nova-powervm/nova_powervm/virt/powervm/driver.py

1907 lines
82 KiB
Python

# Copyright IBM Corp. and contributors
#
# All Rights Reserved.
#
# 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.
from nova import block_device
from nova.compute import task_states
from nova.compute import utils as compute_utils
from nova.compute import vm_states
from nova.console import type as console_type
from nova import context as ctx
from nova import exception
from nova import image
from nova import objects
from nova import utils as n_utils
from nova.virt import configdrive
from nova.virt import driver
import os_resource_classes as orc
from oslo_log import log as logging
from oslo_utils import importutils
import six
from taskflow.patterns import linear_flow as tf_lf
import time
from pypowervm import adapter as pvm_apt
from pypowervm import const as pvm_const
from pypowervm import exceptions as pvm_exc
from pypowervm.helpers import log_helper as log_hlp
from pypowervm.helpers import vios_busy as vio_hlp
from pypowervm.tasks import cna as pvm_cna
from pypowervm.tasks import memory as pvm_mem
from pypowervm.tasks.monitor import host_cpu as pvm_hcpu
from pypowervm.tasks import partition as pvm_par
from pypowervm.tasks import power_opts as pvm_popts
from pypowervm.tasks import scsi_mapper as pvm_smap
from pypowervm.tasks import storage as pvm_stor
from pypowervm.tasks import vterm as pvm_vterm
from pypowervm import util as pvm_util
from pypowervm.wrappers import managed_system as pvm_ms
from nova_powervm import conf as cfg
from nova_powervm.virt.powervm.disk import driver as disk_dvr
from nova_powervm.virt.powervm import event
from nova_powervm.virt.powervm import host as pvm_host
from nova_powervm.virt.powervm.i18n import _
from nova_powervm.virt.powervm import image as img
from nova_powervm.virt.powervm import live_migration as lpm
from nova_powervm.virt.powervm import media
from nova_powervm.virt.powervm.nvram import manager as nvram_manager
from nova_powervm.virt.powervm import slot
from nova_powervm.virt.powervm.tasks import base as tf_base
from nova_powervm.virt.powervm.tasks import image as tf_img
from nova_powervm.virt.powervm.tasks import network as tf_net
from nova_powervm.virt.powervm.tasks import slot as tf_slot
from nova_powervm.virt.powervm.tasks import storage as tf_stg
from nova_powervm.virt.powervm.tasks import vm as tf_vm
from nova_powervm.virt.powervm import vm
from nova_powervm.virt.powervm import volume as vol_attach
from nova_powervm.virt.powervm.volume import iscsi
LOG = logging.getLogger(__name__)
CONF = cfg.CONF
DISK_ADPT_NS = 'nova_powervm.virt.powervm.disk'
DISK_ADPT_MAPPINGS = {
'localdisk': 'localdisk.LocalStorage',
'ssp': 'ssp.SSPDiskAdapter'
}
# NVRAM store APIs for the NVRAM manager to use
NVRAM_NS = 'nova_powervm.virt.powervm.nvram.'
NVRAM_APIS = {
'swift': 'swift.SwiftNvramStore',
}
KEEP_NVRAM_STATES = {vm_states.SHELVED, }
FETCH_NVRAM_STATES = {vm_states.SHELVED, vm_states.SHELVED_OFFLOADED}
class PowerVMDriver(driver.ComputeDriver):
"""PowerVM Implementation of Compute Driver."""
def __init__(self, virtapi):
self.capabilities = {
# NOTE(edmondsw): 'has_imagecache' will be set dynamically
"supports_migrate_to_same_host": False,
"supports_attach_interface": True,
"supports_device_tagging": False,
"supports_tagged_attach_interface": False,
"supports_tagged_attach_volume": False,
"supports_extend_volume": True,
"supports_multiattach": False,
"supports_evacuate": True,
}
super(PowerVMDriver, self).__init__(virtapi)
def init_host(self, host):
"""Initialize anything that is necessary for the driver to function.
Includes catching up with currently running VM's on the given host.
"""
# Live migrations
self.live_migrations = {}
# Set the nvram mgr to None so events are not handled until it's setup
self.nvram_mgr = None
self.store_api = None
# Get an adapter
self._get_adapter()
# First need to resolve the managed host UUID
self._get_host_uuid()
# Make sure the Virtual I/O Server(s) are available.
pvm_par.validate_vios_ready(self.adapter)
# Do a scrub of the I/O plane to make sure the system is in good shape
LOG.info("Clearing stale I/O connections on driver init.")
pvm_stor.ComprehensiveScrub(self.adapter).execute()
# Initialize the disk adapter. Sets self.disk_dvr
self._setup_disk_adapter()
self.image_api = image.API()
self._setup_rebuild_store()
# Init Host CPU Statistics
self.host_cpu_cache = pvm_hcpu.HostCPUMetricCache(self.adapter,
self.host_uuid)
# Cache for instance overhead.
# Key: max_mem (int MB)
# Value: overhead (int MB)
self._inst_overhead_cache = {}
# Clean-up any orphan adapters
self._cleanup_orphan_adapters(CONF.powervm.pvm_vswitch_for_novalink_io)
LOG.info("The compute driver has been initialized.")
def cleanup_host(self, host):
"""Clean up anything that is necessary for the driver gracefully stop.
Includes ending remote sessions. This is optional.
"""
# Stop listening for events
try:
self.session.get_event_listener().shutdown()
except Exception:
pass
LOG.info("The compute driver has been shutdown.")
def _get_adapter(self):
# Build the adapter. May need to attempt the connection multiple times
# in case the REST server is starting.
self.session = pvm_apt.Session(conn_tries=300)
self.adapter = pvm_apt.Adapter(
self.session, helpers=[log_hlp.log_helper,
vio_hlp.vios_busy_retry_helper])
# Register the event handler
eh = event.PowerVMNovaEventHandler(self)
self.session.get_event_listener().subscribe(eh)
def _setup_disk_adapter(self):
"""Set up the nova ephemeral disk adapter."""
self.disk_dvr = importutils.import_object_ns(
DISK_ADPT_NS, DISK_ADPT_MAPPINGS[CONF.powervm.disk_driver.lower()],
self.adapter, self.host_uuid)
has_imagecache = self.disk_dvr.capabilities['has_imagecache']
self.capabilities['has_imagecache'] = has_imagecache
def manage_image_cache(self, context, all_instances):
self.disk_dvr.manage_image_cache(context, all_instances)
def _setup_rebuild_store(self):
"""Set up the store for remote restart objects."""
store = CONF.powervm.nvram_store.lower()
if store != 'none':
self.store_api = importutils.import_object(
NVRAM_NS + NVRAM_APIS[store])
# Events will be handled once the nvram_mgr is set.
self.nvram_mgr = nvram_manager.NvramManager(
self.store_api, self.adapter, self.host_uuid)
# Do host startup for NVRAM for existing VMs on the host
n_utils.spawn(self._nvram_host_startup)
def _nvram_host_startup(self):
"""NVRAM Startup.
When the compute node starts up, it's not known if any NVRAM events
were missed when the compute process was not running. During startup
put each LPAR on the queue to be updated, just incase.
"""
for lpar_w in vm.get_lpars(self.adapter):
# Find the instance for the LPAR.
inst = vm.get_instance(ctx.get_admin_context(), lpar_w.uuid)
if inst is not None and inst.host == CONF.host:
self.nvram_mgr.store(inst)
time.sleep(0)
def _get_host_uuid(self):
"""Get the System wrapper and its UUID for the (single) host."""
syswraps = pvm_ms.System.get(self.adapter)
if len(syswraps) != 1:
raise Exception(
_("Expected exactly one host; found %d"), len(syswraps))
self.host_wrapper = syswraps[0]
self.host_uuid = self.host_wrapper.uuid
LOG.info("Host UUID is:%s", self.host_uuid)
@staticmethod
def _log_operation(op, instance):
"""Log entry point of driver operations."""
LOG.info('Operation: %(op)s. Virtual machine display name: '
'%(display_name)s, name: %(name)s',
{'op': op, 'display_name': instance.display_name,
'name': instance.name}, instance=instance)
def get_info(self, instance, use_cache=True):
"""Get the current status of an instance.
