[LTM1.6] Allow metrics to be enabled selectively and disable all by default. (#362)

* Allow metrics to be enabled selectively and disable all by default.

* Update docs.

Author: progala

development/dev.env

@@ -50,3 +50,5 @@ POSTGRES_HOST=postgres
 MYSQL_DATABASE=nautobot
 MYSQL_USER=nautobot
 MYSQL_ROOT_HOST=%
+
+NAUTOBOT_DLM_ENABLED_METRICS = "nautobot_lcm_software_compliance_per_device_type,nautobot_lcm_software_compliance_per_inventory_item,nautobot_lcm_hw_end_of_support_per_part_number,nautobot_metrics_lcm_hw_end_of_support_site"

development/nautobot_config.py

@@ -234,7 +234,7 @@
 # }
 
 # Expose Prometheus monitoring metrics at the HTTP endpoint '/metrics'
-METRICS_ENABLED = False
+METRICS_ENABLED = True
 
 # Credentials that Nautobot will uses to authenticate to devices when connecting via NAPALM.
 NAPALM_USERNAME = os.environ.get("NAPALM_USERNAME", "")
@@ -262,6 +262,7 @@
         "barchart_bar_width": float(os.environ.get("BARCHART_BAR_WIDTH", 0.1)),
         "barchart_width": int(os.environ.get("BARCHART_WIDTH", 12)),
         "barchart_height": int(os.environ.get("BARCHART_HEIGHT", 5)),
+        "enabled_metrics": [x for x in os.environ.get("NAUTOBOT_DLM_ENABLED_METRICS", "").split(",") if x],
     },
 }
 

