libvirt: Do not reraise DiskNotFound exceptions during resize

When an instance has VERIFY_RESIZE status, the instance disk on the
source compute host has moved to <instance_path>/<instance_uuid>_resize
folder, which leads to disk not found errors if the update available
resource periodic task on the source compute runs before resize is
actually confirmed.

Icec2769bf42455853cbe686fb30fda73df791b25 almost fixed this issue but it
will only set reraise to False when task_state is not None, that isn't
the case when an instance is resized but resize is not yet confirmed.
This patch adds a condition based on vm_state to ensure we don't
reraise DiskNotFound exceptions while resize is not confirmed.

Closes-Bug: 1774249
Co-Authored-By: Vladyslav Drok <vdrok@mirantis.com>
Change-Id: Id687e11e235fd6c2f99bb647184310dfdce9a08d
(cherry picked from commit 966192704c20d1b4e9faf384c8dafac8ea6e06ea)
(cherry picked from commit f1280ab849d20819791f7c4030f570a917d3e91d)
(cherry picked from commit fd5c45473823105d8572d7940980163c6f09169c)
(cherry picked from commit d3bdeb26155c2d3b53850b790d3800a2dd78cada)

2 files changed, 37 insertions, 5 deletions

diff --git a/nova/tests/unit/virt/libvirt/test_driver.py b/nova/tests/unit/virt/libvirt/test_driver.py
index bba1ba07ee..f5595e8eab 100755
--- a/nova/tests/unit/virt/libvirt/test_driver.py
+++ b/nova/tests/unit/virt/libvirt/test_driver.py
@@ -13500,8 +13500,9 @@ class LibvirtConnTestCase(test.NoDBTestCase,
     @mock.patch('nova.objects.InstanceList.get_by_filters',
                 return_value=objects.InstanceList(objects=[
                     objects.Instance(uuid=uuids.instance,
+                                     vm_state=vm_states.ACTIVE,
                                      task_state=task_states.DELETING)]))
-    def test_disk_over_committed_size_total_disk_not_found_ignore(
+    def test_disk_over_committed_size_total_disk_not_found_ignore_task_state(
             self, mock_get, mock_bdms, mock_get_disk_info, mock_list_domains):
         """Tests that we handle DiskNotFound gracefully for an instance that
         is undergoing a task_state transition.
@@ -13521,7 +13522,32 @@ class LibvirtConnTestCase(test.NoDBTestCase,
                 return_value=objects.BlockDeviceMappingList())
     @mock.patch('nova.objects.InstanceList.get_by_filters',
                 return_value=objects.InstanceList(objects=[
-                    objects.Instance(uuid=uuids.instance, task_state=None)]))
+                    objects.Instance(uuid=uuids.instance,
+                                     task_state=None,
+                                     vm_state=vm_states.RESIZED)]))
+    def test_disk_over_committed_size_total_disk_not_found_ignore_vmstate(
+            self, mock_get, mock_bdms, mock_get_disk_info, mock_list_domains):
+        """Tests that we handle DiskNotFound gracefully for an instance that
+        is resized but resize is not confirmed yet.
+        """
+        mock_dom = mock.Mock()
+        mock_dom.XMLDesc.return_value = "<domain/>"
+        mock_dom.UUIDString.return_value = uuids.instance
+        mock_list_domains.return_value = [mock_dom]
+        drvr = libvirt_driver.LibvirtDriver(fake.FakeVirtAPI(), False)
+        self.assertEqual(0, drvr._get_disk_over_committed_size_total())
+
+    @mock.patch('nova.virt.libvirt.host.Host.list_instance_domains')
+    @mock.patch('nova.virt.libvirt.driver.LibvirtDriver.'
+                '_get_instance_disk_info_from_config',
+                side_effect=exception.DiskNotFound(location='/opt/stack/foo'))
+    @mock.patch('nova.objects.BlockDeviceMappingList.bdms_by_instance_uuid',
+                return_value=objects.BlockDeviceMappingList())
+    @mock.patch('nova.objects.InstanceList.get_by_filters',
+                return_value=objects.InstanceList(objects=[
+                    objects.Instance(uuid=uuids.instance,
+                                     vm_state=vm_states.ACTIVE,
+                                     task_state=None)]))
     def test_disk_over_committed_size_total_disk_not_found_reraise(
             self, mock_get, mock_bdms, mock_get_disk_info, mock_list_domains):
         """Tests that we handle DiskNotFound gracefully for an instance that
diff --git a/nova/virt/libvirt/driver.py b/nova/virt/libvirt/driver.py
index 6ef0999165..c616406088 100644
--- a/nova/virt/libvirt/driver.py
+++ b/nova/virt/libvirt/driver.py
@@ -68,6 +68,7 @@ from nova import block_device
 from nova.compute import power_state
 from nova.compute import task_states
 from nova.compute import utils as compute_utils
+from nova.compute import vm_states
 import nova.conf
 from nova.console import serial as serial_console
 from nova.console import type as ctype
@@ -7511,14 +7512,19 @@ class LibvirtDriver(driver.ComputeDriver):
                     # should ignore this instance and move on.
                     if guest.uuid in local_instances:
                         inst = local_instances[guest.uuid]
-                        if inst.task_state is not None:
+                        # bug 1774249 indicated when instance is in RESIZED
+                        # state it might also can't find back disk
+                        if (inst.task_state is not None or
+                            inst.vm_state == vm_states.RESIZED):
                             LOG.info('Periodic task is updating the host '
                                      'stats; it is trying to get disk info '
                                      'for %(i_name)s, but the backing disk '
                                      'was removed by a concurrent operation '
-                                     '(task_state=%(task_state)s)',
+                                     '(task_state=%(task_state)s) and '
+                                     '(vm_state=%(vm_state)s)',
                                      {'i_name': guest.name,
-                                      'task_state': inst.task_state},
+                                      'task_state': inst.task_state,
+                                      'vm_state': inst.vm_state},
                                      instance=inst)
                             err_ctxt.reraise = False