Merge pull request #3323 from grafana/dev

Dev to main

CHANGELOG.md

@@ -5,6 +5,12 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## v1.3.57 (2023-11-10)
+
+### Fixed
+
+- Fix AmazonSNS integration to handle exceptions the same as other integrations @mderynck ([#3315](https://github.com/grafana/oncall/pull/3315))
+
 ## v1.3.56 (2023-11-10)
 
 ## v1.3.55 (2023-11-07)

engine/apps/integrations/views.py

@@ -21,6 +21,7 @@ from apps.integrations.mixins import (
     is_ratelimit_ignored,
 )
 from apps.integrations.tasks import create_alert, create_alertmanager_alerts
+from apps.user_management.exceptions import OrganizationDeletedException, OrganizationMovedException
 from common.api_helpers.utils import create_engine_url
 
 logger = logging.getLogger(__name__)
@@ -31,8 +32,10 @@ class AmazonSNS(BrowsableInstructionMixin, AlertChannelDefiningMixin, Integratio
     def dispatch(self, *args, **kwargs):
         try:
             return super().dispatch(*args, **kwargs)
+        except (OrganizationMovedException, OrganizationDeletedException, PermissionDenied) as oe:
+            raise oe
         except Exception as e:
-            print(e)
+            logger.error(f"AmazonSNS - Bad Request (400) {str(e)}")
             return JsonResponse(status=400, data={})
 
     def handle_message(self, message, payload):

engine/apps/user_management/tests/test_region.py

@@ -9,7 +9,7 @@ from rest_framework.test import APIClient
 from apps.alerts.models import AlertReceiveChannel
 from apps.auth_token.auth import ApiTokenAuthentication, ScheduleExportAuthentication, UserScheduleExportAuthentication
 from apps.auth_token.models import ScheduleExportAuthToken, UserScheduleExportAuthToken
-from apps.integrations.views import AlertManagerAPIView
+from apps.integrations.views import AlertManagerAPIView, AmazonSNS
 from apps.schedules.models import OnCallScheduleWeb
 from apps.user_management.exceptions import OrganizationMovedException
 
@@ -30,19 +30,27 @@ def test_organization_region_delete(
 
 
 @pytest.mark.django_db
+@pytest.mark.parametrize(
+    "integration_type,integration_view",
+    [
+        (AlertReceiveChannel.INTEGRATION_ALERTMANAGER, AlertManagerAPIView()),
+        ("amazon_sns", AmazonSNS()),
+    ],
+)
 def test_integration_does_not_raise_exception_organization_moved(
     make_organization,
     make_alert_receive_channel,
+    integration_type,
+    integration_view,
 ):
     organization = make_organization()
     alert_receive_channel = make_alert_receive_channel(
         organization=organization,
-        integration=AlertReceiveChannel.INTEGRATION_ALERTMANAGER,
+        integration=integration_type,
     )
 
     try:
-        am = AlertManagerAPIView()
-        am.dispatch(alert_channel_key=alert_receive_channel.token)
+        integration_view.dispatch(alert_channel_key=alert_receive_channel.token)
         assert False
     except OrganizationMovedException:
         assert False
@@ -51,21 +59,29 @@ def test_integration_does_not_raise_exception_organization_moved(
 
 
 @pytest.mark.django_db
+@pytest.mark.parametrize(
+    "integration_type,integration_view",
+    [
+        (AlertReceiveChannel.INTEGRATION_ALERTMANAGER, AlertManagerAPIView()),
+        ("amazon_sns", AmazonSNS()),
+    ],
+)
 def test_integration_raises_exception_organization_moved(
     make_organization_and_region,
     make_alert_receive_channel,
+    integration_type,
+    integration_view,
 ):
     organization, region = make_organization_and_region()
     organization.save()
 
     alert_receive_channel = make_alert_receive_channel(
         organization=organization,
-        integration=AlertReceiveChannel.INTEGRATION_ALERTMANAGER,
+        integration=integration_type,
     )
 
     try:
-        am = AlertManagerAPIView()
-        am.dispatch(alert_channel_key=alert_receive_channel.token)
+        integration_view.dispatch(alert_channel_key=alert_receive_channel.token)
         assert False
     except OrganizationMovedException as e:
         assert e.organization == organization
@@ -73,24 +89,29 @@ def test_integration_raises_exception_organization_moved(
 
