, **kwargs: "test-bucket"})(),
+        )
+
+        old_key = "path/compliance/prowler-output-aws-20240101000000_cis_1.4_aws.csv"
+        latest_key = "path/compliance/prowler-output-aws-20240202000000_cis_1.4_aws.csv"
+
+        class FakeS3Client:
+            def list_objects_v2(self, Bucket, Prefix):
+                return {
+                    "Contents": [
+                        {
+                            "Key": old_key,
+                            "LastModified": datetime(2024, 1, 1, tzinfo=timezone.utc),
+                        },
+                        {
+                            "Key": latest_key,
+                            "LastModified": datetime(2024, 2, 2, tzinfo=timezone.utc),
+                        },
+                    ]
+                }
+
+            def generate_presigned_url(self, ClientMethod, Params, ExpiresIn):
+                assert Params["Key"] == latest_key
+                return "https://test-bucket.s3.amazonaws.com/latest"
+
+        monkeypatch.setattr("api.v1.views.get_s3_client", lambda: FakeS3Client())
+
+        url = reverse("scan-compliance", kwargs={"pk": scan.id, "name": "cis_1.4_aws"})
+        resp = authenticated_client.get(url)
+        assert resp.status_code == status.HTTP_302_FOUND
+        assert resp["Location"].endswith("/latest")
+
+    def test_compliance_local_returns_latest_match(
+        self, authenticated_client, scans_fixture, monkeypatch
+    ):
+        """The local branch serves the most recently modified matching file."""
+        scan = scans_fixture[0]
+        scan.state = StateChoices.COMPLETED
+
+        with tempfile.TemporaryDirectory() as tmp:
+            comp_dir = Path(tmp) / "reports" / "compliance"
+            comp_dir.mkdir(parents=True, exist_ok=True)
+
+            old_file = comp_dir / "prowler-output-aws-20240101000000_cis_1.4_aws.csv"
+            old_file.write_bytes(b"old")
+            latest_file = comp_dir / "prowler-output-aws-20240202000000_cis_1.4_aws.csv"
+            latest_file.write_bytes(b"latest")
+            # Make `latest_file` newer regardless of creation order.
+            os.utime(old_file, (1_700_000_000, 1_700_000_000))
+            os.utime(latest_file, (1_700_000_100, 1_700_000_100))
+
+            scan.output_location = str(Path(tmp) / "reports" / "scan.zip")
+            scan.save()
+
+            monkeypatch.setattr(
+                glob,
+                "glob",
+                lambda p: [str(old_file), str(latest_file)],
+            )
+
+            url = reverse(
+                "scan-compliance", kwargs={"pk": scan.id, "name": "cis_1.4_aws"}
+            )
+            resp = authenticated_client.get(url)
+            assert resp.status_code == status.HTTP_200_OK
+            assert resp.content == b"latest"
+            assert resp["Content-Disposition"].endswith(
+                f'filename="{latest_file.name}"'
+            )
+
     def test_compliance_s3_not_found(
         self, authenticated_client, scans_fixture, monkeypatch
     ):
@@ -4294,18 +4383,24 @@ class TestScanViewSet:
             assert cd.startswith('attachment; filename="')
             assert cd.endswith(f'filename="{fname.name}"')
 
-    @patch("api.v1.views.Task.objects.get")
     @patch("api.v1.views.TaskSerializer")
     def test__get_task_status_returns_none_if_task_not_executing(
-        self, mock_task_serializer, mock_task_get, authenticated_client, scans_fixture
+        self, mock_task_serializer, authenticated_client, scans_fixture
     ):
         scan = scans_fixture[0]
         scan.state = StateChoices.COMPLETED
         scan.output_location = "dummy"
         scan.save()
 
