v1.9.3

Author: joeyorlando

.github/workflows/e2e-tests.yml

@@ -144,12 +144,14 @@ jobs:
           git clone https://x-access-token:${{ steps.generate-token.outputs.token }}@github.com/grafana/ops-devenv.git
           git clone https://x-access-token:${{ steps.generate-token.outputs.token }}@github.com/grafana/gops-labels.git
 
-      - name: Tilt CI - standard and expensive E2E tests
+      - name: Tilt CI - Expensive E2E tests
         if: inputs.run-expensive-tests
         shell: bash
         env:
           E2E_TESTS_CMD: "cd ../../grafana-plugin && yarn test:e2e-expensive"
           GRAFANA_VERSION: ${{ inputs.grafana_version }}
+          GF_FEATURE_TOGGLES_ENABLE: "externalServiceAccounts"
+          ONCALL_API_URL: "http://oncall-dev-engine:8080"
           GRAFANA_ADMIN_USERNAME: "irm"
           GRAFANA_ADMIN_PASSWORD: "irm"
           BROWSERS: ${{ inputs.browsers }}

.github/workflows/expensive-e2e-tests.yml

@@ -24,8 +24,8 @@ jobs:
           # - 9.3.16
           # - 9.4.13
           # - 9.5.7
-          - 10.0.11
           - 10.1.7
+          - 10.3.3
           # TODO: fix issues with running e2e tests against Grafana v10.2.x and v10.3.x
           # - 10.2.4
           # - latest
@@ -55,7 +55,7 @@ jobs:
       #
       - uses: slackapi/[email protected]
         with:
-          channel-id: gops-oncall-dev
+          channel-id: gops-irm-dev
           # yamllint disable rule:line-length
           payload: |
             {

.github/workflows/linting-and-tests.yml

@@ -239,9 +239,16 @@ jobs:
   end-to-end-tests:
     name: Standard e2e tests
     uses: ./.github/workflows/e2e-tests.yml
+    strategy:
+      matrix:
+        grafana_version:
+          - 10.1.7
+          - 10.3.3
+          # TODO: fix issues with running e2e tests against Grafana v10.2.x and latest
+          # - 10.2.4
+          # - latest
+      fail-fast: false
     with:
-      # TODO: fix issues with running e2e tests against Grafana v10.2.x and v10.3.x
-      grafana_version: 10.1.7
-      # grafana_version: 10.3.3
+      grafana_version: ${{ matrix.grafana_version }}
       run-expensive-tests: false
       browsers: "chromium"

Tiltfile

@@ -1,6 +1,7 @@
 load('ext://uibutton', 'cmd_button', 'location', 'text_input', 'bool_input')
 load("ext://configmap", "configmap_create")
 
+grafana_url = os.getenv("GRAFANA_URL", "http://grafana:3000")
 running_under_parent_tiltfile = os.getenv("TILT_PARENT", "false") == "true"
 twilio_values=[
     "oncall.twilio.accountSid=" + os.getenv("TWILIO_ACCOUNT_SID", ""),
@@ -29,6 +30,14 @@ def plugin_json():
         return plugin_file
     return 'NOT_A_PLUGIN'
 
+def extra_env():
+    return {
+        "GF_APP_URL": grafana_url,
+        "GF_SERVER_ROOT_URL": grafana_url, 
+        "GF_FEATURE_TOGGLES_ENABLE": "externalServiceAccounts",
+        "ONCALL_API_URL": "http://oncall-dev-engine:8080"
+    }
+
 
 allow_k8s_contexts(["kind-kind"])
 
@@ -83,8 +92,6 @@ def load_grafana():
     # The user/pass that you will login to Grafana with
     grafana_admin_user_pass = os.getenv("GRAFANA_ADMIN_USER_PASS", "oncall")
     grafana_version = os.getenv("GRAFANA_VERSION", "latest")
-    grafana_url = os.getenv("GRAFANA_URL", "http://grafana:3000")
-
 
     if 'plugin' in profiles:
         k8s_resource(

engine/apps/alerts/models/alert_group.py

@@ -201,7 +201,6 @@ class AlertGroup(AlertGroupSlackRenderingMixin, EscalationSnapshotMixin, models.
     resolved_by_user: typing.Optional["User"]
     root_alert_group: typing.Optional["AlertGroup"]
     silenced_by_user: typing.Optional["User"]
-    slack_log_message: typing.Optional["SlackMessage"]
     slack_messages: "RelatedManager['SlackMessage']"
     users: "RelatedManager['User']"
     labels: "RelatedManager['AlertGroupAssociatedLabel']"
@@ -396,6 +395,7 @@ def status(self) -> int:
         related_name="wiped_alert_groups",
     )
 
