openstack/nova
Code Issues Proposed changes
1957f4d5a2
nova/nova/virt/fake.py
John Garbutt e3c5e22d1f Re-use existing ComputeNode on ironic rebalance 
When a nova-compute service dies that is one of several ironic based
nova-compute services running, a node rebalance occurs to ensure there
is still an active nova-compute service dealing with requests for the
given instance that is running.

Today, when this occurs, we create a new ComputeNode entry. This change
alters that logic to detect the case of the ironic node rebalance and in
that case we re-use the existing ComputeNode entry, simply updating the
host field to match the new host it has been rebalanced onto.

Previously we hit problems with placement when we get a new
ComputeNode.uuid for the same ironic_node.uuid. This reusing of the
existing entry keeps the ComputeNode.uuid the same when the rebalance of
the ComputeNode occurs.

Without keeping the same ComputeNode.uuid placement errors out with a 409
because we attempt to create a ResourceProvider that has the same name
as an existing ResourceProvdier. Had that worked, we would have noticed
the race that occurs after we create the ResourceProvider but before we
add back the existing allocations for existing instances. Keeping the
ComputeNode.uuid the same means we simply look up the existing
ResourceProvider in placement, avoiding all this pain and tears.

Closes-Bug: #1714248
Co-Authored-By: Dmitry Tantsur <dtantsur@redhat.com>
Change-Id: I4253cffca3dbf558c875eed7e77711a31e9e3406
2017-12-11 16:23:49 -05:00

709 lines
24 KiB
Python
Raw Blame History

# Copyright 2010 United States Government as represented by the
# Administrator of the National Aeronautics and Space Administration.
# All Rights Reserved.
# Copyright (c) 2010 Citrix Systems, Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License"); you may
# not use this file except in compliance with the License. You may obtain
# a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
# License for the specific language governing permissions and limitations
# under the License.
"""
A fake (in-memory) hypervisor+api.
Allows nova testing w/o a hypervisor. This module also documents the
semantics of real hypervisor connections.
"""
import collections
import contextlib
import copy
import time
from oslo_log import log as logging
from oslo_serialization import jsonutils
from oslo_utils import versionutils
from nova.compute import power_state
from nova.compute import task_states
from nova.compute import vm_states
import nova.conf
from nova.console import type as ctype
from nova import exception
from nova.objects import diagnostics as diagnostics_obj
from nova.objects import fields as obj_fields
from nova.objects import migrate_data
from nova.virt import driver
from nova.virt import hardware
from nova.virt import virtapi
CONF = nova.conf.CONF
LOG = logging.getLogger(__name__)
_FAKE_NODES = None
def set_nodes(nodes):
"""Sets FakeDriver's node.list.
It has effect on the following methods:
get_available_nodes()
get_available_resource
To restore the change, call restore_nodes()
"""
global _FAKE_NODES
_FAKE_NODES = nodes
def restore_nodes():
"""Resets FakeDriver's node list modified by set_nodes().
Usually called from tearDown().
"""
global _FAKE_NODES
_FAKE_NODES = [CONF.host]
class FakeInstance(object):
def __init__(self, name, state, uuid):
self.name = name
self.state = state
self.uuid = uuid
def __getitem__(self, key):
return getattr(self, key)
class Resources(object):
vcpus = 0
memory_mb = 0
local_gb = 0
vcpus_used = 0
memory_mb_used = 0
local_gb_used = 0
def __init__(self, vcpus=8, memory_mb=8000, local_gb=500):
self.vcpus = vcpus
self.memory_mb = memory_mb
self.local_gb = local_gb
def claim(self, vcpus=0, mem=0, disk=0):
self.vcpus_used += vcpus
self.memory_mb_used += mem
self.local_gb_used += disk
def release(self, vcpus=0, mem=0, disk=0):
self.vcpus_used -= vcpus
self.memory_mb_used -= mem
self.local_gb_used -= disk
def dump(self):
return {
'vcpus': self.vcpus,
'memory_mb': self.memory_mb,
'local_gb': self.local_gb,
'vcpus_used': self.vcpus_used,
'memory_mb_used': self.memory_mb_used,
'local_gb_used': self.local_gb_used
}
class FakeDriver(driver.ComputeDriver):
capabilities = {
"has_imagecache": True,
"supports_recreate": True,
"supports_migrate_to_same_host": True,
"supports_attach_interface": True,
"supports_tagged_attach_interface": True,
"supports_tagged_attach_volume": True,
"supports_extend_volume": True,
}
# Since we don't have a real hypervisor, pretend we have lots of
# disk and ram so this driver can be used to test large instances.
vcpus = 1000
memory_mb = 800000
local_gb = 600000
"""Fake hypervisor driver."""
