[postgres] Exclude rdsadmin database from schema collection by default (DataDog#22470)

* [postgres] Exclude rdsadmin database from schema collection by default

The rdsadmin database is an AWS RDS system database that is not accessible
to users. This change adds it to the default exclude_databases list for
collect_schemas to prevent errors when the Agent attempts to collect schema
information from RDS instances.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* [postgres] Exclude system and admin databases from schema collection by default

Exclude the following databases from schema collection and autodiscovery
by default:
- template0, template1 (PostgreSQL system databases)
- rdsadmin (AWS RDS)
- azure_maintenance (Azure)
- cloudsqladmin, alloydbadmin, alloydbmetadata (Google Cloud)

These databases are either system databases or cloud provider admin databases
that are not accessible to users. This prevents errors when the Agent attempts
to collect schema information or autodiscover databases.

Also introduces DEFAULT_EXCLUDED_DATABASES constant in dict_defaults.py to
avoid duplication of the excluded databases list across:
- instance_database_autodiscovery()
- instance_collect_schemas()
- test_config_defaults.py

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>

Author: dalextorres

postgres/assets/configuration/spec.yaml

@@ -471,8 +471,11 @@ files:
               - "cloudsqladmin"
               - "rdsadmin"
             display_default:
+              - "template0"
+              - "template1"
+              - "rdsadmin"
+              - "azure_maintenance"
               - "cloudsqladmin"
-              - "rdsadmin" 
               - "alloydbadmin"
               - "alloydbmetadata"
         - name: refresh
@@ -794,7 +797,7 @@ files:
               - "mydb"
         - name: exclude_databases
           description: |
-            A list of regex patterns to exclude databases. 
+            A list of regex patterns to exclude databases.
             Any database whose name matches any one of these patterns will be excluded.
             If empty, all databases matching other filters are included.
           value:
@@ -803,6 +806,14 @@ files:
               type: string
             example:
               - "privatedb.*"
+            default:
+              - "template0"
+              - "template1"
+              - "rdsadmin"
+              - "azure_maintenance"
+              - "cloudsqladmin"
+              - "alloydbadmin"
+              - "alloydbmetadata"
         - name: include_schemas
           description: |
             A list of regex patterns to include schemas. 

postgres/changelog.d/22470.fixed

@@ -0,0 +1 @@
+Exclude system and cloud provider admin databases from schema collection by default

postgres/datadog_checks/postgres/config_models/dict_defaults.py

@@ -7,6 +7,19 @@
 
 from . import instance
 
+# Default databases to exclude from schema collection and autodiscovery.
+# These are system databases or cloud provider admin databases that are not accessible to users.
+# This list should match the default value for `ignore_databases` in spec.yaml.
+DEFAULT_EXCLUDED_DATABASES = [
+    "template0",
+    "template1",
+    "rdsadmin",
+    "azure_maintenance",
+    "cloudsqladmin",
+    "alloydbadmin",
+    "alloydbmetadata",
+]
+
 
 def instance_database_identifier():
     return instance.DatabaseIdentifier(
@@ -20,7 +33,7 @@ def instance_database_autodiscovery():
         global_view_db="postgres",
         max_databases=100,
         include=[".*"],
-        exclude=["cloudsqladmin", "rdsadmin", "alloydbadmin", "alloydbmetadata"],
+        exclude=list(DEFAULT_EXCLUDED_DATABASES),
         refresh=600,
     )
 
@@ -78,7 +91,7 @@ def instance_collect_schemas():
         max_columns=50,
         collection_interval=600,
         include_databases=[],
-        exclude_databases=[],
+        exclude_databases=list(DEFAULT_EXCLUDED_DATABASES),
         include_schemas=[],
         exclude_schemas=[],
         include_tables=[],

postgres/datadog_checks/postgres/data/conf.yaml.example

@@ -402,7 +402,7 @@ instances:
         #   - master$
         #   - AdventureWorks.*
 
-        ## @param exclude - list of strings - optional - default: ['cloudsqladmin', 'rdsadmin', 'alloydbadmin', 'alloydbmetadata']
+        ## @param exclude - list of strings - optional - default: ['template0', 'template1', 'rdsadmin', 'azure_maintenance', 'cloudsqladmin', 'alloydbadmin', 'alloydbmetadata']
         ## Regular expression for database names to exclude as part of `database_autodiscovery`.
         ## Character casing is ignored. The regular expressions start matching from the beginning,
         ## so to match anywhere, prepend `.*`. For exact matches append `$`.
@@ -594,7 +594,7 @@ instances:
         #   - mydb
 
         ## @param exclude_databases - list of strings - optional
-        ## A list of regex patterns to exclude databases. 
+        ## A list of regex patterns to exclude databases.
         ## Any database whose name matches any one of these patterns will be excluded.
         ## If empty, all databases matching other filters are included.
         #

postgres/tests/test_config_defaults.py

@@ -13,6 +13,7 @@
 import pytest
 
 from datadog_checks.postgres.config import build_config
+from datadog_checks.postgres.config_models.dict_defaults import DEFAULT_EXCLUDED_DATABASES
 
 # Single source of truth for all expected default values
 # Organized by category for readability
@@ -55,15 +56,7 @@
     'table_count_limit': 200,
     'max_relations': 300,
     # === Database filtering ===
-    'ignore_databases': [
-        'template0',
-        'template1',
-        'rdsadmin',
-        'azure_maintenance',
-        'cloudsqladmin',
-        'alloydbadmin',
-        'alloydbmetadata',
-    ],
+    'ignore_databases': list(DEFAULT_EXCLUDED_DATABASES),
     'ignore_schemas_owned_by': [
         'rds_superuser',
         'rdsadmin',
@@ -120,6 +113,12 @@
         'max_columns': 50,
         'collection_interval': 600,
         'max_query_duration': 60,
+        'include_databases': [],
+        'exclude_databases': list(DEFAULT_EXCLUDED_DATABASES),
+        'include_schemas': [],
+        'exclude_schemas': [],
+        'include_tables': [],
+        'exclude_tables': [],
     },
     # === DBM: Obfuscator options ===
     'obfuscator_options': {
@@ -145,7 +144,7 @@
         'global_view_db': 'postgres',
         'max_databases': 100,
         'refresh': 600,
-        'exclude': ['cloudsqladmin', 'rdsadmin', 'alloydbadmin', 'alloydbmetadata'],
+        'exclude': list(DEFAULT_EXCLUDED_DATABASES),
         'include': ['.*'],
     },
     # === DBM: Lock metrics ===