-        task = Task.objects.create(tenant_id=scan.tenant_id)
-        mock_task_get.return_value = task
+        task_result = TaskResult.objects.create(
+            task_id=str(uuid4()),
+            task_name="scan-report",
+            task_kwargs={"scan_id": str(scan.id)},
+        )
+        task = Task.objects.create(
+            tenant_id=scan.tenant_id,
+            task_runner_task=task_result,
+        )
         mock_task_serializer.return_value.data = {
             "id": str(task.id),
             "state": StateChoices.COMPLETED,
@@ -4326,6 +4421,7 @@ class TestScanViewSet:
         scan.save()
 
         task_result = TaskResult.objects.create(
+            task_id=str(uuid4()),
             task_name="scan-report",
             task_kwargs={"scan_id": str(scan.id)},
         )
@@ -4346,6 +4442,51 @@ class TestScanViewSet:
         assert response.status_code == status.HTTP_202_ACCEPTED
         assert response.data["id"] == str(task.id)
 
+    @patch("api.v1.views.TaskSerializer")
+    def test__get_task_status_returns_latest_task(
+        self, mock_task_serializer, authenticated_client, scans_fixture
+    ):
+        """With several scan-report tasks for the scan, the most recent is used."""
+        scan = scans_fixture[0]
+        scan.state = StateChoices.COMPLETED
+        scan.output_location = "dummy"
+        scan.save()
+
+        old_task = Task.objects.create(
+            tenant_id=scan.tenant_id,
+            task_runner_task=TaskResult.objects.create(
+                task_id=str(uuid4()),
+                task_name="scan-report",
+                task_kwargs={"scan_id": str(scan.id)},
+            ),
+        )
+        new_task = Task.objects.create(
+            tenant_id=scan.tenant_id,
+            task_runner_task=TaskResult.objects.create(
+                task_id=str(uuid4()),
+                task_name="scan-report",
+                task_kwargs={"scan_id": str(scan.id)},
+            ),
+        )
+        # `inserted_at` is `auto_now_add`, and within the test transaction the DB
+        # `now()` is constant, so force distinct timestamps to make order_by stable.
+        base = datetime(2024, 1, 1, tzinfo=timezone.utc)
+        Task.objects.filter(pk=old_task.pk).update(inserted_at=base)
+        Task.objects.filter(pk=new_task.pk).update(
+            inserted_at=base + timedelta(hours=1)
+        )
+
+        mock_task_serializer.side_effect = lambda instance, *a, **k: SimpleNamespace(
+            data={"id": str(instance.id), "state": StateChoices.EXECUTING}
+        )
+
+        url = reverse("scan-report", kwargs={"pk": scan.id})
+        response = authenticated_client.get(url)
+
+        assert response.status_code == status.HTTP_202_ACCEPTED
+        assert str(new_task.id) in response["Content-Location"]
+        assert str(old_task.id) not in response["Content-Location"]
+
     @patch("api.v1.views.get_s3_client")
     @patch("api.v1.views.sentry_sdk.capture_exception")
     def test_compliance_list_objects_client_error(
@@ -6916,6 +7057,80 @@ class TestFindingViewSet:
             == findings_fixture[0].status
         )
 