:param instance: nova.objects.instance.Instance object
:param use_cache: unused in this driver
:returns: An InstanceInfo object
"""
return vm.get_vm_info(self.adapter, instance)
def instance_exists(self, instance):
"""Checks existence of an instance on the host.
:param instance: The instance to lookup
Returns True if an instance with the supplied ID exists on
the host, False otherwise.
"""
return vm.instance_exists(self.adapter, instance)
def estimate_instance_overhead(self, instance_info):
"""Estimate the virtualization overhead required to build an instance.
Defaults to zero, Per-instance overhead calculations are desired.
:param instance_info: Instance/flavor to calculate overhead for.
It can be Instance or Flavor object or a simple dict. The dict is
expected to contain:
{ 'memory_mb': <int>, 'extra_specs': {'powervm:max_mem': <int> }}
Values not found will default to zero.
:return: Dict of estimated overhead values {'memory_mb': overhead}
"""
# Check if input passed is an object instance then extract Flavor
if isinstance(instance_info, objects.Instance):
instance_info = instance_info.get_flavor()
# If the instance info passed is dict then create Flavor object.
elif isinstance(instance_info, dict):
instance_info = objects.Flavor(**instance_info)
max_mem = 0
overhead = 0
try:
cur_mem = instance_info.memory_mb
if hasattr(instance_info, 'extra_specs'):
if 'powervm:max_mem' in instance_info.extra_specs.keys():
mem = instance_info.extra_specs.get('powervm:max_mem',
max_mem)
max_mem = int(mem)
max_mem = max(cur_mem, max_mem)
if max_mem in self._inst_overhead_cache:
overhead = self._inst_overhead_cache[max_mem]
else:
overhead, avail = pvm_mem.calculate_memory_overhead_on_host(
self.adapter, self.host_uuid, {'max_mem': max_mem})
self._inst_overhead_cache[max_mem] = overhead
except Exception:
LOG.exception("PowerVM error estimating instance overhead.")
finally:
return {'memory_mb': overhead}
def list_instances(self):
"""Return the names of all the instances known to the virt host.
:return: VM Names as a list.
"""
lpar_list = vm.get_lpar_names(self.adapter)
return lpar_list
def get_host_cpu_stats(self):
"""Return the current CPU state of the host."""
self.host_cpu_cache.refresh()
total_cycles = self.host_cpu_cache.total_cycles
total_user_cycles = self.host_cpu_cache.total_user_cycles
total_fw_cycles = self.host_cpu_cache.total_fw_cycles
return {
'kernel': self.host_cpu_cache.total_fw_cycles,
'user': self.host_cpu_cache.total_user_cycles,
'idle': (total_cycles - total_user_cycles - total_fw_cycles),
# Not reported by PowerVM
'iowait': 0,
'frequency': self.host_cpu_cache.cpu_freq}
def instance_on_disk(self, instance):
"""Checks access of instance files on the host.
:param instance: nova.objects.instance.Instance to lookup
Returns True if files of an instance with the supplied ID accessible on
the host, False otherwise.
.. note::
Used in rebuild for HA implementation and required for validation
of access to instance shared disk files
"""
# If the instance is booted from volume then we shouldn't
# really care if instance "disks" are on shared storage.
context = ctx.get_admin_context()
block_device_info = self._get_block_device_info(context, instance)
if self._is_booted_from_volume(block_device_info):
LOG.debug('Instance booted from volume.', instance=instance)
return True
# If configured for shared storage, see if we can find the disks
if self.disk_dvr.capabilities['shared_storage']:
LOG.debug('Looking for instance disks on shared storage.',
instance=instance)
# Try to get a reference to the disk
try:
if self.disk_dvr.get_disk_ref(instance,
disk_dvr.DiskType.BOOT):
LOG.debug('Disks found on shared storage.',
instance=instance)
return True
except Exception:
LOG.exception("PowerVM error checking instance on disk.",
instance=instance)
LOG.debug('Instance disks not found on this host.', instance=instance)
return False
def spawn(self, context, instance, image_meta, injected_files,
admin_password, allocations, network_info=None,
block_device_info=None):
"""Create a new instance/VM/domain on the virtualization platform.
Once this successfully completes, the instance should be
running (power_state.RUNNING).
If this fails, any partial instance should be completely
cleaned up, and the virtualization platform should be in the state
that it was before this call began.
:param context: security context
:param instance: nova.objects.instance.Instance
This function should use the data there to guide
the creation of the new instance.
:param nova.objects.ImageMeta image_meta:
The metadata of the image of the instance.
:param injected_files: User files to inject into instance.
:param admin_password: Administrator password to set in instance.
:param allocations: Information about resources allocated to the
instance via placement, of the form returned by
SchedulerReportClient.get_allocations_for_consumer.
:param network_info: instance network information
:param block_device_info: Information about block devices to be
attached to the instance.
"""
self._log_operation('spawn', instance)
# Extract the block devices.
bdms = self._extract_bdm(block_device_info)
# Define the flow
flow_spawn = tf_lf.Flow("spawn")
# Determine if this is a VM recreate
task_state = instance.task_state
rebuild_spawning = task_states.REBUILD_SPAWNING
recreate = (task_state == rebuild_spawning and 'id' not in image_meta)
# Create the transaction manager (FeedTask) for Storage I/O.
xag = self._get_inst_xag(instance, bdms, recreate=recreate)
stg_ftsk = pvm_par.build_active_vio_feed_task(self.adapter, xag=xag)
# Build the PowerVM Slot lookup map. Only the recreate action needs
# the volume driver iterator (to look up volumes and their client
# mappings).
vol_drv_iter = (self._vol_drv_iter(instance, bdms, stg_ftsk=stg_ftsk)
if recreate else None)
slot_mgr = slot.build_slot_mgr(
instance, self.store_api, adapter=self.adapter,
vol_drv_iter=vol_drv_iter)
# Create the LPAR, check if NVRAM restore is needed.
vm_state = instance.vm_state
if self.nvram_mgr and (recreate or vm_state in FETCH_NVRAM_STATES):
nvram_mgr = self.nvram_mgr
else:
nvram_mgr = None
# If we're recreating pass None in for the FeedTask. This will make the
# Create task produce a FeedTask that will be used to scrub stale
# adapters immediately after the LPAR is created.
flow_spawn.add(tf_vm.Create(
self.adapter, self.host_wrapper, instance,
stg_ftsk=(None if recreate else stg_ftsk), nvram_mgr=nvram_mgr,
slot_mgr=slot_mgr))
# Create a flow for the IO
flow_spawn.add(tf_net.PlugVifs(
self.virtapi, self.adapter, instance, network_info,
self.host_uuid, slot_mgr))
flow_spawn.add(tf_net.PlugMgmtVif(
self.adapter, instance, self.host_uuid, slot_mgr))
# Only add the image disk if this is from Glance.
if not self._is_booted_from_volume(block_device_info):
# If a recreate, just hookup the existing disk on shared storage.
if recreate:
flow_spawn.add(tf_stg.FindDisk(
self.disk_dvr, context, instance, disk_dvr.DiskType.BOOT))
else:
# Creates the boot image.
flow_spawn.add(tf_stg.CreateDiskForImg(
self.disk_dvr, context, instance, image_meta))
# Connects up the disk to the LPAR
flow_spawn.add(tf_stg.ConnectDisk(
self.disk_dvr, instance, stg_ftsk=stg_ftsk))
# Determine if there are volumes to connect. If so, add a connection
# for each type.
self._add_volume_connection_tasks(
context, instance, bdms, flow_spawn, stg_ftsk, slot_mgr)
# If the config drive is needed, add those steps. Should be done
# after all the other I/O.
if configdrive.required_by(instance) and not recreate:
flow_spawn.add(tf_stg.CreateAndConnectCfgDrive(
self.adapter, instance, injected_files,
network_info, admin_password, stg_ftsk=stg_ftsk))
# Add the transaction manager flow to the end of the 'I/O
# connection' tasks. This will run all the connections in parallel.
flow_spawn.add(stg_ftsk)
# Update load source of IBMi VM
distro = instance.system_metadata.get('image_os_distro', '')
if distro.lower() == img.OSDistro.OS400:
boot_type = self._get_boot_connectivity_type(
bdms, block_device_info)
flow_spawn.add(tf_vm.UpdateIBMiSettings(
self.adapter, instance, boot_type))
# Save the slot map information
flow_spawn.add(tf_slot.SaveSlotStore(instance, slot_mgr))
pwr_opts = pvm_popts.PowerOnOpts()
if CONF.powervm.remove_vopt_media_on_boot:
media_name = media.ConfigDrivePowerVM.get_cfg_drv_name(instance)
pwr_opts.remove_optical(
media_name, time=CONF.powervm.remove_vopt_media_time)
# Last step is to power on the system.
flow_spawn.add(tf_vm.PowerOn(self.adapter, instance,
pwr_opts=pwr_opts))
# Run the flow.
tf_base.run(flow_spawn, instance=instance)
def _add_volume_connection_tasks(self, context, instance, bdms,
flow, stg_ftsk, slot_mgr):
"""Determine if there are volumes to connect to this instance.