def __init__(self, virtapi, read_only=False):
super(FakeDriver, self).__init__(virtapi)
self.instances = {}
self.resources = Resources(
vcpus=self.vcpus,
memory_mb=self.memory_mb,
local_gb=self.local_gb)
self.host_status_base = {
'hypervisor_type': 'fake',
'hypervisor_version': versionutils.convert_version_to_int('1.0'),
'hypervisor_hostname': CONF.host,
'cpu_info': {},
'disk_available_least': 0,
'supported_instances': [(
obj_fields.Architecture.X86_64,
obj_fields.HVType.FAKE,
obj_fields.VMMode.HVM)],
'numa_topology': None,
}
self._mounts = {}
self._interfaces = {}
self.active_migrations = {}
self._nodes = self._init_nodes()
def _init_nodes(self):
if not _FAKE_NODES:
set_nodes([CONF.host])
return copy.copy(_FAKE_NODES)
def init_host(self, host):
return
def list_instances(self):
return [self.instances[uuid].name for uuid in self.instances.keys()]
def list_instance_uuids(self):
return list(self.instances.keys())
def plug_vifs(self, instance, network_info):
"""Plug VIFs into networks."""
pass
def unplug_vifs(self, instance, network_info):
"""Unplug VIFs from networks."""
pass
def spawn(self, context, instance, image_meta, injected_files,
admin_password, allocations, network_info=None,
block_device_info=None):
uuid = instance.uuid
state = power_state.RUNNING
flavor = instance.flavor
self.resources.claim(
vcpus=flavor.vcpus,
mem=flavor.memory_mb,
disk=flavor.root_gb)
fake_instance = FakeInstance(instance.name, state, uuid)
self.instances[uuid] = fake_instance
def snapshot(self, context, instance, image_id, update_task_state):
if instance.uuid not in self.instances:
raise exception.InstanceNotRunning(instance_id=instance.uuid)
update_task_state(task_state=task_states.IMAGE_UPLOADING)
def reboot(self, context, instance, network_info, reboot_type,
block_device_info=None, bad_volumes_callback=None):
pass
def get_host_ip_addr(self):
return '192.168.0.1'
def set_admin_password(self, instance, new_pass):
pass
def inject_file(self, instance, b64_path, b64_contents):
pass
def resume_state_on_host_boot(self, context, instance, network_info,
block_device_info=None):
pass
def rescue(self, context, instance, network_info, image_meta,
rescue_password):
pass
def unrescue(self, instance, network_info):
pass
def poll_rebooting_instances(self, timeout, instances):
pass
def migrate_disk_and_power_off(self, context, instance, dest,
flavor, network_info,
block_device_info=None,
timeout=0, retry_interval=0):
pass
def finish_revert_migration(self, context, instance, network_info,
block_device_info=None, power_on=True):
self.instances[instance.uuid] = FakeInstance(
instance.name, power_state.RUNNING, instance.uuid)
def post_live_migration_at_destination(self, context, instance,
network_info,
block_migration=False,
block_device_info=None):
pass
def power_off(self, instance, timeout=0, retry_interval=0):
if instance.uuid in self.instances:
self.instances[instance.uuid].state = power_state.SHUTDOWN
else:
raise exception.InstanceNotFound(instance_id=instance.uuid)
def power_on(self, context, instance, network_info,
block_device_info=None):
if instance.uuid in self.instances:
self.instances[instance.uuid].state = power_state.RUNNING
else:
raise exception.InstanceNotFound(instance_id=instance.uuid)
def trigger_crash_dump(self, instance):
pass
def soft_delete(self, instance):
pass
def restore(self, instance):
pass
def pause(self, instance):
pass
def unpause(self, instance):
pass
def suspend(self, context, instance):
pass
def resume(self, context, instance, network_info, block_device_info=None):
pass
def destroy(self, context, instance, network_info, block_device_info=None,
destroy_disks=True):
key = instance.uuid
if key in self.instances:
flavor = instance.flavor
self.resources.release(
vcpus=flavor.vcpus,
mem=flavor.memory_mb,
disk=flavor.root_gb)
del self.instances[key]
else:
LOG.warning("Key '%(key)s' not in instances '%(inst)s'",
{'key': key,
'inst': self.instances}, instance=instance)
def cleanup(self, context, instance, network_info, block_device_info=None,
destroy_disks=True, migrate_data=None, destroy_vifs=True):
pass
def attach_volume(self, context, connection_info, instance, mountpoint,
disk_bus=None, device_type=None, encryption=None):
"""Attach the disk to the instance at mountpoint using info."""