+    # TODO: drop this column in future release
     slack_log_message = models.OneToOneField(
         "slack.SlackMessage",
         on_delete=models.SET_NULL,

engine/apps/alerts/signals.py

@@ -36,10 +36,6 @@
     AlertGroupSlackRepresentative.on_alert_group_action_triggered,
 )
 
-alert_group_update_log_report_signal.connect(
-    AlertGroupSlackRepresentative.on_alert_group_update_log_report,
-)
-
 alert_group_update_resolution_note_signal.connect(
     AlertGroupSlackRepresentative.on_alert_group_update_resolution_note,
 )

engine/apps/slack/representatives/alert_group_representative.py

@@ -90,19 +90,7 @@ def on_alert_group_action_triggered_async(log_record_id):
     autoretry_for=(Exception,), retry_backoff=True, max_retries=1 if settings.DEBUG else None
 )
 def on_alert_group_update_log_report_async(alert_group_id):
-    from apps.alerts.models import AlertGroup
-
-    alert_group = AlertGroup.objects.get(pk=alert_group_id)
-    logger.debug(f"Start on_alert_group_update_log_report for alert_group {alert_group_id}")
-    organization = alert_group.channel.organization
-    if alert_group.slack_message and organization.slack_team_identity:
-        logger.debug(f"Process on_alert_group_update_log_report for alert_group {alert_group_id}")
-        UpdateLogReportMessageStep = ScenarioStep.get_step("distribute_alerts", "UpdateLogReportMessageStep")
-        step = UpdateLogReportMessageStep(organization.slack_team_identity, organization)
-        step.process_signal(alert_group)
-    else:
-        logger.debug(f"Drop on_alert_group_update_log_report for alert_group {alert_group_id}")
-    logger.debug(f"Finish on_alert_group_update_log_report for alert_group {alert_group_id}")
+    return "Deprecated, will be removed after queue cleanup"
 
 
 class AlertGroupSlackRepresentative(AlertGroupAbstractRepresentative):
@@ -173,32 +161,6 @@ def on_alert_group_action_triggered(cls, **kwargs):
             logger.debug(f"SLACK on_alert_group_action_triggered: async {log_record_id} {force_sync}")
             on_alert_group_action_triggered_async.apply_async((log_record_id,))
 
-    @classmethod
-    def on_alert_group_update_log_report(cls, **kwargs):
-        from apps.alerts.models import AlertGroup
-
-        alert_group = kwargs["alert_group"]
-
-        if isinstance(alert_group, AlertGroup):
-            alert_group_id = alert_group.pk
-        else:
-            alert_group_id = alert_group
-            try:
-                alert_group = AlertGroup.objects.get(pk=alert_group_id)
-            except AlertGroup.DoesNotExist as e:
-                logger.warning(f"SLACK update log report: alert group {alert_group_id} has been deleted")
-                raise e
-
-        logger.debug(
-            f"Received alert_group_update_log_report signal in SLACK representative for alert_group {alert_group_id}"
-        )
-
-        if alert_group.notify_in_slack_enabled is False:
-            logger.debug(f"Skipping alert_group {alert_group_id} since notify_in_slack is disabled")
-            return
-
-        on_alert_group_update_log_report_async.apply_async((alert_group_id,))
-
     @classmethod
     def on_alert_group_update_resolution_note(cls, **kwargs):
         alert_group = kwargs["alert_group"]