If there are volumes to connect to this instance add a task to the
flow for each volume.
:param context: security context.
:param instance: Instance object as returned by DB layer.
:param bdms: block device mappings.
:param flow: the flow to add the tasks to.
:param stg_ftsk: the storage task flow.
:param slot_mgr: A NovaSlotManager. Used to store/retrieve the client
slots used when a volume is attached to a VM.
"""
for bdm, vol_drv in self._vol_drv_iter(instance, bdms,
stg_ftsk=stg_ftsk):
# First connect the volume. This will update the
# connection_info.
flow.add(tf_stg.ConnectVolume(vol_drv, slot_mgr))
# Save the BDM so that the updated connection info is
# persisted.
flow.add(tf_stg.SaveBDM(bdm, instance))
def _add_volume_disconnection_tasks(self, context, instance, bdms,
flow, stg_ftsk, slot_mgr):
"""Determine if there are volumes to disconnect from this instance.
If there are volumes to disconnect from this instance add a task to the
flow for each volume.
:param context: security context.
:param instance: Instance object as returned by DB layer.
:param bdms: block device mappings.
:param flow: the flow to add the tasks to.
:param stg_ftsk: the storage task flow.
:param slot_mgr: A NovaSlotManager. Used to store/retrieve the client
slots used when a volume is detached from a VM.
"""
# TODO(thorst) Do we need to do something on the disconnect for slots?
for bdm, vol_drv in self._vol_drv_iter(instance, bdms,
stg_ftsk=stg_ftsk):
flow.add(tf_stg.DisconnectVolume(vol_drv, slot_mgr))
def _get_block_device_info(self, context, instance):
"""Retrieves the instance's block_device_info."""
bdms = objects.BlockDeviceMappingList.get_by_instance_uuid(
context, instance.uuid)
return driver.get_block_device_info(instance, bdms)
def _is_booted_from_volume(self, block_device_info):
"""Determine whether the root device is listed in block_device_info.
If it is, this can be considered a 'boot from Cinder Volume'.
:param block_device_info: The block device info from the compute
manager.
:return: True if the root device is in block_device_info and False if
it is not.
"""
if block_device_info is None:
return False
root_bdm = block_device.get_root_bdm(
driver.block_device_info_get_mapping(block_device_info))
return root_bdm is not None
@property
def need_legacy_block_device_info(self):
return False
def _destroy(self, context, instance, block_device_info=None,
network_info=None, destroy_disks=True, shutdown=True):
"""Internal destroy method used by multiple operations.
:param context: security context
:param instance: Instance object as returned by DB layer.
:param block_device_info: Information about block devices that should
be detached from the instance.
This can be None when destroying the original
VM during confirm resize/migration. In that
case, the storage mappings have already been
removed from the original VM, so no work to
do.
:param network_info: The network information associated with the
instance
:param destroy_disks: Indicates if disks should be destroyed
:param shutdown: Indicate whether to shutdown the VM first
"""
def _setup_flow_and_run():
# Extract the block devices.
bdms = self._extract_bdm(block_device_info)
# Define the flow
flow = tf_lf.Flow("destroy")
if shutdown:
# Power Off the LPAR. If its disks are about to be deleted,
# VSP hard shutdown it.
flow.add(tf_vm.PowerOff(self.adapter, instance,
force_immediate=destroy_disks))
# Create the transaction manager (FeedTask) for Storage I/O.
xag = self._get_inst_xag(instance, bdms)
stg_ftsk = pvm_par.build_active_vio_feed_task(self.adapter,
xag=xag)
# Build the PowerVM Slot lookup map.
slot_mgr = slot.build_slot_mgr(instance, self.store_api)
# Call the unplug VIFs task. While CNAs get removed from the LPAR
# directly on the destroy, this clears up the I/O Host side.
flow.add(tf_vm.Get(self.adapter, self.host_uuid, instance))
flow.add(tf_net.UnplugVifs(self.adapter, instance, network_info,
self.host_uuid, slot_mgr))
# Add the disconnect/deletion of the vOpt to the transaction
# manager.
if configdrive.required_by(instance):
flow.add(tf_stg.DeleteVOpt(
self.adapter, instance, stg_ftsk=stg_ftsk))
# Determine if there are volumes to disconnect. If so, remove each
# volume (within the transaction manager)
self._add_volume_disconnection_tasks(
context, instance, bdms, flow, stg_ftsk, slot_mgr)
# Only detach the disk adapters if this is not a boot from volume
# since volumes are handled above. This is only for disks.
destroy_disk_task = None
if not self._is_booted_from_volume(block_device_info):
# Detach the disk storage adapters (when the stg_ftsk runs)
flow.add(tf_stg.DetachDisk(
self.disk_dvr, instance, stg_ftsk=stg_ftsk))
# Delete the storage disks
if destroy_disks:
destroy_disk_task = tf_stg.DeleteDisk(
self.disk_dvr, instance)
# It's possible that volume disconnection may have failed for disks
# which had been removed from the VIOS by the storage back end
# (e.g. if we're destroying an evacuated instance). In that case,
# the storage is already gone, but we had no way to identify its
# mappings because no disk name/UDID. We now remove such mappings
# based on their association with the LPAR ID.
def _rm_vscsi_maps(vwrap):
removals = pvm_smap.remove_maps(vwrap, pvm_inst_uuid)
if removals:
LOG.warning("Removing %(num_maps)d storage-less VSCSI "
"mappings associated with LPAR ID "
"%(lpar_uuid)s from VIOS %(vios_name)s.",
{'num_maps': len(removals),
'lpar_uuid': pvm_inst_uuid,
'vios_name': vwrap.name}, instance=instance)
return removals
stg_ftsk.add_functor_subtask(_rm_vscsi_maps)
# Add the transaction manager flow to the end of the 'storage
# connection' tasks. This will run all the disconnection ops
# in parallel
flow.add(stg_ftsk)
# The disks shouldn't be destroyed until the unmappings are done.
if destroy_disk_task:
flow.add(destroy_disk_task)
# Last step is to delete the LPAR from the system and delete
# the NVRAM from the store.
# Note: If moving to a Graph Flow, will need to change to depend on
# the prior step.
flow.add(tf_vm.Delete(self.adapter, instance))
if (destroy_disks and instance.vm_state not in KEEP_NVRAM_STATES):
if instance.host in [None, CONF.host]:
# If the disks are being destroyed and not one of the
# operations that we should keep the NVRAM around for, then
# it's probably safe to delete the NVRAM from the store.
flow.add(tf_vm.DeleteNvram(self.nvram_mgr, instance))
flow.add(tf_slot.DeleteSlotStore(instance, slot_mgr))
# Run the flow
tf_base.run(flow, instance=instance)
try:
pvm_inst_uuid = vm.get_pvm_uuid(instance)
_setup_flow_and_run()
except exception.InstanceNotFound:
LOG.warning('VM was not found during destroy operation.',
instance=instance)
return
except Exception as e:
LOG.exception("PowerVM error destroying instance.",
instance=instance)
# Convert to a Nova exception
raise exception.InstanceTerminationFailure(
reason=six.text_type(e))
def destroy(self, context, instance, network_info, block_device_info=None,
destroy_disks=True, migrate_data=None):
"""Destroy (shutdown and delete) the specified instance.
If the instance is not found (for example if networking failed), this
function should still succeed. It's probably a good idea to log a
warning in that case.
:param context: security context
:param instance: Instance object as returned by DB layer.
:param network_info:
:py:meth:`~nova.network.manager.NetworkManager.get_instance_nw_info`
:param block_device_info: Information about block devices that should
be detached from the instance.