instance_name = instance.name
if instance_name not in self._mounts:
self._mounts[instance_name] = {}
self._mounts[instance_name][mountpoint] = connection_info
def detach_volume(self, connection_info, instance, mountpoint,
encryption=None):
"""Detach the disk attached to the instance."""
try:
del self._mounts[instance.name][mountpoint]
except KeyError:
pass
def swap_volume(self, old_connection_info, new_connection_info,
instance, mountpoint, resize_to):
"""Replace the disk attached to the instance."""
instance_name = instance.name
if instance_name not in self._mounts:
self._mounts[instance_name] = {}
self._mounts[instance_name][mountpoint] = new_connection_info
def extend_volume(self, connection_info, instance):
"""Extend the disk attached to the instance."""
pass
def attach_interface(self, context, instance, image_meta, vif):
if vif['id'] in self._interfaces:
raise exception.InterfaceAttachFailed(
instance_uuid=instance.uuid)
self._interfaces[vif['id']] = vif
def detach_interface(self, context, instance, vif):
try:
del self._interfaces[vif['id']]
except KeyError:
raise exception.InterfaceDetachFailed(
instance_uuid=instance.uuid)
def get_info(self, instance):
if instance.uuid not in self.instances:
raise exception.InstanceNotFound(instance_id=instance.uuid)
i = self.instances[instance.uuid]
return hardware.InstanceInfo(state=i.state)
def get_diagnostics(self, instance):
return {'cpu0_time': 17300000000,
'memory': 524288,
'vda_errors': -1,
'vda_read': 262144,
'vda_read_req': 112,
'vda_write': 5778432,
'vda_write_req': 488,
'vnet1_rx': 2070139,
'vnet1_rx_drop': 0,
'vnet1_rx_errors': 0,
'vnet1_rx_packets': 26701,
'vnet1_tx': 140208,
'vnet1_tx_drop': 0,
'vnet1_tx_errors': 0,
'vnet1_tx_packets': 662,
}
def get_instance_diagnostics(self, instance):
diags = diagnostics_obj.Diagnostics(
state='running', driver='libvirt', hypervisor='kvm',
hypervisor_os='ubuntu', uptime=46664, config_drive=True)
diags.add_cpu(id=0, time=17300000000, utilisation=15)
diags.add_nic(mac_address='01:23:45:67:89:ab',
rx_octets=2070139,
rx_errors=100,
rx_drop=200,
rx_packets=26701,
rx_rate=300,
tx_octets=140208,
tx_errors=400,
tx_drop=500,
tx_packets = 662,
tx_rate=600)
diags.add_disk(read_bytes=262144,
read_requests=112,
write_bytes=5778432,
write_requests=488,
errors_count=1)
diags.memory_details = diagnostics_obj.MemoryDiagnostics(
maximum=524288, used=0)
return diags
def get_all_bw_counters(self, instances):
"""Return bandwidth usage counters for each interface on each
running VM.
"""
bw = []
for instance in instances:
bw.append({'uuid': instance.uuid,
'mac_address': 'fa:16:3e:4c:2c:30',
'bw_in': 0,
'bw_out': 0})
return bw
def get_all_volume_usage(self, context, compute_host_bdms):
"""Return usage info for volumes attached to vms on
a given host.
