AAP-68024 Remove dead Host.last_job and last_job_host_summary FK fields (#16529)

* Remove dead Host.last_job and Host.last_job_host_summary FK fields

PR #16332 replaced these denormalized ForeignKey columns with dynamic
derivation via Host.latest_summary, but left the columns in place.
Nothing writes to or reads from them anymore — drop the dead weight.

- Remove last_job and last_job_host_summary ForeignKey fields from Host
- Convert HostSerializer to use SerializerMethodField instead of FK-backed fields
- Remove redundant JobHostSummary.latest_for_host/latest_job_for_host classmethods
- Add migration 0206 to drop the columns
- Update tests to use Host.latest_summary instead of removed fields/methods

* Remove dead last_job_host_summary_id UPDATE from cleanup_jobs

The Host.last_job_host_summary FK was removed in the prior commit,
so the raw SQL UPDATE that NULLed it before deleting JHS rows now
references a non-existent column and crashes.

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

awx/api/serializers.py

@@ -174,8 +174,6 @@ SUMMARIZABLE_FK_FIELDS = {
     'workflow_approval': DEFAULT_SUMMARY_FIELDS + ('timeout',),
     'schedule': DEFAULT_SUMMARY_FIELDS + ('next_run',),
     'unified_job_template': DEFAULT_SUMMARY_FIELDS + ('unified_job_type',),
-    # last_job and last_job_host_summary are derived from JobHostSummary in HostSerializer,
-    # not from the stale FK fields on Host.
     'last_update': DEFAULT_SUMMARY_FIELDS + ('status', 'failed', 'license_error'),
     'current_update': DEFAULT_SUMMARY_FIELDS + ('status', 'failed', 'license_error'),
     'current_job': DEFAULT_SUMMARY_FIELDS + ('status', 'failed', 'license_error'),
@@ -1803,6 +1801,8 @@ class HostSerializer(BaseSerializerWithVariables):
 
     has_active_failures = serializers.SerializerMethodField()
     has_inventory_sources = serializers.SerializerMethodField()
+    last_job = serializers.SerializerMethodField()
+    last_job_host_summary = serializers.SerializerMethodField()
 
     class Meta:
         model = Host
@@ -1818,7 +1818,7 @@ class HostSerializer(BaseSerializerWithVariables):
             'last_job_host_summary',
             'ansible_facts_modified',
         )
-        read_only_fields = ('last_job', 'last_job_host_summary', 'ansible_facts_modified')
+        read_only_fields = ('ansible_facts_modified',)
 
     def build_relational_field(self, field_name, relation_info):
         field_class, field_kwargs = super(HostSerializer, self).build_relational_field(field_name, relation_info)
@@ -1953,13 +1953,16 @@ class HostSerializer(BaseSerializerWithVariables):
             return ret
         if 'inventory' in ret and not obj.inventory:
             ret['inventory'] = None
-        last_summary = obj.latest_summary
-        if 'last_job' in ret:
-            ret['last_job'] = last_summary.job_id if last_summary else None
-        if 'last_job_host_summary' in ret:
-            ret['last_job_host_summary'] = last_summary.pk if last_summary else None
         return ret
 
+    def get_last_job(self, obj):
+        last_summary = obj.latest_summary
+        return last_summary.job_id if last_summary else None
+
+    def get_last_job_host_summary(self, obj):
+        last_summary = obj.latest_summary
+        return last_summary.pk if last_summary else None
+
     def get_has_active_failures(self, obj):
         last_summary = obj.latest_summary
         return bool(last_summary and last_summary.failed)

awx/main/management/commands/cleanup_jobs.py

@@ -53,28 +53,19 @@ JHS_CHUNK_SIZE = 1000
 
 
 def _pre_delete_job_host_summaries(job_pks, logger=None):
-    """Pre-delete JobHostSummary rows and clear Host FK references in batches.
+    """Pre-delete JobHostSummary rows in batches.
 