engine/apps/slack/scenarios/distribute_alerts.py

@@ -4,7 +4,6 @@
 from datetime import datetime
 
 from django.core.cache import cache
-from django.utils import timezone
 
 from apps.alerts.constants import ActionSource
 from apps.alerts.incident_appearance.renderers.constants import DEFAULT_BACKUP_TITLE
@@ -14,25 +13,17 @@
 from apps.slack.chatops_proxy_routing import make_private_metadata, make_value
 from apps.slack.constants import CACHE_UPDATE_INCIDENT_SLACK_MESSAGE_LIFETIME
 from apps.slack.errors import (
-    SlackAPICantUpdateMessageError,
     SlackAPIChannelArchivedError,
-    SlackAPIChannelInactiveError,
     SlackAPIChannelNotFoundError,
     SlackAPIError,
-    SlackAPIInvalidAuthError,
     SlackAPIMessageNotFoundError,
     SlackAPIRatelimitError,
     SlackAPIRestrictedActionError,
     SlackAPITokenError,
 )
 from apps.slack.scenarios import scenario_step
-from apps.slack.scenarios.slack_renderer import AlertGroupLogSlackRenderer
 from apps.slack.slack_formatter import SlackFormatter
-from apps.slack.tasks import (
-    post_or_update_log_report_message_task,
-    send_message_to_thread_if_bot_not_in_channel,
-    update_incident_slack_message,
-)
+from apps.slack.tasks import send_message_to_thread_if_bot_not_in_channel, update_incident_slack_message
 from apps.slack.types import (
     Block,
     BlockActionType,
@@ -95,7 +86,6 @@ def process_signal(self, alert: Alert) -> None:
             else:
                 # check if alert group was posted to slack before posting message to thread
                 if not alert.group.skip_escalation_in_slack:
-                    self._send_log_report_message(alert.group, channel_id)
                     self._send_message_to_thread_if_bot_not_in_channel(alert.group, channel_id)
         else:
             # check if alert group was posted to slack before updating its message
@@ -208,11 +198,6 @@ def _send_debug_mode_notice(self, alert_group: AlertGroup, channel_id: str) -> N
             blocks=blocks,
         )
 
-    def _send_log_report_message(self, alert_group: AlertGroup, channel_id: str) -> None:
-        post_or_update_log_report_message_task.apply_async(
-            (alert_group.pk, self.slack_team_identity.pk),
-        )
-
     def _send_message_to_thread_if_bot_not_in_channel(self, alert_group: AlertGroup, channel_id: str) -> None:
         send_message_to_thread_if_bot_not_in_channel.apply_async(
             (alert_group.pk, self.slack_team_identity.pk, channel_id),
@@ -895,76 +880,6 @@ def process_signal(self, log_record: AlertGroupLogRecord) -> None:
             message.delete()
 
 
-class UpdateLogReportMessageStep(scenario_step.ScenarioStep):
-    def process_signal(self, alert_group: AlertGroup) -> None:
-        if alert_group.skip_escalation_in_slack or alert_group.channel.is_rate_limited_in_slack:
-            return
-
-        self.update_log_message(alert_group)
-
-    def update_log_message(self, alert_group: AlertGroup) -> None:
-        slack_message = alert_group.slack_message
-        if slack_message is None:
-            logger.info(
-                f"Cannot update log message for alert_group {alert_group.pk} because SlackMessage doesn't exist"
-            )
-            return None
-
-        slack_log_message = alert_group.slack_log_message
-
-        if slack_log_message is not None:
-            # prevent too frequent updates
-            if timezone.now() <= slack_log_message.last_updated + timezone.timedelta(seconds=5):
-                return
-
-            attachments = AlertGroupLogSlackRenderer.render_incident_log_report_for_slack(alert_group)
-            logger.debug(
-                f"Update log message for alert_group {alert_group.pk}, slack_log_message {slack_log_message.pk}"
-            )
-            try:
-                self._slack_client.chat_update(
-                    channel=slack_message.channel_id,
-                    text="Alert Group log",
-                    ts=slack_log_message.slack_id,
-                    attachments=attachments,
-                )
-            except SlackAPIRatelimitError as e:
-                if not alert_group.channel.is_rate_limited_in_slack:
-                    alert_group.channel.start_send_rate_limit_message_task(e.retry_after)
-            except SlackAPIMessageNotFoundError:
-                alert_group.slack_log_message = None
-                alert_group.save(update_fields=["slack_log_message"])
-            except (
-                SlackAPITokenError,
-                SlackAPIChannelNotFoundError,
-                SlackAPIChannelArchivedError,
-                SlackAPIChannelInactiveError,
-                SlackAPIInvalidAuthError,
-                SlackAPICantUpdateMessageError,
-            ):
-                pass
-            else:
-                slack_log_message.last_updated = timezone.now()
-                slack_log_message.save(update_fields=["last_updated"])
-                logger.debug(
-                    f"Finished update log message for alert_group {alert_group.pk}, "
-                    f"slack_log_message {slack_log_message.pk}"
-                )
-        # check how much time has passed since slack message was created
-        # to prevent eternal loop of restarting update log message task
-        elif timezone.now() <= slack_message.created_at + timezone.timedelta(minutes=5):
-            logger.debug(
-                f"Update log message failed for alert_group {alert_group.pk}: "
-                f"log message does not exist yet. Restarting post_or_update_log_report_message_task..."
-            )
-            post_or_update_log_report_message_task.apply_async(
-                (alert_group.pk, self.slack_team_identity.pk, True),
-                countdown=3,
-            )
-        else:
-            logger.debug(f"Update log message failed for alert_group {alert_group.pk}: " f"log message does not exist.")
-
-
 STEPS_ROUTING: ScenarioRoute.RoutingSteps = [
     {
         "payload_type": PayloadType.INTERACTIVE_MESSAGE,