"""
volusage = []
return volusage
def get_host_cpu_stats(self):
stats = {'kernel': 5664160000000,
'idle': 1592705190000000,
'user': 26728850000000,
'iowait': 6121490000000}
stats['frequency'] = 800
return stats
def block_stats(self, instance, disk_id):
return [0, 0, 0, 0, None]
def get_console_output(self, context, instance):
return 'FAKE CONSOLE OUTPUT\nANOTHER\nLAST LINE'
def get_vnc_console(self, context, instance):
return ctype.ConsoleVNC(internal_access_path='FAKE',
host='fakevncconsole.com',
port=6969)
def get_spice_console(self, context, instance):
return ctype.ConsoleSpice(internal_access_path='FAKE',
host='fakespiceconsole.com',
port=6969,
tlsPort=6970)
def get_rdp_console(self, context, instance):
return ctype.ConsoleRDP(internal_access_path='FAKE',
host='fakerdpconsole.com',
port=6969)
def get_serial_console(self, context, instance):
return ctype.ConsoleSerial(internal_access_path='FAKE',
host='fakerdpconsole.com',
port=6969)
def get_mks_console(self, context, instance):
return ctype.ConsoleMKS(internal_access_path='FAKE',
host='fakemksconsole.com',
port=6969)
def get_console_pool_info(self, console_type):
return {'address': '127.0.0.1',
'username': 'fakeuser',
'password': 'fakepassword'}
def refresh_security_group_rules(self, security_group_id):
return True
def refresh_instance_security_rules(self, instance):
return True
def get_available_resource(self, nodename):
"""Updates compute manager resource info on ComputeNode table.
Since we don't have a real hypervisor, pretend we have lots of
disk and ram.
"""
cpu_info = collections.OrderedDict([
('arch', 'x86_64'),
('model', 'Nehalem'),
('vendor', 'Intel'),
('features', ['pge', 'clflush']),
('topology', {
'cores': 1,
'threads': 1,
'sockets': 4,
}),
])
if nodename not in self._nodes:
return {}
host_status = self.host_status_base.copy()
host_status.update(self.resources.dump())
host_status['hypervisor_hostname'] = nodename
host_status['host_hostname'] = nodename
host_status['host_name_label'] = nodename
host_status['cpu_info'] = jsonutils.dumps(cpu_info)
return host_status
def ensure_filtering_rules_for_instance(self, instance, network_info):
return
def get_instance_disk_info(self, instance, block_device_info=None):
return
def live_migration(self, context, instance, dest,
post_method, recover_method, block_migration=False,
migrate_data=None):
post_method(context, instance, dest, block_migration,
migrate_data)
return
def live_migration_force_complete(self, instance):
return
def live_migration_abort(self, instance):
return
def cleanup_live_migration_destination_check(self, context,
dest_check_data):
return
def check_can_live_migrate_destination(self, context, instance,
src_compute_info, dst_compute_info,
block_migration=False,
disk_over_commit=False):
data = migrate_data.LibvirtLiveMigrateData()
data.filename = 'fake'
data.image_type = CONF.libvirt.images_type
data.graphics_listen_addr_vnc = CONF.vnc.server_listen
data.graphics_listen_addr_spice = CONF.spice.server_listen
data.serial_listen_addr = None
# Notes(eliqiao): block_migration and disk_over_commit are not
# nullable, so just don't set them if they are None
if block_migration is not None:
data.block_migration = block_migration
if disk_over_commit is not None:
data.disk_over_commit = disk_over_commit
data.disk_available_mb = 100000
data.is_shared_block_storage = True
data.is_shared_instance_path = True
return data
def check_can_live_migrate_source(self, context, instance,
dest_check_data, block_device_info=None):
return dest_check_data
def finish_migration(self, context, migration, instance, disk_info,
network_info, image_meta, resize_instance,
block_device_info=None, power_on=True):
return
def confirm_migration(self, context, migration, instance, network_info):
return
def pre_live_migration(self, context, instance, block_device_info,
network_info, disk_info, migrate_data):
return migrate_data
def unfilter_instance(self, instance, network_info):
return
def _test_remove_vm(self, instance_uuid):
"""Removes the named VM, as if it crashed. For testing."""
self.instances.pop(instance_uuid)
def host_power_action(self, action):
"""Reboots, shuts down or powers up the host."""
return action
def host_maintenance_mode(self, host, mode):
"""Start/Stop host maintenance window. On start, it triggers
guest VMs evacuation.