-    Django's cascade collector materializes all JHS IDs into a single
+    Deleting via raw SQL in chunks avoids Django's cascade collector
-    UPDATE ... IN (...) to SET_NULL on Host.last_job_host_summary.
+    materializing all JHS IDs into memory at once, which can exceed
-    With many jobs x hosts this exceeds PostgreSQL's 1GB alloc limit.
+    PostgreSQL's 1GB alloc limit when many jobs x hosts are involved.
-    Doing it in chunks with raw SQL avoids that.
     """
     if not job_pks:
         return
 
-    # ANY(%s) is PostgreSQL-specific; AWX only supports PostgreSQL
     with connection.cursor() as cursor:
         for i in range(0, len(job_pks), JHS_CHUNK_SIZE):
             chunk = list(job_pks[i : i + JHS_CHUNK_SIZE])
 
-            cursor.execute(
-                "UPDATE main_host SET last_job_host_summary_id = NULL"
-                " WHERE last_job_host_summary_id IN"
-                " (SELECT id FROM main_jobhostsummary WHERE job_id = ANY(%s))",
-                [chunk],
-            )
-
             cursor.execute(
                 "DELETE FROM main_jobhostsummary WHERE job_id = ANY(%s)",
                 [chunk],

awx/main/migrations/0210_remove_host_last_job_fields.py

@@ -0,0 +1,18 @@
+from django.db import migrations
+
+
+class Migration(migrations.Migration):
+    dependencies = [
+        ('main', '0209_fix_system_auditor_migration'),
+    ]
+
+    operations = [
+        migrations.RemoveField(
+            model_name='host',
+            name='last_job',
+        ),
+        migrations.RemoveField(
+            model_name='host',
+            name='last_job_host_summary',
+        ),
+    ]

awx/main/models/events.py

@@ -589,9 +589,6 @@ class JobEvent(BasePlaybookEvent):
 
             JobHostSummary.objects.bulk_create(summaries.values())
 
-            # last_job and last_job_host_summary are now derived via
-            # JobHostSummary.latest_for_host / latest_job_for_host
-
             # Create/update Host Metrics
             self._update_host_metrics(updated_hosts_list)
 

awx/main/models/inventory.py

@@ -550,23 +550,6 @@ class Host(CommonModelNameNotUnique, RelatedJobsMixin):
             help_text=_('Host variables in JSON or YAML format.'),
         )
     )
-    last_job = models.ForeignKey(
-        'Job',
-        related_name='hosts_as_last_job+',
-        null=True,
-        default=None,
-        editable=False,
-        on_delete=models.SET_NULL,
-    )
-    last_job_host_summary = models.ForeignKey(
-        'JobHostSummary',
-        related_name='hosts_as_last_job_summary+',
-        blank=True,
-        null=True,
-        default=None,
-        editable=False,
-        on_delete=models.SET_NULL,
-    )
     inventory_sources = models.ManyToManyField(
         'InventorySource',
         related_name='hosts',

awx/main/models/jobs.py

@@ -1140,22 +1140,6 @@ class JobHostSummary(CreatedModifiedModel):
             self.skipped,
         )
 
-    @classmethod
-    def latest_for_host(cls, host_id):
-        """Return the most recent JobHostSummary for a given host, or None."""
-        return cls.objects.filter(host_id=host_id).order_by('-id').first()
-
-    @classmethod
-    def latest_job_for_host(cls, host_id):
-        """Return the Job from the most recent JobHostSummary for a host, or None."""
-        summary = cls.latest_for_host(host_id)
-        if summary:
-            try:
-                return summary.job
-            except cls.job.field.related_model.DoesNotExist:
-                return None
-        return None
-
     def get_absolute_url(self, request=None):
         return reverse('api:job_host_summary_detail', kwargs={'pk': self.pk}, request=request)
 

awx/main/signals.py

@@ -302,11 +302,6 @@ def migrate_children_from_deleted_group_to_parent_groups(sender, **kwargs):
                         pass
 
 
-# Host.last_job and Host.last_job_host_summary are now derived from
-# JobHostSummary.latest_for_host / latest_job_for_host.
-# No signal handlers needed to maintain these denormalized FKs.
-
-
 # Set via ActivityStreamRegistrar to record activity stream events
 
