🐛 🗃️ fix resource tracking not starting + adding project uuid to tracking table

packages/postgres-database/src/simcore_postgres_database/migration/versions/417f9eb848ce_adding_study_id_to_resource_tracking_.py

@@ -0,0 +1,30 @@
+"""adding study id to resource-tracking-container table
+
+Revision ID: 417f9eb848ce
+Revises: add0afaaf728
+Create Date: 2023-06-23 14:15:20.721005+00:00
+
+"""
+import sqlalchemy as sa
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision = "417f9eb848ce"
+down_revision = "add0afaaf728"
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.add_column(
+        "resource_tracker_container",
+        sa.Column("project_uuid", sa.String(), nullable=False),
+    )
+    # ### end Alembic commands ###
+
+
+def downgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_column("resource_tracker_container", "project_uuid")
+    # ### end Alembic commands ###

packages/postgres-database/src/simcore_postgres_database/models/resource_tracker.py

@@ -31,6 +31,12 @@
         nullable=False,
         doc="user_id label scraped via Prometheus (taken from container labels)",
     ),
+    sa.Column(
+        "project_uuid",
+        sa.String,
+        nullable=False,
+        doc="project_uuid label scraped via Prometheus (taken from container labels)",
+    ),
     sa.Column(
         "product_name",
         sa.String,

services/resource-usage-tracker/src/simcore_service_resource_usage_tracker/models/resource_tracker_container.py

@@ -2,6 +2,7 @@
 
 from arrow import Arrow
 from models_library.products import ProductName
+from models_library.projects import ProjectID
 from models_library.users import UserID
 from pydantic import BaseModel
 
@@ -11,6 +12,7 @@ class ContainerResourceUsageMetric(BaseModel):
     image: str
     user_id: UserID
     product_name: ProductName
+    project_uuid: ProjectID
     service_settings_reservation_nano_cpus: int | None
     service_settings_reservation_memory_bytes: int | None
     service_settings_reservation_additional_info: dict[str, Any] = {}

services/resource-usage-tracker/src/simcore_service_resource_usage_tracker/modules/db/repositories/resource_tracker.py

@@ -19,6 +19,7 @@ async def upsert_resource_tracker_container_data_(
                 container_id=data.container_id,
                 image=data.image,
                 user_id=data.user_id,
+                project_uuid=str(data.project_uuid),
                 product_name=data.product_name,
                 service_settings_reservation_nano_cpus=data.service_settings_reservation_nano_cpus,
                 service_settings_reservation_memory_bytes=data.service_settings_reservation_memory_bytes,

services/resource-usage-tracker/src/simcore_service_resource_usage_tracker/resource_tracker_core.py

@@ -16,13 +16,11 @@
 _logger = logging.getLogger(__name__)
 
 
-async def _prometheus_client_custom_query(
+def _prometheus_sync_client_custom_query(
     prometheus_client: PrometheusConnect, promql_cpu_query: str
 ) -> list[dict]:
     _logger.info("Querying prometheus with: %s", promql_cpu_query)
-    data: list[dict] = await asyncio.get_event_loop().run_in_executor(
-        None, prometheus_client.custom_query(promql_cpu_query)
-    )
+    data: list[dict] = prometheus_client.custom_query(promql_cpu_query)
     return data
 
 
@@ -33,8 +31,10 @@ async def _scrape_and_upload_container_resource_usage(
 ) -> None:
     # Query CPU seconds
     promql_cpu_query = f"sum without (cpu) (container_cpu_usage_seconds_total{{image=~'{image_regex}'}})[30m:1m]"
-    containers_cpu_seconds_usage: list = await _prometheus_client_custom_query(
-        prometheus_client, promql_cpu_query
+    containers_cpu_seconds_usage: list[
+        dict
+    ] = await asyncio.get_event_loop().run_in_executor(
+        None, _prometheus_sync_client_custom_query, prometheus_client, promql_cpu_query
     )
     _logger.info(
         "Received %s containers from Prometheus", len(containers_cpu_seconds_usage)
@@ -74,6 +74,7 @@ async def _scrape_and_upload_container_resource_usage(
             image=metric["image"],
             user_id=metric["container_label_user_id"],
             product_name=metric["container_label_product_name"],
+            project_uuid=metric["container_label_study_id"],
             service_settings_reservation_nano_cpus=int(nano_cpus)
             if nano_cpus
             else None,

services/resource-usage-tracker/tests/unit/with_dbs/test_collect_container_resource_usage_task.py