+    def test_findings_list_resource_tags_no_n_plus_one(
+        self, authenticated_client, findings_fixture
+    ):
+        """Listing findings must load every resource's tags in a constant
+        number of queries, no matter how many findings/resources are returned.
+
+        This guards ``FindingViewSet._optimize_tags_loading`` against
+        regressions that would reintroduce one extra query per resource (the
+        N+1 the prefetch was added to remove).
+        """
+        scan = findings_fixture[0].scan
+        tenant_id = findings_fixture[0].tenant_id
+        provider = scan.provider
+
+        def _create_finding_with_tagged_resource(index):
+            resource = Resource.objects.create(
+                tenant_id=tenant_id,
+                provider=provider,
+                uid=f"arn:aws:ec2:us-east-1:123456789012:instance/n-plus-one-{index}",
+                name=f"N+1 Instance {index}",
+                region="us-east-1",
+                service="ec2",
+                type="prowler-test",
+            )
+            resource.upsert_or_delete_tags(
+                [
+                    ResourceTag.objects.create(
+                        tenant_id=tenant_id,
+                        key=f"key-{index}",
+                        value=f"value-{index}",
+                    )
+                ]
+            )
+            finding = Finding.objects.create(
+                tenant_id=tenant_id,
+                uid=f"n_plus_one_finding_{index}",
+                scan=scan,
+                status=Status.FAIL,
+                status_extended="n+1 status",
+                impact=Severity.medium,
+                severity=Severity.medium,
+                check_id="test_check_id",
+                check_metadata={"CheckId": "test_check_id", "servicename": "ec2"},
+                first_seen_at="2024-01-02T00:00:00Z",
+            )
+            finding.add_resources([resource])
+            return finding
+
+        params = {"filter[inserted_at]": TODAY, "include": "resources"}
+
+        # Baseline: the two findings provided by the fixture.
+        with CaptureQueriesContext(connection) as baseline:
+            response = authenticated_client.get(reverse("finding-list"), params)
+        assert response.status_code == status.HTTP_200_OK
+
+        # Add more findings, each with its own resource carrying tags.
+        extra_findings = 5
+        for index in range(extra_findings):
+            _create_finding_with_tagged_resource(index)
+
+        with CaptureQueriesContext(connection) as scaled:
+            response = authenticated_client.get(reverse("finding-list"), params)
+        assert response.status_code == status.HTTP_200_OK
+        assert len(response.json()["data"]) == len(findings_fixture) + extra_findings
+
+        # The query count must not grow with the number of findings/resources.
+        assert len(scaled.captured_queries) == len(baseline.captured_queries), (
+            "Resource tags are not being prefetched: "
+            f"{len(baseline.captured_queries)} queries for {len(findings_fixture)} "
+            f"findings vs {len(scaled.captured_queries)} for "
+            f"{len(findings_fixture) + extra_findings}. Likely an N+1 regression "
+            "in FindingViewSet._optimize_tags_loading."
+        )
+
     @pytest.mark.parametrize(
         "include_values, expected_resources",
         [

api/src/backend/api/v1/views.py

@@ -2059,12 +2059,17 @@ class ScanViewSet(BaseRLSViewSet):
         if scan_instance.state == StateChoices.EXECUTING and scan_instance.task:
             task = scan_instance.task
         else:
-            try:
-                task = Task.objects.get(
+            # A scan can have several `scan-report` tasks (e.g. re-runs); take the
+            # most recent one. `.first()` also avoids `MultipleObjectsReturned`.
+            task = (
+                Task.objects.filter(
                     task_runner_task__task_name="scan-report",
                     task_runner_task__task_kwargs__contains=str(scan_instance.id),
                 )
-            except Task.DoesNotExist:
+                .order_by("-inserted_at")
+                .first()
+            )
+            if task is None:
                 return None
 
         self.response_serializer_class = TaskSerializer
@@ -2139,27 +2144,32 @@ class ScanViewSet(BaseRLSViewSet):
                         status=status.HTTP_502_BAD_GATEWAY,
                     )
                 contents = resp.get("Contents", [])
-                keys = []
+                matches = []
                 for obj in contents:
                     key = obj["Key"]
                     key_basename = os.path.basename(key)
                     if any(ch in suffix for ch in ("*", "?", "[")):
                         if fnmatch.fnmatch(key_basename, suffix):
-                            keys.append(key)
+                            matches.append(obj)
                     elif key_basename == suffix:
-                        keys.append(key)
+                        matches.append(obj)
                     elif key.endswith(suffix):
                         # Backward compatibility if suffix already includes directories
-                        keys.append(key)
-                if not keys:
+                        matches.append(obj)
+                if not matches:
                     return Response(
                         {
                             "detail": f"No compliance file found for name '{os.path.splitext(suffix)[0]}'."
                         },
                         status=status.HTTP_404_NOT_FOUND,
                     )
-                # path_pattern here is prefix, but in compliance we build correct suffix check before
-                key = keys[0]
+                # Return the most recently modified match (latest report) when
+                # several files share the prefix/suffix. `list_objects_v2` always
+                # returns `LastModified`; the fallback keeps ordering deterministic
+                # if it is ever absent.
+                key = max(matches, key=lambda o: (o.get("LastModified", ""), o["Key"]))[
+                    "Key"
+                ]
             else:
                 # path_pattern is exact key; HEAD before presigning to preserve the 404 contract.
                 key = path_pattern
@@ -2209,7 +2219,9 @@ class ScanViewSet(BaseRLSViewSet):
                     },
                     status=status.HTTP_404_NOT_FOUND,
                 )
-            filepath = files[0]
+            # Return the most recently modified match (latest report) when the
+            # pattern resolves to several files.
+            filepath = max(files, key=os.path.getmtime)
             with open(filepath, "rb") as f:
                 content = f.read()
             filename = os.path.basename(filepath)
@@ -3749,6 +3761,16 @@ class FindingViewSet(PaginateByPkMixin, BaseRLSViewSet):
             return queryset
         return super().filter_queryset(queryset)
 
