openstack
/
nova
Code Issues Proposed changes

Merge "Fix race condition in reshaper handler"

This commit is contained in:
Zuul 2018-08-30 17:35:09 +00:00 committed by Gerrit Code Review
parent 18e801d4de d98986415c
commit 3779211031
3 changed files with 20 additions and 32 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
nova/api/openstack/placement/handlers/reshaper.py
View File

@ -80,7 +80,8 @@ def reshape(req):
inv_obj = inventory.make_inventory_object(
resource_provider, res_class, **inv_data)
inv_list.append(inv_obj)
inventory_by_rp[rp_uuid] = rp_obj.InventoryList(objects=inv_list)
inventory_by_rp[resource_provider] = rp_obj.InventoryList(
objects=inv_list)
# Make the consumer objects associated with the allocations.
consumers, new_consumers_created = allocation.inspect_consumers(

33
nova/api/openstack/placement/objects/resource_provider.py
View File

@ -4171,9 +4171,9 @@ def reshape(ctx, inventories, allocations):
:param ctx: `nova.api.openstack.placement.context.RequestContext` object
containing the DB transaction context.
:param inventories: dict, keyed by resource provider UUID, of
`InventoryList` objects representing the replaced
inventory information for the provider.
:param inventories: dict, keyed by ResourceProvider, of `InventoryList`
objects representing the replaced inventory information
for the provider.
:param allocations: `AllocationList` object containing all allocations for
all consumers being modified by the reshape operation.
:raises: `exception.ConcurrentUpdateDetected` when any resource provider or
@ -4201,25 +4201,12 @@ def reshape(ctx, inventories, allocations):
# allocated, but that's allowed by the code. If the final picture is
# overcommitted, we'll get an appropriate exception when we replace the
# allocations at the end.
for rp_uuid, new_inv_list in inventories.items():
for rp, new_inv_list in inventories.items():
LOG.debug("reshaping: *interim* inventory replacement for provider %s",
rp_uuid)
# We need the resource provider object.
if new_inv_list:
# If the inventory list is nonempty, grab it from the first entry.
# This overrides (and will replace, below) an object that was
# prepopulated from the allocations.
rp = new_inv_list[0].resource_provider
elif rp_uuid in affected_providers:
# No inventory object; second choice is from an allocation object.
rp = affected_providers[rp_uuid]
else:
# No inventory or allocations for this provider - which makes sense
# when we're "clearing out" a provider. Look it up.
rp = ResourceProvider.get_by_uuid(ctx, rp_uuid)
rp.uuid)
# Update the cache. This may be replacing an entry that came from
# allocations, or adding a new entry from inventories and/or db lookup.
affected_providers[rp_uuid] = rp
# allocations, or adding a new entry from inventories.
affected_providers[rp.uuid] = rp
# Optimization: If the new inventory is empty, the below would be
# replacing it with itself (and incrementing the generation)
@ -4253,7 +4240,7 @@ def reshape(ctx, inventories, allocations):
allocations.replace_all()
# And finally, we can set the inventories to their actual desired state.
for rp_uuid, new_inv_list in inventories.items():
for rp, new_inv_list in inventories.items():
LOG.debug("reshaping: *final* inventory replacement for provider %s",
rp_uuid)
affected_providers[rp_uuid].set_inventory(new_inv_list)
rp.uuid)
rp.set_inventory(new_inv_list)

16
nova/tests/functional/api/openstack/placement/db/test_reshape.py
View File

@ -104,7 +104,7 @@ class ReshapeTestCase(tb.PlacementDbBaseTestCase):
# storage provider.
after_inventories = {
# cn1 keeps the RAM only
cn1.uuid: rp_obj.InventoryList(self.ctx, objects=[
cn1: rp_obj.InventoryList(self.ctx, objects=[
rp_obj.Inventory(
self.ctx, resource_provider=cn1,
resource_class='MEMORY_MB', total=32768, reserved=0,
@ -112,14 +112,14 @@ class ReshapeTestCase(tb.PlacementDbBaseTestCase):
allocation_ratio=1.0),
]),
# each NUMA node gets half of the CPUs
cn1_numa0.uuid: rp_obj.InventoryList(self.ctx, objects=[
cn1_numa0: rp_obj.InventoryList(self.ctx, objects=[
rp_obj.Inventory(
self.ctx, resource_provider=cn1_numa0,
resource_class='VCPU', total=8, reserved=0,
max_unit=8, min_unit=1, step_size=1,
allocation_ratio=1.0),
]),
cn1_numa1.uuid: rp_obj.InventoryList(self.ctx, objects=[
cn1_numa1: rp_obj.InventoryList(self.ctx, objects=[
rp_obj.Inventory(
self.ctx, resource_provider=cn1_numa1,
resource_class='VCPU', total=8, reserved=0,
@ -127,7 +127,7 @@ class ReshapeTestCase(tb.PlacementDbBaseTestCase):
allocation_ratio=1.0),
]),
# The sharing provider gets a bunch of disk
ss.uuid: rp_obj.InventoryList(self.ctx, objects=[
ss: rp_obj.InventoryList(self.ctx, objects=[
rp_obj.Inventory(
self.ctx, resource_provider=ss,
resource_class='DISK_GB', total=100000, reserved=0,
@ -283,7 +283,7 @@ class ReshapeTestCase(tb.PlacementDbBaseTestCase):
# storage provider.
after_inventories = {
# cn1 keeps the RAM only
cn1.uuid: rp_obj.InventoryList(self.ctx, objects=[
cn1: rp_obj.InventoryList(self.ctx, objects=[
rp_obj.Inventory(
self.ctx, resource_provider=cn1,
resource_class='MEMORY_MB', total=32768, reserved=0,
@ -291,14 +291,14 @@ class ReshapeTestCase(tb.PlacementDbBaseTestCase):
allocation_ratio=1.0),
]),
# each NUMA node gets half of the CPUs
cn1_numa0.uuid: rp_obj.InventoryList(self.ctx, objects=[
cn1_numa0: rp_obj.InventoryList(self.ctx, objects=[
rp_obj.Inventory(
self.ctx, resource_provider=cn1_numa0,
resource_class='VCPU', total=8, reserved=0,
max_unit=8, min_unit=1, step_size=1,
allocation_ratio=1.0),
]),
cn1_numa1.uuid: rp_obj.InventoryList(self.ctx, objects=[
cn1_numa1: rp_obj.InventoryList(self.ctx, objects=[
rp_obj.Inventory(
self.ctx, resource_provider=cn1_numa1,
resource_class='VCPU', total=8, reserved=0,
@ -306,7 +306,7 @@ class ReshapeTestCase(tb.PlacementDbBaseTestCase):
allocation_ratio=1.0),
]),
# The sharing provider gets a bunch of disk
ss.uuid: rp_obj.InventoryList(self.ctx, objects=[
ss: rp_obj.InventoryList(self.ctx, objects=[
rp_obj.Inventory(
self.ctx, resource_provider=ss,
resource_class='DISK_GB', total=100000, reserved=0,