@@ -34,7 +34,7 @@ def mocked_prometheus_client_custom_query(
         data = json.load(file)
 
     mocked_get_prometheus_api_client = mocker.patch(
-        "simcore_service_resource_usage_tracker.resource_tracker_core._prometheus_client_custom_query",
+        "simcore_service_resource_usage_tracker.resource_tracker_core._prometheus_sync_client_custom_query",
         autospec=True,
         return_value=data,
     )

services/resource-usage-tracker/tests/unit/with_dbs/test_collect_container_resource_usage_task__on_update_set.py

@@ -77,7 +77,7 @@ def random_promql_output_generator():
                 "container_label_product_name": "osparc",
                 "container_label_simcore_service_settings": '[{"name": "ports", "type": "int", "value": 8888}, {"name": "env", "type": "string", "value": ["DISPLAY=:0"]}, {"name": "env", "type": "string", "value": ["SYM_SERVER_HOSTNAME=sym-server_%service_uuid%"]}, {"name": "mount", "type": "object", "value": [{"ReadOnly": true, "Source": "/tmp/.X11-unix", "Target": "/tmp/.X11-unix", "Type": "bind"}]}, {"name": "constraints", "type": "string", "value": ["node.platform.os == linux"]}, {"name": "Resources", "type": "Resources", "value": {"Limits": {"NanoCPUs": 4000000000, "MemoryBytes": 17179869184}, "Reservations": {"NanoCPUs": 100000000, "MemoryBytes": 536870912, "GenericResources": [{"DiscreteResourceSpec": {"Kind": "VRAM", "Value": 1}}]}}}]',
                 "container_label_simcore_user_agent": "puppeteer",
-                "container_label_study_id": "52d7e1a8-0c27-11ee-bec2-024201234c7",
+                "container_label_study_id": "46449cc3-7d83-4081-a44e-fc75a0c85f2c",
                 "container_label_user_id": "43820",
                 "container_label_uuid": "2b231c38-0ebc-5cc0-1234-1ffe573f54e9",
                 "id": "/docker/58e1138d51eb5eafd737024d0df0b01ef88f2087e5a3922565c59130d57ac7a3",
@@ -106,7 +106,7 @@ def mocked_prometheus_client_custom_query(
     mocker: MockerFixture, project_slug_dir: Path, random_promql_output_generator
 ) -> dict[str, mock.Mock]:
     mocked_get_prometheus_api_client = mocker.patch(
-        "simcore_service_resource_usage_tracker.resource_tracker_core._prometheus_client_custom_query",
+        "simcore_service_resource_usage_tracker.resource_tracker_core._prometheus_sync_client_custom_query",
         autospec=True,
         return_value=random_promql_output_generator["data"],
     )
@@ -135,13 +135,13 @@ async def test_collect_container_resource_usage_task(
     assert len(db_rows) == 1
 
     assert (
-        random_promql_output_generator["max_float"] == db_rows[0][7]
+        random_promql_output_generator["max_float"] == db_rows[0][8]
     )  # <-- container_cpu_usage_seconds_total
     assert (
         arrow.get(random_promql_output_generator["min_timestamp"]).datetime
-        == db_rows[0][8]
+        == db_rows[0][9]
     )  # <-- prometheus_created
     assert (
         arrow.get(random_promql_output_generator["max_timestamp"]).datetime
-        == db_rows[0][9]
+        == db_rows[0][10]
     )  # <-- prometheus_last_scraped