 @patch("apps.user_management.middlewares.OrganizationMovedMiddleware.make_request")
 @pytest.mark.django_db
+@pytest.mark.parametrize(
+    "integration_type",
+    [
+        AlertReceiveChannel.INTEGRATION_ALERTMANAGER,
+        "amazon_sns",
+    ],
+)
 def test_organization_moved_middleware(
-    mocked_make_request,
-    make_organization_and_region,
-    make_alert_receive_channel,
+    mocked_make_request, make_organization_and_region, make_alert_receive_channel, integration_type
 ):
     organization, region = make_organization_and_region()
     organization.save()
 
     alert_receive_channel = make_alert_receive_channel(
         organization=organization,
-        integration=AlertReceiveChannel.INTEGRATION_ALERTMANAGER,
+        integration=integration_type,
     )
 
     expected_message = bytes(f"Redirected to {region.oncall_backend_url}", "utf-8")
     mocked_make_request.return_value = HttpResponse(expected_message, status=status.HTTP_200_OK)
 
     client = APIClient()
-    url = reverse("integrations:alertmanager", kwargs={"alert_channel_key": alert_receive_channel.token})
+    url = reverse(f"integrations:{integration_type}", kwargs={"alert_channel_key": alert_receive_channel.token})
 
     data = {"value": "test"}
     response = client.post(url, data, format="json")

engine/engine/management/commands/restart_acknowledge_reminder.py

@@ -0,0 +1,73 @@
+from celery import uuid as celery_uuid
+from django.core.management import BaseCommand
+
+from apps.alerts.models import AlertGroup
+from apps.alerts.tasks import acknowledge_reminder_task
+from apps.user_management.models import Organization
+
+
+class Command(BaseCommand):
+    """
+    Restart acknowledge_reminder_task for organization. Used for migrated organizations.
+
+    Usage example:
+    `python manage.py restart_acknowledge_reminder -ppk "organization ppk"` - restart task for alert groups from
+    organizations with this public pk
+    """
+
+    def add_arguments(self, parser):
+        group = parser.add_mutually_exclusive_group(required=True)
+
+        group.add_argument(
+            "-ppk", "--organization_ppk", type=str, help="Organization public pks to restart reminder for."
+        )
+
+    def handle(self, *args, **options):
+        organization_ppk = options["organization_ppk"]
+        organization = Organization.objects.get(public_primary_key=organization_ppk)
+        self.stdout.write(
+            f"Processing restart acknowledge reminder for alert groups from organization "
+            f"(id: {organization.id}, ppk: {organization.public_primary_key})..."
+        )
+        if organization.acknowledge_remind_timeout == 0:
+            self.stdout.write("Organization doesn't have acknowledge reminder setting set")
+            return
+
+        alert_groups = AlertGroup.objects.filter(
+            acknowledged=True,
+            resolved=False,
+            silenced=False,
+            maintenance_uuid__isnull=True,
+            root_alert_group=None,
+            channel__organization=organization,
+        )
+        if not alert_groups:
+            self.stdout.write("No affected alert groups.")
+            return
+
+        self.stdout.write(f"Affected alert groups count: {alert_groups.count()}.")
+
+        tasks = []
+        alert_groups_to_update = []
+
+        for alert_group in alert_groups:
+            task_id = celery_uuid()
+            countdown = Organization.ACKNOWLEDGE_REMIND_DELAY[organization.acknowledge_remind_timeout]
+            alert_group.last_unique_unacknowledge_process_id = task_id
+            alert_groups_to_update.append(alert_group)
+            tasks.append(
+                acknowledge_reminder_task.signature(
+                    args=(alert_group.pk, task_id), immutable=True, task_id=task_id, countdown=countdown
+                )
+            )
+
+        AlertGroup.objects.bulk_update(
+            alert_groups_to_update,
+            ["last_unique_unacknowledge_process_id"],
+            batch_size=5000,
+        )
+
+        for task in tasks:
+            task.apply_async()
+
+        self.stdout.write("Acknowledge reminder has been restarted for affected alert groups")