+    def _optimize_tags_loading(self, queryset):
+        """Prefetch resource tags to avoid N+1 queries when serializing findings"""
+        return queryset.prefetch_related(
+            Prefetch(
+                "resources__tags",
+                queryset=ResourceTag.objects.filter(tenant_id=self.request.tenant_id),
+                to_attr="prefetched_tags",
+            )
+        )
+
     def list(self, request, *args, **kwargs):
         filtered_queryset = self.filter_queryset(self.get_queryset())
         return self.paginate_by_pk(

api/src/backend/tasks/tasks.py

@@ -467,8 +467,31 @@ def delete_tenant_task(tenant_id: str):
     return delete_tenant(pk=tenant_id)
 
 
+def _scan_tmp_output_directory(tenant_id: str, scan_id: str) -> Path:
+    """Root tmp output directory for a scan ({tmp}/{tenant_id}/{scan_id})."""
+    return Path(DJANGO_TMP_OUTPUT_DIRECTORY) / str(tenant_id) / str(scan_id)
+
+
+class ScanReportRLSTask(RLSTask):
+    """
+    RLS task that removes the scan's tmp output directory when the task fails.
+
+    Covers failures both inside and outside the task body (e.g. ENOSPC mid-write,
+    or setup errors) so partial artifacts do not accumulate on the worker disk.
+    """
+
+    def on_failure(self, exc, task_id, args, kwargs, _einfo):  # noqa: ARG002
+        del args  # Required by Celery's Task.on_failure signature; not used.
+        tenant_id = kwargs.get("tenant_id")
+        scan_id = kwargs.get("scan_id")
+
+        if tenant_id and scan_id:
+            logger.error(f"Scan report task {task_id} failed: {exc}")
+            rmtree(_scan_tmp_output_directory(tenant_id, scan_id), ignore_errors=True)
+
+
 @shared_task(
-    base=RLSTask,
+    base=ScanReportRLSTask,
     name="scan-report",
     queue="scan-reports",
 )
@@ -518,6 +541,9 @@ def generate_outputs_task(scan_id: str, provider_id: str, tenant_id: str):
     out_dir, comp_dir = _generate_output_directory(
         DJANGO_TMP_OUTPUT_DIRECTORY, provider_uid, tenant_id, scan_id
     )
+    # Removed on success here and on failure by ScanReportRLSTask.on_failure,
+    # so partial artifacts do not accumulate and fill the disk (ENOSPC).
+    scan_tmp_dir = _scan_tmp_output_directory(tenant_id, scan_id)
 
     def get_writer(writer_map, name, factory, is_last):
         """
@@ -666,7 +692,7 @@ def generate_outputs_task(scan_id: str, provider_id: str, tenant_id: str):
         # TODO: We need to create a new periodic task to delete the output files
         # This task shouldn't be responsible for deleting the output files
         try:
-            rmtree(Path(compressed).parent, ignore_errors=True)
+            rmtree(scan_tmp_dir, ignore_errors=True)
         except Exception as e:
             logger.error(f"Error deleting output files: {e}")
         final_location, did_upload = upload_uri, True