:param destroy_disks: Indicates if disks should be destroyed
:param migrate_data: a LiveMigrateData object
"""
if instance.task_state == task_states.RESIZE_REVERTING:
LOG.info('Destroy called for migrated/resized instance.',
instance=instance)
# This destroy is part of resize or migrate. It's called to
# revert the resize/migration on the destination host.
# Get the VM and see if we've renamed it to the resize name,
# if not delete as usual because then we know it's not the
# original VM.
pvm_inst_uuid = vm.get_pvm_uuid(instance)
vm_name = vm.get_vm_qp(self.adapter, pvm_inst_uuid,
qprop='PartitionName', log_errors=False)
if vm_name == self._gen_resize_name(instance, same_host=True):
# Since it matches it must have been a resize, don't delete it!
LOG.info('Ignoring destroy call during resize revert.',
instance=instance)
return
# Run the destroy
self._log_operation('destroy', instance)
self._destroy(
context, instance, block_device_info=block_device_info,
network_info=network_info, destroy_disks=destroy_disks,
shutdown=True)
def attach_volume(self, context, connection_info, instance, mountpoint,
disk_bus=None, device_type=None, encryption=None):
"""Attach the volume to the instance at mountpoint using info."""
self._log_operation('attach_volume', instance)
# Define the flow
flow = tf_lf.Flow("attach_volume")
# Get the LPAR Wrapper
flow.add(tf_vm.Get(self.adapter, self.host_uuid, instance))
# Determine if there are volumes to connect. If so, add a connection
# for each type.
slot_mgr = slot.build_slot_mgr(instance, self.store_api)
vol_drv = vol_attach.build_volume_driver(
self.adapter, self.host_uuid, instance, connection_info)
flow.add(tf_stg.ConnectVolume(vol_drv, slot_mgr))
# Save the new slot info
flow.add(tf_slot.SaveSlotStore(instance, slot_mgr))
# Run the flow
tf_base.run(flow, instance=instance)
# The volume connector may have updated the system metadata. Save
# the instance to persist the data. Spawn/destroy auto saves instance,
# but the attach does not. Detach does not need this save - as the
# detach flows do not (currently) modify system metadata. May need
# to revise in the future as volume connectors evolve.
instance.save()
def extend_volume(self, connection_info, instance, requested_size):
"""Extend the disk attached to the instance.
:param dict connection_info: The connection for the extended volume.
:param nova.objects.instance.Instance instance:
The instance whose volume gets extended.
:param int requested_size: The requested new volume size in bytes. This
parameter is unused by this driver.
:return: None
"""
vol_drv = vol_attach.build_volume_driver(
self.adapter, self.host_uuid, instance, connection_info)
vol_drv.extend_volume()
def detach_volume(self, context, connection_info, instance, mountpoint,
encryption=None):
"""Detach the volume attached to the instance."""
self._log_operation('detach_volume', instance)
# Define the flow
flow = tf_lf.Flow("detach_volume")
# Get a volume adapter for this volume
vol_drv = vol_attach.build_volume_driver(
self.adapter, self.host_uuid, instance, connection_info)
# Before attempting to detach a volume, ensure the instance exists
# If a live migration fails, the compute manager will call detach
# for each volume attached to the instance, against the destination
# host. If the migration failed, then the VM is probably not on
# the destination host.
if not vm.instance_exists(self.adapter, instance):
LOG.info('During volume detach, the instance was not found on '
'this host.', instance=instance)
# Check if there is live migration cleanup to do on this volume.
mig = self.live_migrations.get(instance.uuid, None)
if mig is not None and isinstance(mig, lpm.LiveMigrationDest):
mig.cleanup_volume(vol_drv)
return
# Add a task to detach the volume
slot_mgr = slot.build_slot_mgr(instance, self.store_api)
flow.add(tf_stg.DisconnectVolume(vol_drv, slot_mgr))
# Save the new slot info
flow.add(tf_slot.SaveSlotStore(instance, slot_mgr))
# Run the flow
tf_base.run(flow, instance=instance)
def snapshot(self, context, instance, image_id, update_task_state):
"""Snapshots the specified instance.
:param context: security context
:param instance: nova.objects.instance.Instance
:param image_id: Reference to a pre-created image that will hold the
snapshot.
:param update_task_state: Callback function to update the task_state
on the instance while the snapshot operation progresses. The
function takes a task_state argument and an optional
expected_task_state kwarg which defaults to
nova.compute.task_states.IMAGE_SNAPSHOT. See
nova.objects.instance.Instance.save for expected_task_state usage.
"""
if not self.disk_dvr.capabilities.get('snapshot'):
raise exception.NotSupportedWithOption(
message=_("The snapshot operation is not supported in "
"conjunction with a [powervm]/disk_driver setting "
"of %s.") % CONF.powervm.disk_driver)
self._log_operation('snapshot', instance)
# Define the flow
flow = tf_lf.Flow("snapshot")
# Notify that we're starting the process
flow.add(tf_img.UpdateTaskState(update_task_state,
task_states.IMAGE_PENDING_UPLOAD))
# Connect the instance's boot disk to the management partition, and
# scan the scsi bus and bring the device into the management partition.
flow.add(tf_stg.InstanceDiskToMgmt(self.disk_dvr, instance))
# Notify that the upload is in progress
flow.add(tf_img.UpdateTaskState(
update_task_state, task_states.IMAGE_UPLOADING,
expected_state=task_states.IMAGE_PENDING_UPLOAD))
# Stream the disk to glance
flow.add(tf_img.StreamToGlance(context, self.image_api, image_id,
instance))
# Disconnect the boot disk from the management partition and delete the
# device
flow.add(tf_stg.RemoveInstanceDiskFromMgmt(self.disk_dvr, instance))
# Run the flow
tf_base.run(flow, instance=instance)
def rescue(self, context, instance, network_info, image_meta,
rescue_password):
"""Rescue the specified instance.
:param nova.context.RequestContext context:
The context for the rescue.
:param nova.objects.instance.Instance instance:
The instance being rescued.
:param nova.network.model.NetworkInfo network_info:
Necessary network information for the resume.
:param nova.objects.ImageMeta image_meta:
The metadata of the image of the instance.
:param rescue_password: new root password to set for rescue.
"""
self._log_operation('rescue', instance)
# Define the flow
flow = tf_lf.Flow("rescue")
# Power Off the LPAR
flow.add(tf_vm.PowerOff(self.adapter, instance))
# Create and populate the rescue disk.
flow.add(tf_stg.CreateDiskForImg(
self.disk_dvr, context, instance, image_meta,
image_type=disk_dvr.DiskType.RESCUE))
# Connect up the disk to the LPAR
flow.add(tf_stg.ConnectDisk(self.disk_dvr, instance))
# Last step is to power on the system.
flow.add(tf_vm.PowerOn(
self.adapter, instance,
pwr_opts=pvm_popts.PowerOnOpts().bootmode(pvm_popts.BootMode.SMS)))
# Run the flow
tf_base.run(flow, instance=instance)
def unrescue(self, instance, network_info):
"""Unrescue the specified instance.
:param instance: nova.objects.instance.Instance
"""
self._log_operation('unrescue', instance)
# Define the flow
flow = tf_lf.Flow("unrescue")
# Power Off the LPAR
flow.add(tf_vm.PowerOff(self.adapter, instance))
# Detach the disk adapter for the rescue image
flow.add(tf_stg.DetachDisk(
self.disk_dvr, instance, disk_type=[disk_dvr.DiskType.RESCUE]))
# Delete the storage disk for the rescue image
flow.add(tf_stg.DeleteDisk(self.disk_dvr, instance))
# Last step is to power on the system.
flow.add(tf_vm.PowerOn(self.adapter, instance))
# Run the flow
tf_base.run(flow, instance=instance)
def power_off(self, instance, timeout=0, retry_interval=0):
"""Power off the specified instance.
:param instance: nova.objects.instance.Instance
:param timeout: time to wait for GuestOS to shutdown
:param retry_interval: How often to signal guest while
waiting for it to shutdown
"""
self._log_operation('power_off', instance)
force_immediate = (timeout == 0)
timeout = timeout or None
vm.power_off(self.adapter, instance, force_immediate=force_immediate,
timeout=timeout)
def power_on(self, context, instance, network_info,
block_device_info=None):
"""Power on the specified instance.