docs/admin/install.md

@@ -44,6 +44,7 @@ PLUGINS_CONFIG = {
         "barchart_bar_width": float(os.environ.get("BARCHART_BAR_WIDTH", 0.1)),
         "barchart_width": int(os.environ.get("BARCHART_WIDTH", 12)),
         "barchart_height": int(os.environ.get("BARCHART_HEIGHT", 5)),
+        "enabled_metrics": [x for x in os.environ.get("NAUTOBOT_DLM_ENABLED_METRICS", "").split(",") if x],
     },
 }
 ```
@@ -72,8 +73,21 @@ sudo systemctl restart nautobot nautobot-worker nautobot-scheduler
 
 The plugin behavior can be controlled with the following list of settings.
 
-| Key     | Example | Default | Description                          |
-| ------- | ------ | -------- | ------------------------------------- |
-| enable_backup | True | True | A boolean to represent whether or not to run backup configurations within the plugin. |
-| platform_slug_map | {"cisco_wlc": "cisco_aireos"} | None | A dictionary in which the key is the platform slug and the value is what netutils uses in any "network_os" parameter. |
-| per_feature_bar_width | 0.15 | 0.15 | The width of the table bar within the overview report |
+| Key     | ENV VAR | Example | Default | Description                          |
+| ------- | ------ |  ------ | -------- | ------------------------------------- |
+| barchart_bar_width | BARCHART_BAR_WIDTH | 0.15 | 0.1 | The width of the table bar within the overview report. |
+| barchart_width | BARCHART_WIDTH | 15 | 12 | The width of the barchart within the overview report. |
+| barchart_height | BARCHART_HEIGHT | 8 | 5 | The height of the barchart within the overview report. |
+| enabled_metrics | NAUTOBOT_DLM_ENABLED_METRICS | `["nautobot_metrics_lcm_hw_end_of_support_site"]` | `[]` | Enables metrics corresponding to the provided, comma separated, entries. |
+
+### Available Metric Names
+
+Following are the metric names that can be defined in `enabled_metrics`:
+
+- `nautobot_lcm_software_compliance_per_device_type`: Number of devices with valid/invalid software by device_type.
+
+- `nautobot_lcm_software_compliance_per_inventory_item`: Number of inventory items with valid/invalid software.
+
+- `nautobot_lcm_hw_end_of_support_per_part_number`: Number of End of Support devices and inventory items per Part Number.
+
+- `nautobot_metrics_lcm_hw_end_of_support_site`: Number of End of Support devices and inventory items per Site.

nautobot_device_lifecycle_mgmt/__init__.py

@@ -29,6 +29,7 @@ class DeviceLifeCycleConfig(PluginConfig):
         "barchart_bar_width": 0.1,
         "barchart_width": 12,
         "barchart_height": 5,
+        "enabled_metrics": [],
     }
     caching_config = {}
 

nautobot_device_lifecycle_mgmt/metrics.py

@@ -1,6 +1,7 @@
 """Nautobot Device LCM plugin application level metrics ."""
 from datetime import datetime
 
+from django.conf import settings
 from django.db.models import Count, F, IntegerField, OuterRef, Q, Subquery, Value
 from django.db.models.functions import Coalesce
 from nautobot.dcim.models import Device, DeviceType, InventoryItem, Site
@@ -12,6 +13,8 @@
     InventoryItemSoftwareValidationResult,
 )
 
+PLUGIN_CFG = settings.PLUGINS_CONFIG["nautobot_device_lifecycle_mgmt"]
+
 
 def metrics_lcm_validation_report_device_type():
     """Calculate number of devices with valid/invalid software by device_type.
@@ -111,8 +114,8 @@ def metrics_lcm_validation_report_inventory_item():
     yield inventory_item_software_compliance_gauge
 
 
-def metrics_lcm_hw_end_of_support():  # pylint: disable=too-many-locals
-    """Calculate number of End of Support devices and inventory items per Part Number and per Site.
+def metrics_lcm_hw_end_of_support_part_number():  # pylint: disable=too-many-locals
+    """Calculate number of End of Support devices and inventory items per Part Number.
 
     Yields:
         GaugeMetricFamily: Prometheus Metrics
@@ -122,9 +125,6 @@ def metrics_lcm_hw_end_of_support():  # pylint: disable=too-many-locals
         "Nautobot LCM Hardware End of Support per Part Number",
         labels=["part_number"],
     )
-    hw_end_of_support_site_gauge = GaugeMetricFamily(
-        "nautobot_lcm_hw_end_of_support_per_site", "Nautobot LCM Hardware End of Support per Site", labels=["site"]
-    )
 
     today = datetime.today().date()
     hw_end_of_support = HardwareLCM.objects.filter(end_of_support__lt=today)
@@ -177,6 +177,25 @@ def metrics_lcm_hw_end_of_support():  # pylint: disable=too-many-locals
 
     yield hw_end_of_support_part_number_gauge
 
+
+def metrics_lcm_hw_end_of_support_site():  # pylint: disable=too-many-locals
+    """Calculate number of End of Support devices and inventory items per Site.
+
+    Yields:
+        GaugeMetricFamily: Prometheus Metrics
+    """
+    hw_end_of_support_site_gauge = GaugeMetricFamily(
+        "nautobot_lcm_hw_end_of_support_per_site", "Nautobot LCM Hardware End of Support per Site", labels=["site"]
+    )
+
+    today = datetime.today().date()
+    hw_end_of_support = HardwareLCM.objects.filter(end_of_support__lt=today)
+    hw_end_of_support_device_types = hw_end_of_support.exclude(device_type__isnull=True).values_list(
+        "device_type", flat=True
+    )
+    hw_end_of_support_invitems = hw_end_of_support.exclude(inventory_item__isnull=True).values_list(
+        "inventory_item", flat=True
+    )
     # Initialize per site count to 0 for all sites
     init_site_counts = Site.objects.values(site_slug=F("slug")).annotate(
         site_count=Value(0, output_field=IntegerField())
@@ -214,8 +233,12 @@ def metrics_lcm_hw_end_of_support():  # pylint: disable=too-many-locals
     yield hw_end_of_support_site_gauge
 
 
-metrics = [
-    metrics_lcm_hw_end_of_support,
-    metrics_lcm_validation_report_device_type,
-    metrics_lcm_validation_report_inventory_item,
-]
+metrics = []
+if "nautobot_lcm_software_compliance_per_device_type" in PLUGIN_CFG["enabled_metrics"]:
+    metrics.append(metrics_lcm_validation_report_device_type)
+if "nautobot_lcm_software_compliance_per_inventory_item" in PLUGIN_CFG["enabled_metrics"]:
+    metrics.append(metrics_lcm_validation_report_inventory_item)
+if "nautobot_lcm_hw_end_of_support_per_part_number" in PLUGIN_CFG["enabled_metrics"]:
+    metrics.append(metrics_lcm_hw_end_of_support_part_number)
+if "nautobot_metrics_lcm_hw_end_of_support_site" in PLUGIN_CFG["enabled_metrics"]:
+    metrics.append(metrics_lcm_hw_end_of_support_site)

nautobot_device_lifecycle_mgmt/tests/test_metrics.py

@@ -3,7 +3,8 @@
 from django.test import TestCase
 
 from nautobot_device_lifecycle_mgmt.metrics import (
-    metrics_lcm_hw_end_of_support,
+    metrics_lcm_hw_end_of_support_part_number,
+    metrics_lcm_hw_end_of_support_site,
     metrics_lcm_validation_report_device_type,
     metrics_lcm_validation_report_inventory_item,
 )
@@ -53,22 +54,20 @@ def test_metrics_lcm_validation_report_inventory_item(self):
             sample_labels = tuple(sample.labels.items())
             self.assertEqual(expected_ts_samples[sample_labels], sample.value)
 
-    def test_metrics_lcm_hw_end_of_support_does_not_error(self):
+    def test_metrics_lcm_hw_end_of_support_site_does_not_error(self):
         """Query providing data to hw_end_of_support_site_gauge metric should not error out.
         Guards against https://github.com/nautobot/nautobot-app-device-lifecycle-mgmt/issues/309
         """
-        metric_gen = metrics_lcm_hw_end_of_support()
-        # skip hw_end_of_support_part_number_gauge
-        next(metric_gen)
+        metric_gen = metrics_lcm_hw_end_of_support_site()
         try:
             # Get hw_end_of_support_site_gauge
             next(metric_gen)
         except ProgrammingError:
-            self.fail("hw_end_of_support_site_gauge query bug")
+            self.fail("hw_end_of_support_site query bug")
 
     def test_metrics_lcm_hw_end_of_support_part_number(self):
         """Test metric hw_end_of_support_part_number_gauge."""
-        metric_gen = metrics_lcm_hw_end_of_support()
+        metric_gen = metrics_lcm_hw_end_of_support_part_number()
 
         # Get hw_end_of_support_part_number_gauge
         metric = next(metric_gen)
@@ -86,10 +85,7 @@ def test_metrics_lcm_hw_end_of_support_part_number(self):
 
     def test_metrics_lcm_hw_end_of_support_site_gauge(self):
         """Test metric hw_end_of_support_site_gauge."""
-        metric_gen = metrics_lcm_hw_end_of_support()
-
-        # skip hw_end_of_support_part_number_gauge
-        next(metric_gen)
+        metric_gen = metrics_lcm_hw_end_of_support_site()
 
         # Get hw_end_of_support_site_gauge
         metric = next(metric_gen)