"""
if not mode:
return 'off_maintenance'
return 'on_maintenance'
def set_host_enabled(self, enabled):
"""Sets the specified host's ability to accept new instances."""
if enabled:
return 'enabled'
return 'disabled'
def get_volume_connector(self, instance):
return {'ip': CONF.my_block_storage_ip,
'initiator': 'fake',
'host': 'fakehost'}
def get_available_nodes(self, refresh=False):
return self._nodes
def instance_on_disk(self, instance):
return False
def quiesce(self, context, instance, image_meta):
pass
def unquiesce(self, context, instance, image_meta):
pass
class FakeVirtAPI(virtapi.VirtAPI):
@contextlib.contextmanager
def wait_for_instance_event(self, instance, event_names, deadline=300,
error_callback=None):
# NOTE(danms): Don't actually wait for any events, just
# fall through
yield
class SmallFakeDriver(FakeDriver):
# The api samples expect specific cpu memory and disk sizes. In order to
# allow the FakeVirt driver to be used outside of the unit tests, provide
# a separate class that has the values expected by the api samples. So
# instead of requiring new samples every time those
# values are adjusted allow them to be overwritten here.
vcpus = 2
memory_mb = 8192
local_gb = 1028
class MediumFakeDriver(FakeDriver):
# Fake driver that has enough resources to host more than one instance
# but not that much that cannot be exhausted easily
vcpus = 10
memory_mb = 8192
local_gb = 1028
class FakeRescheduleDriver(FakeDriver):
"""FakeDriver derivative that triggers a reschedule on the first spawn
attempt. This is expected to only be used in tests that have more than
one compute service.
"""
# dict, keyed by instance uuid, mapped to a boolean telling us if the
# instance has been rescheduled or not
rescheduled = {}
def spawn(self, context, instance, image_meta, injected_files,
admin_password, allocations, network_info=None,
block_device_info=None):
if not self.rescheduled.get(instance.uuid, False):
# We only reschedule on the first time something hits spawn().
self.rescheduled[instance.uuid] = True
raise exception.ComputeResourcesUnavailable(
reason='FakeRescheduleDriver')
super(FakeRescheduleDriver, self).spawn(
context, instance, image_meta, injected_files,
admin_password, allocations, network_info, block_device_info)
class FakeBuildAbortDriver(FakeDriver):
"""FakeDriver derivative that always fails on spawn() with a
BuildAbortException so no reschedule is attempted.
"""
def spawn(self, context, instance, image_meta, injected_files,
admin_password, allocations, network_info=None,
block_device_info=None):
raise exception.BuildAbortException(
instance_uuid=instance.uuid, reason='FakeBuildAbortDriver')
class FakeUnshelveSpawnFailDriver(FakeDriver):
"""FakeDriver derivative that always fails on spawn() with a
VirtualInterfaceCreateException when unshelving an offloaded instance.
"""
def spawn(self, context, instance, image_meta, injected_files,
admin_password, allocations, network_info=None,
block_device_info=None):
if instance.vm_state == vm_states.SHELVED_OFFLOADED:
raise exception.VirtualInterfaceCreateException(
'FakeUnshelveSpawnFailDriver')
# Otherwise spawn normally during the initial build.
super(FakeUnshelveSpawnFailDriver, self).spawn(
context, instance, image_meta, injected_files,
admin_password, allocations, network_info, block_device_info)
class FakeLiveMigrateDriver(FakeDriver):
"""FakeDriver derivative to handle force_complete and abort calls.
This module serves those tests that need to abort or force-complete
the live migration, thus the live migration will never be finished
without the force_complete_migration or delete_migration API calls.
"""
def __init__(self, virtapi, read_only=False):
super(FakeLiveMigrateDriver, self).__init__(virtapi, read_only)
self._migrating = True
self._abort_migration = True
def live_migration(self, context, instance, dest,
post_method, recover_method, block_migration=False,
migrate_data=None):
self._abort_migration = False
self._migrating = True
while self._migrating:
time.sleep(0.1)
if self._abort_migration:
recover_method(context, instance, dest, migrate_data,
migration_status='cancelled')
else:
post_method(context, instance, dest, block_migration,
migrate_data)
def live_migration_force_complete(self, instance):
self._migrating = False
def live_migration_abort(self, instance):
self._abort_migration = True
self._migrating = False
View Git Blame Copy Permalink