:param instance: nova.objects.instance.Instance
"""
self._log_operation('power_on', instance)
vm.power_on(self.adapter, instance)
def reboot(self, context, instance, network_info, reboot_type,
block_device_info=None, bad_volumes_callback=None):
"""Reboot the specified instance.
After this is called successfully, the instance's state
goes back to power_state.RUNNING. The virtualization
platform should ensure that the reboot action has completed
successfully even in cases in which the underlying domain/vm
is paused or halted/stopped.
:param instance: nova.objects.instance.Instance
:param network_info:
:py:meth:`~nova.network.manager.NetworkManager.get_instance_nw_info`
:param reboot_type: Either a HARD or SOFT reboot
:param block_device_info: Info pertaining to attached volumes
:param bad_volumes_callback: Function to handle any bad volumes
encountered
"""
self._log_operation(reboot_type + ' reboot', instance)
vm.reboot(self.adapter, instance, reboot_type == 'HARD')
# pypowervm exceptions are sufficient to indicate real failure.
# Otherwise, pypowervm thinks the instance is up.
return True
def get_available_resource(self, nodename):
"""Retrieve resource information.
This method is called when nova-compute launches, and
as part of a periodic task
:param nodename:
node which the caller want to get resources from
a driver that manages only one node can safely ignore this
:return: Dictionary describing resources
"""
# Do this here so it refreshes each time this method is called.
self.host_wrapper = pvm_ms.System.get(self.adapter)[0]
return self._get_available_resource()
def _get_available_resource(self):
# Get host information
data = pvm_host.build_host_resource_from_ms(self.host_wrapper)
# Add the disk information
data["local_gb"] = self.disk_dvr.capacity
data["local_gb_used"] = self.disk_dvr.capacity_used
return data
def update_provider_tree(self, provider_tree, nodename, allocations=None):
"""Update a ProviderTree with current provider and inventory data.
:param nova.compute.provider_tree.ProviderTree provider_tree:
A nova.compute.provider_tree.ProviderTree object representing all
the providers in the tree associated with the compute node, and any
sharing providers (those with the ``MISC_SHARES_VIA_AGGREGATE``
trait) associated via aggregate with any of those providers (but
not *their* tree- or aggregate-associated providers), as currently
known by placement.
:param nodename:
String name of the compute node (i.e.
ComputeNode.hypervisor_hostname) for which the caller is requesting
updated provider information.
:param allocations: Currently ignored by this driver.
"""
# Get (legacy) resource information. Same as get_available_resource,
# but we don't need to refresh self.host_wrapper as it was *just*
# refreshed by get_available_resource in the resource tracker's
# update_available_resource flow.
data = self._get_available_resource()
# TODO(efried): Fix these to reflect something like reality
# For now, duplicate the logic the resource tracker uses via
# update_compute_node when get_inventory/update_provider_tree is not
# implemented.
cpu_alloc_ratio = CONF.cpu_allocation_ratio or 16.0
cpu_reserved = CONF.reserved_host_cpus
mem_alloc_ratio = CONF.ram_allocation_ratio or 1.5
mem_reserved = CONF.reserved_host_memory_mb
disk_alloc_ratio = CONF.disk_allocation_ratio or 1.0
disk_reserved = self._get_reserved_host_disk_gb_from_config()
inventory = {
orc.VCPU: {
'total': data['vcpus'],
'max_unit': data['vcpus'],
'allocation_ratio': cpu_alloc_ratio,
'reserved': cpu_reserved,
},
orc.MEMORY_MB: {
'total': data['memory_mb'],
'max_unit': data['memory_mb'],
'allocation_ratio': mem_alloc_ratio,
'reserved': mem_reserved,
},
orc.DISK_GB: {
# TODO(efried): Proper DISK_GB sharing when SSP driver in play
'total': int(data['local_gb']),
'max_unit': int(data['local_gb']),
'allocation_ratio': disk_alloc_ratio,
'reserved': disk_reserved,
},
}
provider_tree.update_inventory(nodename, inventory)
def get_host_uptime(self):
"""Returns the result of calling "uptime" on the target host."""
# trivial implementation from libvirt/driver.py for consistency
out, err = n_utils.execute('env', 'LANG=C', 'uptime')
return out
def attach_interface(self, context, instance, image_meta, vif):
"""Attach an interface to the instance."""
self.plug_vifs(instance, [vif])
def detach_interface(self, context, instance, vif):
"""Detach an interface from the instance."""
self.unplug_vifs(instance, [vif])
def plug_vifs(self, instance, network_info):
"""Plug VIFs into networks."""
self._log_operation('plug_vifs', instance)
# Define the flow
flow = tf_lf.Flow("plug_vifs")
# Get the LPAR Wrapper
flow.add(tf_vm.Get(self.adapter, self.host_uuid, instance))
# Run the attach
slot_mgr = slot.build_slot_mgr(instance, self.store_api)
flow.add(tf_net.PlugVifs(self.virtapi, self.adapter, instance,
network_info, self.host_uuid, slot_mgr))
# Save the new slot info
flow.add(tf_slot.SaveSlotStore(instance, slot_mgr))
# Run the flow
try:
tf_base.run(flow, instance=instance)
except exception.InstanceNotFound:
raise exception.VirtualInterfacePlugException(
_("Plug vif failed because instance %s was not found.")
% instance.name)
except Exception:
LOG.exception("PowerVM error plugging vifs.", instance=instance)
raise exception.VirtualInterfacePlugException(
_("Plug vif failed because of an unexpected error."))
def unplug_vifs(self, instance, network_info):
"""Unplug VIFs from networks."""
self._log_operation('unplug_vifs', instance)
# Define the flow
flow = tf_lf.Flow("unplug_vifs")
# Get the LPAR Wrapper
flow.add(tf_vm.Get(self.adapter, self.host_uuid, instance))
# Run the detach
slot_mgr = slot.build_slot_mgr(instance, self.store_api)
flow.add(tf_net.UnplugVifs(self.adapter, instance, network_info,
self.host_uuid, slot_mgr))
# Save the new slot info
flow.add(tf_slot.SaveSlotStore(instance, slot_mgr))
# Run the flow
try:
tf_base.run(flow, instance=instance)
except exception.InstanceNotFound:
LOG.warning('VM was not found during unplug operation as it is '
'already possibly deleted.', instance=instance)
except Exception:
LOG.exception("PowerVM error trying to unplug vifs.",
instance=instance)
raise exception.InterfaceDetachFailed(instance_uuid=instance.uuid)
def get_available_nodes(self, refresh=False):
"""Returns nodenames of all nodes managed by the compute service.
This method is for multi compute-nodes support. If a driver supports
multi compute-nodes, this method returns a list of nodenames managed
by the service. Otherwise, this method should return
[hypervisor_hostname].
"""
return [CONF.host]
def legacy_nwinfo(self):
"""Indicate if the driver requires the legacy network_info format."""
return False
def get_host_ip_addr(self):
"""Retrieves the IP address of the Host."""
# This code was pulled from the libvirt driver.
ips = compute_utils.get_machine_ips()
if CONF.my_ip not in ips:
LOG.warning('my_ip address (%(my_ip)s) was not found on '
'any of the interfaces: %(ifaces)s',
{'my_ip': CONF.my_ip, 'ifaces': ", ".join(ips)})
return CONF.my_ip
def get_volume_connector(self, instance):
"""Get connector information for the instance for attaching to volumes.
Connector information is a dictionary representing the ip of the
machine that will be making the connection, the name of the iscsi
initiator and the hostname of the machine as follows::
{
'ip': ip,
'initiator': initiator,
'host': hostname
}
"""
# Put the values in the connector
connector = {}
wwpn_list = vol_attach.get_wwpns_for_volume_connector(
self.adapter, self.host_uuid, instance)
if wwpn_list is not None:
connector["wwpns"] = wwpn_list
connector["multipath"] = CONF.powervm.volume_use_multipath
connector['host'] = vol_attach.get_hostname_for_volume(instance)
initiator_dict = iscsi.get_iscsi_initiators(self.adapter)
if initiator_dict:
connector['initiator'] = list(initiator_dict.values())[0]
return connector
def migrate_disk_and_power_off(self, context, instance, dest,
flavor, network_info,
block_device_info=None,
timeout=0, retry_interval=0):
disk_info = {}
if flavor and flavor.root_gb < instance.root_gb:
raise exception.InstanceFaultRollback(
exception.ResizeError(reason=_('Cannot reduce disk size.')))
same_host = dest == self.get_host_ip_addr()
if same_host:
self._log_operation('resize', instance)
else:
self._log_operation('migration', instance)
# Can't migrate the disks if they are not on shared storage
if not self._is_booted_from_volume(block_device_info):
if not self.disk_dvr.capabilities['shared_storage']:
raise exception.InstanceFaultRollback(
exception.ResizeError(
reason=_('Cannot migrate local disks.')))