api/src/backend/tasks/tests/test_tasks.py

@@ -15,8 +15,10 @@ from tasks.jobs.lighthouse_providers import (
 from tasks.tasks import (
     DJANGO_TMP_OUTPUT_DIRECTORY,
     STALE_TMP_OUTPUT_MAX_AGE_HOURS,
+    ScanReportRLSTask,
     _cleanup_orphan_scheduled_scans,
     _perform_scan_complete_tasks,
+    _scan_tmp_output_directory,
     check_integrations_task,
     check_lighthouse_provider_connection_task,
     generate_outputs_task,
@@ -771,6 +773,38 @@ class TestGenerateOutputs:
             mock_s3_task.assert_called_once()
 
 
+class TestScanReportRLSTaskOnFailure:
+    def test_on_failure_removes_scan_tmp_directory(self):
+        task = ScanReportRLSTask()
+
+        with patch("tasks.tasks.rmtree") as mock_rmtree:
+            task.on_failure(
+                exc=OSError("No space left on device"),
+                task_id="task-abc",
+                args=(),
+                kwargs={"tenant_id": "t-1", "scan_id": "s-1"},
+                _einfo=None,
+            )
+
+        mock_rmtree.assert_called_once_with(
+            _scan_tmp_output_directory("t-1", "s-1"), ignore_errors=True
+        )
+
+    def test_on_failure_skips_when_missing_kwargs(self):
+        task = ScanReportRLSTask()
+
+        with patch("tasks.tasks.rmtree") as mock_rmtree:
+            task.on_failure(
+                exc=OSError("No space left on device"),
+                task_id="task-abc",
+                args=(),
+                kwargs={},
+                _einfo=None,
+            )
+
+        mock_rmtree.assert_not_called()
+
+
 class TestScanCompleteTasks:
     @patch("tasks.tasks.aggregate_attack_surface_task.apply_async")
     @patch("tasks.tasks.chain")

api/uv.lock

@@ -4410,8 +4410,8 @@ wheels = [
 
 [[package]]
 name = "prowler"
-version = "5.27.0"
-source = { git = "https://github.com/prowler-cloud/prowler.git?rev=master#0abbb7fc590eaf7de6ed354dd5a217bca261d2b0" }
+version = "5.29.0"
+source = { git = "https://github.com/prowler-cloud/prowler.git?rev=v5.29#a769e3761532d9332cb64078ef09ebf7ffb15292" }
 dependencies = [
     { name = "alibabacloud-actiontrail20200706" },
     { name = "alibabacloud-credentials" },
@@ -4484,9 +4484,13 @@ dependencies = [
     { name = "pygithub" },
     { name = "python-dateutil" },
     { name = "pytz" },
+    { name = "scaleway" },
     { name = "schema" },
     { name = "shodan" },
     { name = "slack-sdk" },
+    { name = "stackit-core" },
+    { name = "stackit-iaas" },
+    { name = "stackit-resourcemanager" },
     { name = "tabulate" },
     { name = "tzlocal" },
     { name = "uuid6" },
@@ -4494,7 +4498,7 @@ dependencies = [
 
 [[package]]
 name = "prowler-api"
-version = "1.30.0"
+version = "1.30.1"
 source = { virtual = "." }
 dependencies = [
     { name = "cartography" },
@@ -4590,7 +4594,7 @@ requires-dist = [
     { name = "matplotlib", specifier = "==3.10.8" },
     { name = "neo4j", specifier = "==6.1.0" },
     { name = "openai", specifier = "==1.109.1" },
-    { name = "prowler", git = "https://github.com/prowler-cloud/prowler.git?rev=master" },
+    { name = "prowler", git = "https://github.com/prowler-cloud/prowler.git?rev=v5.29" },
     { name = "psycopg2-binary", specifier = "==2.9.9" },
     { name = "pytest-celery", extras = ["redis"], specifier = "==1.3.0" },
     { name = "reportlab", specifier = "==4.4.10" },
@@ -5526,6 +5530,52 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/49/4b/359f28a903c13438ef59ebeee215fb25da53066db67b305c125f1c6d2a25/sqlparse-0.5.5-py3-none-any.whl", hash = "sha256:12a08b3bf3eec877c519589833aed092e2444e68240a3577e8e26148acc7b1ba", size = 46138, upload-time = "2025-12-19T07:17:46.573Z" },
 ]
 
