RBD: Don't query Ceph on stats for exclusive pools

Collecting stats for provisioned_capacity_gb takes a long time since we
have to query each individual image for the provisioned size.  If we are
using the pool just for Cinder and/or are willing to accept a potential
deviation in Cinder stats we could just not retrieve this information
and calculate this based on the DB information for the volumes.

This patch adds configuration option `rbd_exclusive_cinder_pool` that
allows us to disable the size collection and thus improve the stats
reporting speed.

Change-Id: I32c7746fa9149bce6cdec96ee9aa87b303de4271
Closes-Bug: #1704106

cinder/tests/unit/volume/drivers/test_rbd.py

@@ -141,6 +141,19 @@ CEPH_MON_DUMP = r"""dumped monmap epoch 1
 """
 
 
+class MockDriverConfig(object):
+    def __init__(self, **kwargs):
+        my_dict = vars(self)
+        my_dict.update(kwargs)
+        my_dict.setdefault('max_over_subscription_ratio', 1.0)
+        my_dict.setdefault('reserved_percentage', 0)
+        my_dict.setdefault('volume_backend_name', 'RBD')
+        my_dict.setdefault('_default', None)
+
+    def __call__(self, value):
+        return getattr(self, value, self._default)
+
+
 def mock_driver_configuration(value):
     if value == 'max_over_subscription_ratio':
         return 1.0
@@ -1191,8 +1204,9 @@ class RBDTestCase(test.TestCase):
             expected['replication_targets'] = [t['backend_id']for t in targets]
             expected['replication_targets'].append('default')
 
+        my_safe_get = MockDriverConfig(rbd_exclusive_cinder_pool=False)
         self.mock_object(self.driver.configuration, 'safe_get',
-                         mock_driver_configuration)
+                         my_safe_get)
 
         with mock.patch.object(self.driver, '_get_fsid') as mock_get_fsid:
             mock_get_fsid.return_value = expected_fsid
@@ -1202,9 +1216,46 @@ class RBDTestCase(test.TestCase):
     @common_mocks
     @mock.patch('cinder.volume.drivers.rbd.RBDDriver._get_usage_info')
     @mock.patch('cinder.volume.drivers.rbd.RBDDriver._get_pool_stats')
-    def test_update_volume_stats_error(self, stats_mock, usage_mock):
+    def test_update_volume_stats_exclusive_pool(self, stats_mock, usage_mock):
+        stats_mock.return_value = (mock.sentinel.free_capacity_gb,
+                                   mock.sentinel.total_capacity_gb)
+
+        expected_fsid = 'abc'
+        expected_location_info = ('nondefault:%s:%s:%s:rbd' %
+                                  (self.cfg.rbd_ceph_conf, expected_fsid,
+                                   self.cfg.rbd_user))
+        expected = dict(
+            volume_backend_name='RBD',
+            replication_enabled=False,
+            vendor_name='Open Source',
+            driver_version=self.driver.VERSION,
+            storage_protocol='ceph',
+            total_capacity_gb=mock.sentinel.total_capacity_gb,
+            free_capacity_gb=mock.sentinel.free_capacity_gb,
+            reserved_percentage=0,
+            thin_provisioning_support=True,
+            max_over_subscription_ratio=1.0,
+            multiattach=False,
+            location_info=expected_location_info)
+
+        my_safe_get = MockDriverConfig(rbd_exclusive_cinder_pool=True)
         self.mock_object(self.driver.configuration, 'safe_get',
-                         mock_driver_configuration)
+                         my_safe_get)
+
+        with mock.patch.object(self.driver, '_get_fsid',
+                               return_value=expected_fsid):
+            actual = self.driver.get_volume_stats(True)
+
+        self.assertDictEqual(expected, actual)
+        usage_mock.assert_not_called()
+
+    @common_mocks
+    @mock.patch('cinder.volume.drivers.rbd.RBDDriver._get_usage_info')
+    @mock.patch('cinder.volume.drivers.rbd.RBDDriver._get_pool_stats')
+    def test_update_volume_stats_error(self, stats_mock, usage_mock):
+        my_safe_get = MockDriverConfig(rbd_exclusive_cinder_pool=False)
+        self.mock_object(self.driver.configuration, 'safe_get',
+                         my_safe_get)
 
         expected_fsid = 'abc'
         expected_location_info = ('nondefault:%s:%s:%s:rbd' %
@@ -1219,7 +1270,6 @@ class RBDTestCase(test.TestCase):
                         free_capacity_gb='unknown',
                         reserved_percentage=0,
                         multiattach=False,
-                        provisioned_capacity_gb=0,
                         max_over_subscription_ratio=1.0,
                         thin_provisioning_support=True,
                         location_info=expected_location_info)

cinder/volume/drivers/rbd.py

@@ -103,6 +103,13 @@ RBD_OPTS = [
                      'dynamic value -used + current free- and to False to '
                      'report a static value -quota max bytes if defined and '
                      'global size of cluster if not-.'),
+    cfg.BoolOpt('rbd_exclusive_cinder_pool', default=False,
+                help="Set to True if the pool is used exclusively by Cinder. "
+                     "On exclusive use driver won't query images' provisioned "
+                     "size as they will match the value calculated by the "
+                     "Cinder core code for allocated_capacity_gb. This "
+                     "reduces the load on the Ceph cluster as well as on the "
+                     "volume service."),
 ]
 
 CONF = cfg.CONF
@@ -470,7 +477,6 @@ class RBDDriver(driver.CloneableImageVD,
             'storage_protocol': 'ceph',
             'total_capacity_gb': 'unknown',
             'free_capacity_gb': 'unknown',
-            'provisioned_capacity_gb': 0,
             'reserved_percentage': (
                 self.configuration.safe_get('reserved_percentage')),
             'multiattach': False,
@@ -492,10 +498,14 @@ class RBDDriver(driver.CloneableImageVD,
             stats['free_capacity_gb'] = free_capacity
             stats['total_capacity_gb'] = total_capacity
 
-            total_gbi = self._get_usage_info()
-            stats['provisioned_capacity_gb'] = total_gbi
+            # For exclusive pools let scheduler set provisioned_capacity_gb to
+            # allocated_capacity_gb, and for non exclusive query the value.
+            if not self.configuration.safe_get('rbd_exclusive_cinder_pool'):
+                total_gbi = self._get_usage_info()
+                stats['provisioned_capacity_gb'] = total_gbi
         except self.rados.Error:
-            # just log and return unknown capacities
+            # just log and return unknown capacities and let scheduler set
+            # provisioned_capacity_gb = allocated_capacity_gb
             LOG.exception('error refreshing volume stats')
         self._stats = stats
 

releasenotes/notes/feature-rbd-exclusive-pool-a9bdebdeb1f0bf37.yaml

@@ -0,0 +1,15 @@
+---
+features:
+  - |
+    When using the RBD pool exclusively for Cinder we can now set
+    `rbd_exclusive_cinder_pool` to `true` and Cinder will use DB information
+    to calculate provisioned size instead of querying all volumes in the
+    backend, which will reduce the load on the Ceph cluster and the volume
+    service.
+issues:
+  - |
+    If RBD stats collection is taking too long in your environment maybe even
+    leading to the service appearing as down you'll want to use the
+    `rbd_exclusive_cinder_pool = true` configuration option if you are using
+    the pool exclusively for Cinder and maybe even if you are not and can live
+    with the innacuracy.