 

awx/main/tests/functional/models/test_events.py

@@ -71,7 +71,7 @@ class TestEvents:
             assert s.skipped == 0
 
         for host in Host.objects.all():
-            latest_summary = JobHostSummary.latest_for_host(host.id)
+            latest_summary = host.latest_summary
             assert latest_summary is not None
             assert latest_summary.job_id == self.job.id
             assert latest_summary.host == host
@@ -106,13 +106,12 @@ class TestEvents:
         # be related to the appropriate Host)
         assert JobHostSummary.objects.count() == 1
         for h in Host.objects.all():
-            latest_summary = JobHostSummary.latest_for_host(h.id)
+            latest_summary = h.latest_summary
             if h.name == 'Host 1':
                 assert latest_summary is not None
                 assert latest_summary.job_id == self.job.id
                 assert latest_summary.id == JobHostSummary.objects.first().id
             else:
-                # all other hosts in the inventory should have no summary
                 assert latest_summary is None
 
     def test_host_metrics_insert(self):

awx/main/tests/functional/models/test_host_summary_fields.py

@@ -47,10 +47,6 @@ class TestHostSummaryFields:
         ).save()
 
         summary = JobHostSummary.objects.filter(host=host, job=job).first()
-        host.last_job = job
-        host.last_job_host_summary = summary
-        host.save(update_fields=['last_job', 'last_job_host_summary'])
-        host.refresh_from_db()
 
         return host, job, summary
 

awx/main/tests/unit/commands/test_cleanup_jobs.py

@@ -18,13 +18,8 @@ class TestPreDeleteJobHostSummaries:
 
             _pre_delete_job_host_summaries(job_pks)
 
-            assert mock_cursor.execute.call_count == 2
+            assert mock_cursor.execute.call_count == 1
-            update_call = mock_cursor.execute.call_args_list[0]
+            delete_call = mock_cursor.execute.call_args_list[0]
-            assert 'UPDATE main_host SET last_job_host_summary_id = NULL' in update_call[0][0]
-            assert 'ANY(%s)' in update_call[0][0]
-            assert update_call[0][1] == [[1, 2, 3]]
-
-            delete_call = mock_cursor.execute.call_args_list[1]
             assert 'DELETE FROM main_jobhostsummary' in delete_call[0][0]
             assert 'ANY(%s)' in delete_call[0][0]
             assert delete_call[0][1] == [[1, 2, 3]]
@@ -38,16 +33,16 @@ class TestPreDeleteJobHostSummaries:
 
             _pre_delete_job_host_summaries(job_pks)
 
-            # 2 chunks x 2 SQL statements each = 4 execute calls
+            # 2 chunks x 1 DELETE each = 2 execute calls
-            assert mock_cursor.execute.call_count == 4
+            assert mock_cursor.execute.call_count == 2
 
             # First chunk should have JHS_CHUNK_SIZE items
-            first_update = mock_cursor.execute.call_args_list[0]
+            first_delete = mock_cursor.execute.call_args_list[0]
-            assert len(first_update[0][1][0]) == JHS_CHUNK_SIZE
+            assert len(first_delete[0][1][0]) == JHS_CHUNK_SIZE
 
             # Second chunk should have the remainder
-            second_update = mock_cursor.execute.call_args_list[2]
+            second_delete = mock_cursor.execute.call_args_list[1]
-            assert len(second_update[0][1][0]) == 499
+            assert len(second_delete[0][1][0]) == 499
 
     def test_sql_is_fully_static(self):
         """SQL strings contain no interpolated values — only ANY(%s) placeholders."""
@@ -77,21 +72,6 @@ class TestPreDeleteJobHostSummaries:
 
             logger.debug.assert_called_once()
 
-    def test_update_runs_before_delete(self):
-        """Host FK must be NULLed before JHS rows are deleted."""
-        job_pks = [1]
-        with mock.patch('awx.main.management.commands.cleanup_jobs.connection') as mock_conn:
-            mock_cursor = mock.MagicMock()
-            mock_conn.cursor.return_value.__enter__ = mock.Mock(return_value=mock_cursor)
-            mock_conn.cursor.return_value.__exit__ = mock.Mock(return_value=False)
-
-            _pre_delete_job_host_summaries(job_pks)
-
-            first_sql = mock_cursor.execute.call_args_list[0][0][0]
-            second_sql = mock_cursor.execute.call_args_list[1][0][0]
-            assert 'UPDATE' in first_sql
-            assert 'DELETE' in second_sql
-
 
 class TestDeleteMetaPreDelete:
     """Verify DeleteMeta.delete_jobs() calls _pre_delete_job_host_summaries correctly."""