+[[package]]
+name = "stackit-core"
+version = "0.2.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "cryptography" },
+    { name = "pydantic" },
+    { name = "pyjwt" },
+    { name = "requests" },
+    { name = "urllib3" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/24/90/20f9ec7387eec4067cfd3d29055d0e2b5e1e0322c601a7f48125fd8ea35f/stackit_core-0.2.0.tar.gz", hash = "sha256:b8af91877cdb060d6969a303d8cf20bc0b33b345afd91f679c44a987381e2d47", size = 8987, upload-time = "2025-06-12T08:24:45.251Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ab/b4/7b53187ce68956870d864ccb9ccfb68066c9df9de1c9568fd2feb03c4504/stackit_core-0.2.0-py3-none-any.whl", hash = "sha256:04632fc6742790d08ddfcb7f2313e04d1254827397a80250f838a2f81b92645b", size = 10240, upload-time = "2025-06-12T08:24:44.214Z" },
+]
+
+[[package]]
+name = "stackit-iaas"
+version = "1.4.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pydantic" },
+    { name = "python-dateutil" },
+    { name = "requests" },
+    { name = "stackit-core" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/52/07/24e65278300d5c3cb19cb1660bff924c80812cf8aad3e715f826bae5aa80/stackit_iaas-1.4.0.tar.gz", hash = "sha256:93523b23442350c7ebefd9129485c4c2a539f694a9c36a0f8edfaba9862057ea", size = 116236, upload-time = "2026-05-13T09:43:15.996Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/08/51/2201164d7bfacf47539888c735f10f6320c188252384957aa1b23121a210/stackit_iaas-1.4.0-py3-none-any.whl", hash = "sha256:3f4a32321b57ac238f73e5d660c6428186b92cc0425c1f0783ba801e377149d9", size = 316588, upload-time = "2026-05-13T09:43:14.943Z" },
+]
+
+[[package]]
+name = "stackit-resourcemanager"
+version = "0.8.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pydantic" },
+    { name = "python-dateutil" },
+    { name = "requests" },
+    { name = "stackit-core" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/23/2d/f458f18e48ed2b1c83df52cff7dbdfd5dd904fb2980ffd9385876e47bbd9/stackit_resourcemanager-0.8.0.tar.gz", hash = "sha256:f44542beab4130857f5a7f465cf02defeef657bdf63c1beeb3102f0ba3c003fe", size = 33943, upload-time = "2026-05-13T09:43:08.667Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c7/9c/38a74d0f7a89b4320f6d2366fb660638bda8860daa08748b12c713d84381/stackit_resourcemanager-0.8.0-py3-none-any.whl", hash = "sha256:dd04bb8353d041a137c4dcba190beabded7acfaff1bc98b218fce20a99389ebc", size = 81288, upload-time = "2026-05-13T09:43:07.81Z" },
+]
+
 [[package]]
 name = "statsd"
 version = "4.0.1"

prowler/CHANGELOG.md

@@ -2,6 +2,14 @@
 
 All notable changes to the **Prowler SDK** are documented in this file.
 
+## [5.29.1] (Prowler v5.29.1)
+
+### 🐞 Fixed
+
+- OCSF output writer now re-raises I/O errors (e.g. `ENOSPC`) instead of logging them per finding and leaving a truncated file [(#11421)](https://github.com/prowler-cloud/prowler/pull/11421)
+
+---
+
 ## [5.29.0] (Prowler v5.29.0)
 