engine/apps/slack/scenarios/slack_renderer.py

@@ -39,17 +39,3 @@ def render_alert_group_future_log_report_text(alert_group: "AlertGroup"):
                     for plan_line in escalation_policies_plan[time]:
                         result += f"*{humanize.naturaldelta(time)}:* {plan_line}\n"
         return result
-
-    @staticmethod
-    def render_incident_log_report_for_slack(alert_group: "AlertGroup"):
-        attachments = []
-        past = AlertGroupLogSlackRenderer.render_alert_group_past_log_report_text(alert_group)
-        future = AlertGroupLogSlackRenderer.render_alert_group_future_log_report_text(alert_group)
-        text = past + future
-        if len(text) > 0:
-            attachments.append(
-                {
-                    "text": text,
-                }
-            )
-        return attachments

engine/apps/slack/tasks.py

@@ -20,7 +20,6 @@
     SlackAPITokenError,
     SlackAPIUsergroupNotFoundError,
 )
-from apps.slack.scenarios.scenario_step import ScenarioStep
 from apps.slack.utils import (
     get_cache_key_update_incident_slack_message,
     get_populate_slack_channel_task_id_key,
@@ -289,27 +288,7 @@ def populate_slack_user_identities(organization_pk):
     autoretry_for=(Exception,), retry_backoff=True, max_retries=1 if settings.DEBUG else None
 )
 def post_or_update_log_report_message_task(alert_group_pk, slack_team_identity_pk, update=False):
-    logger.debug(f"Start post_or_update_log_report_message_task for alert_group {alert_group_pk}")
-    from apps.alerts.models import AlertGroup
-    from apps.slack.models import SlackTeamIdentity
-
-    UpdateLogReportMessageStep = ScenarioStep.get_step("distribute_alerts", "UpdateLogReportMessageStep")
-
-    slack_team_identity = SlackTeamIdentity.objects.get(pk=slack_team_identity_pk)
-    alert_group = AlertGroup.objects.get(pk=alert_group_pk)
-    step = UpdateLogReportMessageStep(slack_team_identity, alert_group.channel.organization)
-
-    if alert_group.skip_escalation_in_slack or alert_group.channel.is_rate_limited_in_slack:
-        return
-
-    if update:  # flag to prevent multiple posting log message to slack
-        step.update_log_message(alert_group)
-    else:
-        # don't post a new message, as it is available from the button
-        # this is an intermediate step, so we will only update posted messages but not post new ones
-        # once majority of messages are updated, we can remove this step (https://github.com/grafana/oncall/pull/4686)
-        pass
-    logger.debug(f"Finish post_or_update_log_report_message_task for alert_group {alert_group_pk}")
+    return "Deprecated, will be removed after queue cleanup"
 
 
 @shared_dedicated_queue_retry_task(