# Get disk info from disk driver.
disk_info = dict(disk_info, **self.disk_dvr.get_info())
# Define the migrate flow
flow = tf_lf.Flow("migrate_vm")
# Power off the VM
flow.add(tf_vm.PowerOff(self.adapter, instance))
if not same_host:
# If VM is moving to a new host make sure the NVRAM is at the very
# latest.
flow.add(tf_vm.StoreNvram(self.nvram_mgr, instance,
immediate=True))
if flavor.root_gb > instance.root_gb:
# Resize the root disk
flow.add(tf_stg.ExtendDisk(
self.disk_dvr, instance, dict(type='boot'), flavor.root_gb))
# Disconnect any volumes that are attached. They are reattached
# on the new VM (or existing VM if this is just a resize.)
# Extract the block devices.
bdms = self._extract_bdm(block_device_info)
if bdms:
# Create the transaction manager (FeedTask) for Storage I/O.
xag = self._get_inst_xag(instance, bdms)
stg_ftsk = pvm_par.build_active_vio_feed_task(self.adapter,
xag=xag)
# Get the slot map. This is so we build the client
# adapters in the same slots.
slot_mgr = slot.build_slot_mgr(
instance, self.store_api, adapter=self.adapter,
vol_drv_iter=self._vol_drv_iter(instance, bdms,
stg_ftsk=stg_ftsk))
# Determine if there are volumes to disconnect. If so, remove each
# volume (within the transaction manager)
self._add_volume_disconnection_tasks(context, instance, bdms, flow,
stg_ftsk, slot_mgr)
# Add the transaction manager flow to the end of the 'storage
# disconnection' tasks. This will run all the disconnections in
# parallel
flow.add(stg_ftsk)
# We rename the VM to help identify if this is a resize and so it's
# easy to see the VM is being migrated from pvmctl. We use the resize
# name so we don't destroy it on a revert when it's on the same host.
new_name = self._gen_resize_name(instance, same_host=same_host)
flow.add(tf_vm.Rename(self.adapter, instance, new_name))
try:
tf_base.run(flow, instance=instance)
except Exception as e:
raise exception.InstanceFaultRollback(e)
return disk_info
@staticmethod
def _gen_resize_name(instance, same_host=False):
"""Generate a temporary name for the source VM being resized/migrated.
:param instance: nova.objects.instance.Instance being migrated/resized.
:param same_host: Boolean indicating whether this resize is being
performed for the sake of a resize (True) or a
migration (False).
:return: A new name which can be assigned to the source VM.
"""
prefix = 'resize_' if same_host else 'migrate_'
return pvm_util.sanitize_partition_name_for_api(prefix + instance.name)
def finish_migration(self, context, migration, instance, disk_info,
network_info, image_meta, resize_instance,
block_device_info=None, power_on=True):
"""Completes a resize or cold migration.
:param context: the context for the migration/resize
:param migration: the migrate/resize information
:param instance: nova.objects.instance.Instance being migrated/resized
:param disk_info: the newly transferred disk information
:param network_info:
:py:meth:`~nova.network.manager.NetworkManager.get_instance_nw_info`
:param nova.objects.ImageMeta image_meta:
The metadata of the image of the instance.
:param resize_instance: True if the instance disks are being resized,
False otherwise
:param block_device_info: instance volume block device info
:param power_on: True if the instance should be powered on, False
otherwise
"""
# See if this was to the same host
same_host = migration.source_compute == migration.dest_compute
if same_host:
self._log_operation('finish resize', instance)
else:
self._log_operation('finish migration', instance)
# Ensure the disk drivers are compatible.
booted_from_vol = self._is_booted_from_volume(block_device_info)
if (not same_host and not booted_from_vol):
# Can't migrate the disks if they are not on shared storage
if not self.disk_dvr.capabilities['shared_storage']:
raise exception.InstanceFaultRollback(
exception.ResizeError(
reason=_('Cannot migrate local disks.')))
# Call the disk driver to evaluate the disk info
reason = self.disk_dvr.validate(disk_info)
if reason:
raise exception.InstanceFaultRollback(
exception.ResizeError(reason=reason))
# Extract the block devices.
bdms = self._extract_bdm(block_device_info)
# Define the flow
flow = tf_lf.Flow("finish_migration")
# If attaching disks or volumes
if bdms or not same_host:
# Create the transaction manager (FeedTask) for Storage I/O.
xag = self._get_inst_xag(instance, bdms)
stg_ftsk = pvm_par.build_active_vio_feed_task(self.adapter,
xag=xag)
# We need the slot manager
# a) If migrating to a different host: to restore the proper slots;
# b) If adding/removing block devices, to register the slots.
slot_mgr = slot.build_slot_mgr(
instance, self.store_api, adapter=self.adapter,
vol_drv_iter=self._vol_drv_iter(instance, bdms,
stg_ftsk=stg_ftsk))
else:
stg_ftsk = None
if same_host:
# This is just a resize.
new_name = self._gen_resize_name(instance, same_host=True)
flow.add(tf_vm.Resize(self.adapter, self.host_wrapper, instance,
name=new_name))
else:
# This is a migration over to another host. We have a lot of work.
# Create the LPAR
flow.add(tf_vm.Create(self.adapter, self.host_wrapper, instance,
stg_ftsk=stg_ftsk, nvram_mgr=self.nvram_mgr,
slot_mgr=slot_mgr))
# Create a flow for the network IO
flow.add(tf_net.PlugVifs(self.virtapi, self.adapter, instance,
network_info, self.host_uuid, slot_mgr))
flow.add(tf_net.PlugMgmtVif(
self.adapter, instance, self.host_uuid, slot_mgr))
# Need to attach the boot disk, if present.
if not self._is_booted_from_volume(block_device_info):
flow.add(tf_stg.FindDisk(self.disk_dvr, context, instance,
disk_dvr.DiskType.BOOT))
# Connects up the disk to the LPAR
# TODO(manas) Connect the disk flow into the slot lookup map
flow.add(tf_stg.ConnectDisk(
self.disk_dvr, instance, stg_ftsk=stg_ftsk))
if bdms:
# Determine if there are volumes to connect. If so, add a
# connection for each type.
self._add_volume_connection_tasks(
context, instance, bdms, flow, stg_ftsk, slot_mgr)
if stg_ftsk:
# Add the transaction manager flow to the end of the 'storage
# connection' tasks to run all the connections in parallel
flow.add(stg_ftsk)
if power_on:
flow.add(tf_vm.PowerOn(self.adapter, instance))
# Run the flow
try:
tf_base.run(flow, instance=instance)
except Exception as e:
raise exception.InstanceFaultRollback(e)
def confirm_migration(self, context, migration, instance, network_info):
"""Confirms a resize, destroying the source VM.
:param migration: the migrate/resize information
:param instance: nova.objects.instance.Instance
:param network_info:
:py:meth:`~nova.network.manager.NetworkManager.get_instance_nw_info`
"""
# See if this was to the same host
same_host = migration.source_compute == migration.dest_compute
if same_host:
# This was a local resize, don't delete our only VM!
self._log_operation('confirm resize', instance)
vm.rename(self.adapter, instance, instance.name)
return
# Confirming the migrate means we need to delete source VM.
self._log_operation('confirm migration', instance)
# Destroy the old VM.
destroy_disks = not self.disk_dvr.capabilities['shared_storage']
self._destroy(context, instance, block_device_info=None,
destroy_disks=destroy_disks, shutdown=False)
def finish_revert_migration(self, context, instance, network_info,
block_device_info=None, power_on=True):
"""Finish reverting a resize on the source host.