 ### 🚀 Added

prowler/config/config.py

@@ -48,7 +48,7 @@ class _MutableTimestamp:
 
 timestamp = _MutableTimestamp(datetime.today())
 timestamp_utc = _MutableTimestamp(datetime.now(timezone.utc))
-prowler_version = "5.29.0"
+prowler_version = "5.29.1"
 html_logo_url = "https://github.com/prowler-cloud/prowler/"
 square_logo_img = "https://raw.githubusercontent.com/prowler-cloud/prowler/dc7d2d5aeb92fdf12e8604f42ef6472cd3e8e889/docs/img/prowler-logo-black.png"
 aws_logo = "https://user-images.githubusercontent.com/38561120/235953920-3e3fba08-0795-41dc-b480-9bea57db9f2e.png"

prowler/lib/outputs/ocsf/ocsf.py

@@ -227,6 +227,10 @@ class OCSF(Output):
                             json_output = finding.json(exclude_none=True, indent=4)
                         self._file_descriptor.write(json_output)
                         self._file_descriptor.write(",")
+                    except OSError:
+                        # I/O errors (e.g. ENOSPC) are not recoverable per finding:
+                        # fail fast instead of logging once per finding.
+                        raise
                     except Exception as error:
                         logger.error(
                             f"{error.__class__.__name__}[{error.__traceback__.tb_lineno}]: {error}"
@@ -239,6 +243,10 @@ class OCSF(Output):
                     self._file_descriptor.truncate()
                     self._file_descriptor.write("]")
                     self._file_descriptor.close()
+        except OSError:
+            # Propagate unrecoverable I/O errors (e.g. ENOSPC) so the caller can
+            # fail fast instead of producing a corrupt output file.
+            raise
         except Exception as error:
             logger.error(
                 f"{error.__class__.__name__}[{error.__traceback__.tb_lineno}]: {error}"

pyproject.toml

@@ -123,7 +123,7 @@ maintainers = [{name = "Prowler Engineering", email = "engineering@prowler.com"}
 name = "prowler"
 readme = "README.md"
 requires-python = ">=3.10,<3.13"
-version = "5.29.0"
+version = "5.29.1"
 
 [project.scripts]
 prowler = "prowler.__main__:prowler"

tests/lib/outputs/ocsf/ocsf_test.py

@@ -2,8 +2,10 @@ import json
 from datetime import datetime, timezone
 from io import StringIO
 from typing import Optional
+from unittest.mock import MagicMock
 from uuid import UUID
 
+import pytest
 import requests
 from freezegun import freeze_time
 from mock import patch
@@ -300,6 +302,36 @@ class TestOCSF:
     def test_batch_write_data_to_file_without_findings(self):
         assert not OCSF([])._file_descriptor
 
+    def test_batch_write_data_to_file_propagates_oserror(self):
+        """An I/O error (e.g. ENOSPC) while writing a finding must propagate
+        instead of being swallowed, so the caller can fail fast."""
+        findings = [
+            generate_finding_output(
+                status="FAIL",
+                severity="low",
+                muted=False,
+                region=AWS_REGION_EU_WEST_1,
+                timestamp=datetime.now(),
+                resource_details="resource_details",
+                resource_name="resource_name",
+                resource_uid="resource-id",
+                status_extended="status extended",
+            )
+        ]
+
+        output = OCSF(findings)
+        mock_file = MagicMock()
+        mock_file.closed = False
+        # Non-zero so the "[" prelude is skipped and the failure happens on the
+        # per-finding write, the exact path that hit ENOSPC in production.
+        mock_file.tell.return_value = 1
+        mock_file.write.side_effect = OSError(28, "No space left on device")
+        output._file_descriptor = mock_file
+
+        with pytest.raises(OSError) as excinfo:
+            output.batch_write_data_to_file()
+        assert excinfo.value.errno == 28
+
     def test_finding_output_cloud_pass_low_muted(self):
         finding_output = generate_finding_output(
             status="PASS",

uv.lock

@@ -3241,7 +3241,7 @@ wheels = [
 
 [[package]]
 name = "prowler"
-version = "5.29.0"
+version = "5.29.1"
 source = { editable = "." }
 dependencies = [
     { name = "alibabacloud-actiontrail20200706" },