:param context: the context for the finish_revert_migration
:param instance: nova.objects.instance.Instance being migrated/resized
:param network_info:
:py:meth:`~nova.network.manager.NetworkManager.get_instance_nw_info`
:param block_device_info: instance volume block device info
:param power_on: True if the instance should be powered on, False
otherwise
"""
self._log_operation('revert resize/migration', instance)
# This method is always run on the source host, so we just need to
# revert the VM back to it's old sizings, if it was even changed
# at all. If it was a migration, then it wasn't changed but it
# shouldn't hurt to "update" it with the prescribed flavor. This
# makes it easy to handle both resize and migrate.
#
# The flavor should be the 'old' flavor now.
vm.power_off(self.adapter, instance)
vm.update(self.adapter, self.host_wrapper, instance)
if power_on:
vm.power_on(self.adapter, instance)
def ensure_filtering_rules_for_instance(self, instance, network_info):
"""Setting up filtering rules and waiting for its completion.
To migrate an instance, filtering rules to hypervisors
and firewalls are inevitable on destination host.
( Waiting only for filtering rules to hypervisor,
since filtering rules to firewall rules can be set faster).
Concretely, the below method must be called.
- setup_basic_filtering (for nova-basic, etc.)
- prepare_instance_filter(for nova-instance-instance-xxx, etc.)
to_xml may have to be called since it defines PROJNET, PROJMASK.
but libvirt migrates those value through migrateToURI(),
so , no need to be called.
Don't use thread for this method since migration should
not be started when setting-up filtering rules operations
are not completed.
:param instance: nova.objects.instance.Instance object
"""
# No op for PowerVM
pass
def check_can_live_migrate_destination(self, context, instance,
src_compute_info, dst_compute_info,
block_migration=False,
disk_over_commit=False):
"""Check if it is possible to execute live migration.
This runs checks on the destination host, and then calls
back to the source host to check the results.
:param context: security context
:param instance: nova.db.sqlalchemy.models.Instance
:param src_compute_info: Info about the sending machine
:param dst_compute_info: Info about the receiving machine
:param block_migration: if true, prepare for block migration
:param disk_over_commit: if true, allow disk over commit
:returns: a dict containing migration info (hypervisor-dependent)
"""
LOG.info("Checking live migration capability on destination host.",
instance=instance)
mig = lpm.LiveMigrationDest(self, instance)
self.live_migrations[instance.uuid] = mig
return mig.check_destination(context, src_compute_info,
dst_compute_info)
def cleanup_live_migration_destination_check(self, context,
dest_check_data):
"""Do required cleanup on dest host after check_can_live_migrate calls
:param context: security context
:param dest_check_data: result of check_can_live_migrate_destination
"""
LOG.info("Cleaning up from checking live migration capability "
"on destination.")
def check_can_live_migrate_source(self, context, instance,
dest_check_data, block_device_info=None):
"""Check if it is possible to execute live migration.
This checks if the live migration can succeed, based on the
results from check_can_live_migrate_destination.
:param context: security context
:param instance: nova.db.sqlalchemy.models.Instance
:param dest_check_data: result of check_can_live_migrate_destination
:param block_device_info: result of _get_instance_block_device_info
:returns: a dict containing migration info (hypervisor-dependent)
"""
LOG.info("Checking live migration capability on source host.",
instance=instance)
mig = lpm.LiveMigrationSrc(self, instance, dest_check_data)
self.live_migrations[instance.uuid] = mig
# Get a volume driver for each volume
vol_drvs = self._build_vol_drivers(context, instance,
block_device_info)
return mig.check_source(context, block_device_info, vol_drvs)
def pre_live_migration(self, context, instance, block_device_info,
network_info, disk_info, migrate_data=None):
"""Prepare an instance for live migration
:param context: security context
:param instance: nova.objects.instance.Instance object
:param block_device_info: instance block device information
:param network_info: instance network information
:param disk_info: instance disk information
:param migrate_data: a LiveMigrateData object
"""
LOG.info("Pre live migration processing.", instance=instance)
mig = self.live_migrations[instance.uuid]
# Get a volume driver for each volume
vol_drvs = self._build_vol_drivers(context, instance,
block_device_info)
# Run pre-live migration
return mig.pre_live_migration(context, block_device_info, network_info,
disk_info, migrate_data, vol_drvs)
def live_migration(self, context, instance, dest,
post_method, recover_method, block_migration=False,
migrate_data=None):
"""Live migration of an instance to another host.
:param context: security context
:param instance:
nova.db.sqlalchemy.models.Instance object
instance object that is migrated.
:param dest: destination host
:param post_method:
post operation method.
expected nova.compute.manager._post_live_migration.
:param recover_method:
recovery method when any exception occurs.
expected nova.compute.manager._rollback_live_migration.
:param block_migration: if true, migrate VM disk.
:param migrate_data: a LiveMigrateData object
"""
self._log_operation('live_migration', instance)
try:
mig = self.live_migrations[instance.uuid]
try:
mig.live_migration(context, migrate_data)
except pvm_exc.JobRequestTimedOut as timeout_ex:
# If the migration operation exceeds configured timeout
LOG.error("Live migration timed out. Aborting migration",
instance=instance)
mig.migration_abort()
self._migration_exception_util(context, instance, dest,
recover_method,
migrate_data,
mig, ex=timeout_ex)
except Exception as e:
LOG.exception("PowerVM error during live migration.",
instance=instance)
self._migration_exception_util(context, instance, dest,
recover_method,
migrate_data,
mig, ex=e)
LOG.debug("Calling post live migration method.", instance=instance)
# Post method to update host in OpenStack and finish live-migration
post_method(context, instance, dest, block_migration, migrate_data)
finally:
# Remove the migration record on the source side.
del self.live_migrations[instance.uuid]
def _migration_exception_util(self, context, instance, dest,
recover_method, migrate_data, mig, ex):
"""Migration exception utility.
:param context: security context
:param instance:
nova.db.sqlalchemy.models.Instance object
instance object that is migrated.
:param dest: destination host
:param recover_method:
recovery method when any exception occurs.
expected nova.compute.manager._rollback_live_migration.
:param migrate_data: a LiveMigrateData object
:param mig: live_migration object
:param ex: exception reason
"""
LOG.warning("Rolling back live migration.", instance=instance)
try:
mig.rollback_live_migration(context)
recover_method(context, instance, dest, migrate_data=migrate_data)
except Exception:
LOG.exception("PowerVM error rolling back live migration.",
instance=instance)
raise lpm.LiveMigrationFailed(name=instance.name,
reason=six.text_type(ex))
def rollback_live_migration_at_destination(self, context, instance,
network_info,
block_device_info,
destroy_disks=True,
migrate_data=None):
"""Clean up destination node after a failed live migration.
:param context: security context
:param instance: instance object that was being migrated
:param network_info: instance network information
:param block_device_info: instance block device information
:param destroy_disks:
if true, destroy disks at destination during cleanup
:param migrate_data: a LiveMigrateData object
"""
# Run the rollback
mig = self.live_migrations[instance.uuid]
mig.rollback_live_migration_at_destination(
context, instance, network_info, block_device_info,
destroy_disks=destroy_disks, migrate_data=migrate_data)
# Remove the active migration
del self.live_migrations[instance.uuid]
def check_instance_shared_storage_local(self, context, instance):
"""Check if instance files located on shared storage.
This runs check on the destination host, and then calls
back to the source host to check the results.
:param context: security context
:param instance: nova.objects.instance.Instance object
"""
# Defer to the disk driver method.
return self.disk_dvr.check_instance_shared_storage_local(
context, instance)
def check_instance_shared_storage_remote(self, context, data):
"""Check if instance files located on shared storage.
:param context: security context
:param data: result of check_instance_shared_storage_local
"""
# Defer to the disk driver method.
return self.disk_dvr.check_instance_shared_storage_remote(
context, data)
def check_instance_shared_storage_cleanup(self, context, data):
"""Do cleanup on host after check_instance_shared_storage calls
:param context: security context
:param data: result of check_instance_shared_storage_local
"""
# Defer to the disk driver method.
return self.disk_dvr.check_instance_shared_storage_cleanup(
context, data)
def post_live_migration(self, context, instance, block_device_info,
migrate_data=None):
"""Post operation of live migration at source host.
:param context: security context
:instance: instance object that was migrated
:block_device_info: instance block device information
:param migrate_data: a LiveMigrateData object
"""
# Build the volume drivers
vol_drvs = self._build_vol_drivers(context, instance,
block_device_info)
mig = self.live_migrations[instance.uuid]
mig.post_live_migration(vol_drvs, migrate_data)
def post_live_migration_at_source(self, context, instance, network_info):
"""Unplug VIFs from networks at source.
:param context: security context
:param instance: instance object reference
:param network_info: instance network information
"""
LOG.info("Post live migration processing on source host.",
instance=instance)
mig = self.live_migrations[instance.uuid]
mig.post_live_migration_at_source(network_info)
def post_live_migration_at_destination(self, context, instance,
network_info,
block_migration=False,
block_device_info=None):
"""Post operation of live migration at destination host.
:param context: security context
:param instance: instance object that is migrated
:param network_info: instance network information
:param block_migration: if true, post operation of block_migration.
:param block_device_info: instance block device information.
"""
LOG.info("Post live migration processing on destination host.",
instance=instance)
mig = self.live_migrations[instance.uuid]
mig.instance = instance
bdms = self._extract_bdm(block_device_info)
# Get a volume driver iterator for volume and BDM mapping
vol_drv_iter = self._vol_drv_iter(instance, bdms)
# Run post live migration
mig.post_live_migration_at_destination(network_info, vol_drv_iter)
del self.live_migrations[instance.uuid]
def _vol_drv_iter(self, instance, bdms, stg_ftsk=None):
"""Yields a bdm and volume driver."""
# Get a volume driver for each volume
for bdm in bdms or []:
vol_drv = vol_attach.build_volume_driver(
self.adapter, self.host_uuid, instance,
bdm.get('connection_info'), stg_ftsk=stg_ftsk)
yield bdm, vol_drv
def _build_vol_drivers(self, context, instance, block_device_info):
"""Builds the volume connector drivers for a block device info."""
# Get a volume driver for each volume
bdms = self._extract_bdm(block_device_info)
return [vol_drv for bdm, vol_drv in self._vol_drv_iter(instance, bdms)]
def unfilter_instance(self, instance, network_info):
"""Stop filtering instance."""
# No op for PowerVM
pass
@staticmethod
def _extract_bdm(block_device_info):
"""Returns the block device mapping out of the block device info.
The block device mapping is a list of instances of block device
classes from nova.virt.block_device. Each block device
represents one volume connection.
An example string representation of a DriverVolumeBlockDevice
from the early Liberty time frame is:
{'guest_format': None,
'boot_index': 0,
'mount_device': u'/dev/sda',
'connection_info': {u'driver_volume_type': u'fibre_channel',
u'serial': u'e11765ea-dd14-4aa9-a953-4fd6b4999635',
u'data': {u'initiator_target_map':
{u'21000024ff747e59':
[u'500507680220E522',
u'500507680210E522'],
u'21000024ff747e58':
[u'500507680220E522',
u'500507680210E522']},
u'vendor': u'IBM',
u'target_discovered':False,
u'target_UID': u'600507680282...',
u'qos_specs': None,
u'volume_id': u'e11765ea-...',
u'target_lun': u'2',
u'access_mode': u'rw',
u'target_wwn': u'500507680220E522'}
},
'disk_bus': None,
'device_type': u'disk',
'delete_on_termination': True}
"""
if block_device_info is None:
return []
return block_device_info.get('block_device_mapping', [])
def get_vnc_console(self, context, instance):
"""Get connection info for a vnc console.
:param context: security context
:param instance: nova.objects.instance.Instance
:return: An instance of console.type.ConsoleVNC
"""
self._log_operation('get_vnc_console', instance)
lpar_uuid = vm.get_pvm_uuid(instance)
# Build the connection to the VNC.
host = CONF.vnc.server_proxyclient_address
use_x509_auth = CONF.powervm.vnc_use_x509_auth
ca_certs = CONF.powervm.vnc_ca_certs
server_key = CONF.powervm.vnc_server_key
server_cert = CONF.powervm.vnc_server_cert
try:
# Open up a remote vterm with the host and certificates configured
# This will only use TLS if the use_x509_auth is set to True
port = pvm_vterm.open_remotable_vnc_vterm(
self.adapter, lpar_uuid, host, vnc_path=lpar_uuid,
use_x509_auth=use_x509_auth, ca_certs=ca_certs,
server_cert=server_cert, server_key=server_key)
except pvm_exc.HttpNotFound:
raise exception.InstanceNotFound(instance_id=instance.uuid)
except pvm_exc.Error as exc:
# Otherwise wrapper the error in an exception that can be handled
LOG.exception("Unable to open console.", instance=instance)
msg = (_("VNC based terminal for instance %(instance_name)s "
"failed to open: %(exc_msg)s")
% {'instance_name': instance.name,
'exc_msg': exc.args[0]})
# Need to raise ConsoleTypeUnavailable with overwritten message
# because otherwise the exception will not be caught. It is
# disallowed to send a non-nova exception over the wire.
raise exception.ConsoleTypeUnavailable(msg)
# Note that the VNC viewer will wrap the internal_access_path with
# the HTTP content.
return console_type.ConsoleVNC(host=host, port=port,
internal_access_path=lpar_uuid)
def _get_inst_xag(self, instance, bdms, recreate=False):
"""Returns the extended attributes required for a given instance.
This is used in coordination with the FeedTask. It identifies ahead
of time what each request requires for its general operations.
:param instance: Nova instance for which the volume adapter is needed.
:param bdms: The BDMs for the operation.
:param recreate: (Optional, Default: False) If set to true, will return
all of the storage XAGs so that a full scrub can be
done (since specific slots are needed).
:return: List of extended attributes required for the operation.
"""
if recreate:
return {pvm_const.XAG.VIO_FMAP, pvm_const.XAG.VIO_SMAP,
pvm_const.XAG.VIO_STOR}
# All operations for deploy/destroy require scsi by default. This is
# either vopt, local/SSP disks, etc...
xags = {pvm_const.XAG.VIO_SMAP}
# BDMs could be none, if there are no cinder volumes.
bdms = bdms if bdms else []
# If we have any volumes, add the volumes required mapping XAGs.
for bdm in bdms:
driver_type = bdm.get('connection_info').get('driver_volume_type')
vol_cls = vol_attach.get_volume_class(driver_type)
xags.update(set(vol_cls.min_xags()))
LOG.debug('Instance XAGs: %(xags)s.', {'xags': ','.join(xags)},
instance=instance)
return list(xags)
def _get_boot_connectivity_type(self, bdms, block_device_info):
"""Get connectivity information for the instance.
:param bdms: The BDMs for the operation. If boot volume of
the instance is ssp lu or local disk, the bdms is None.
:param block_device_info: Instance volume block device info.
:return: The boot connectivity type.
If boot volume is an npiv volume, returns 'fibre_channel'.
Otherwise, returns 'vscsi'.
"""
if self._is_booted_from_volume(block_device_info) and bdms is not None:
for bdm in bdms:
if bdm.get('boot_index') == 0:
return self._get_connectivity_type(bdm)
# Default connectivity type is 'vscsi'
return 'vscsi'
@staticmethod
def _get_connectivity_type(bdm):
conn_info = bdm.get('connection_info')
if 'connection-type' in conn_info['data']:
connectivity_type = conn_info['data']['connection-type']
return ('vscsi' if connectivity_type == 'pv_vscsi'
else connectivity_type)
# Seemingly bogus path (driver_volume_type shouldn't be in 'data'),
# preserved for potential compatibility.
if 'driver_volume_type' in conn_info['data']:
return conn_info['data']['driver_volume_type']
# Actual location for driver_volume_type. Default to vscsi.
return conn_info.get('driver_volume_type', 'vscsi')
def deallocate_networks_on_reschedule(self, instance):
"""Does the driver want networks deallocated on reschedule?
:param instance: the instance object.
:returns: Boolean value. If True deallocate networks on reschedule.
"""
return True
def _cleanup_orphan_adapters(self, vswitch_name):
"""Finds and removes trunk VEAs that have no corresponding CNA."""
orphans = pvm_cna.find_orphaned_trunks(self.adapter, vswitch_name)
for orphan in orphans:
LOG.info("Deleting orphan CNA: %s", orphan.dev_name)
orphan.delete()