The nova.virt.xenapi.vmops Module

Management class for VM-related functions (spawn, reboot, etc).

class VMOps(session, virtapi)

Bases: object

Management class for VM-related tasks

agent_enabled

change_instance_metadata(instance, diff)

Apply changes to instance metadata to xenstore.

check_can_live_migrate_destination(ctxt, instance_ref, block_migration=False, disk_over_commit=False)

Check if it is possible to execute live migration.

Parameters:

• context – security context
• instance_ref – nova.db.sqlalchemy.models.Instance object
• block_migration – if true, prepare for block migration
• disk_over_commit – if true, allow disk over commit

check_can_live_migrate_source(ctxt, instance_ref, dest_check_data)

Check if it’s possible to execute live migration on the source side.

Parameters:

• context – security context
• instance_ref – nova.db.sqlalchemy.models.Instance object
• dest_check_data – data returned by the check on the destination, includes block_migration flag

check_resize_func_name()

Check the function name used to resize an instance based on product_brand and product_version.

confirm_migration(migration, instance, network_info)

destroy(instance, network_info, block_device_info=None, destroy_disks=True)

Destroy VM instance.

This is the method exposed by xenapi_conn.destroy(). The rest of the destroy_* methods are internal.

finish_migration(context, migration, instance, disk_info, network_info, image_meta, resize_instance, block_device_info=None)

finish_revert_migration(instance, block_device_info=None)

get_all_bw_counters()

Return running bandwidth counter for each interface on each running VM

get_console_output(instance)

Return snapshot of console.

get_diagnostics(instance)

Return data about VM diagnostics.

get_info(instance, vm_ref=None)

Return data about VM instance.

get_per_instance_usage()

Get usage info about each active instance.

get_vnc_console(instance)

Return connection info for a vnc console.

inject_file(instance, path, contents)

Write a file to the VM instance.

inject_hostname(instance, vm_ref, hostname)

Inject the hostname of the instance into the xenstore.

inject_instance_metadata(instance, vm_ref)

Inject instance metadata into xenstore.

inject_network_info(instance, network_info, vm_ref=None)

Generate the network info and make calls to place it into the xenstore and the xenstore param list. vm_ref can be passed in because it will sometimes be different than what vm_utils.lookup(session, instance[‘name’]) will find (ex: rescue)

list_instance_uuids()

Get the list of nova instance uuids for VMs found on the hypervisor.

list_instances()

List VM instances.

live_migrate(context, instance, destination_hostname, post_method, recover_method, block_migration, migrate_data=None)

migrate_disk_and_power_off(context, instance, dest, instance_type)

Copies a VHD from one host machine to another, possibly resizing filesystem before hand.

Parameters:

• instance – the instance that owns the VHD in question.
• dest – the destination host machine.
• instance_type – instance_type to resize to

pause(instance)

Pause VM instance.

plug_vifs(instance, network_info)

Set up VIF networking on the host.

poll_rebooting_instances(timeout, instances)

Look for expirable rebooting instances.

• issue a “hard” reboot to any instance that has been stuck in a reboot state for >= the given timeout

power_off(instance)

Power off the specified instance.

power_on(instance)

Power on the specified instance.

reboot(instance, reboot_type, bad_volumes_callback=None)

Reboot VM instance.

refresh_instance_security_rules(instance)

recreates security group rules for specified instance.

refresh_provider_fw_rules()

refresh_security_group_members(security_group_id)

recreates security group rules for every instance.

refresh_security_group_rules(security_group_id)

recreates security group rules for every instance.

rescue(context, instance, network_info, image_meta, rescue_password)

Rescue the specified instance.

• shutdown the instance VM.
• set ‘bootlock’ to prevent the instance from starting in rescue.
• spawn a rescue VM (the vm name-label will be instance-N-rescue).

reset_network(instance)

Calls resetnetwork method in agent.

restore(instance)

Restore the specified instance.

resume(instance)

Resume the specified instance.

set_admin_password(instance, new_pass)

Set the root/admin password on the VM instance.

snapshot(context, instance, image_id, update_task_state)

Create snapshot from a running VM instance.

Parameters:

• context – request context
• instance – instance to be snapshotted
• image_id – id of image to upload to

Steps involved in a XenServer snapshot:

1. XAPI-Snapshot: Snapshotting the instance using XenAPI. This creates: Snapshot (Template) VM, Snapshot VBD, Snapshot VDI, Snapshot VHD
2. Wait-for-coalesce: The Snapshot VDI and Instance VDI both point to a ‘base-copy’ VDI. The base_copy is immutable and may be chained with other base_copies. If chained, the base_copies coalesce together, so, we must wait for this coalescing to occur to get a stable representation of the data on disk.
3. Push-to-data-store: Once coalesced, we call a plugin on the XenServer that will bundle the VHDs together and then push the bundle. Depending on the configured value of ‘xenapi_image_upload_handler’, image data may be pushed to Glance or the specified data store.

soft_delete(instance)

Soft delete the specified instance.

spawn(context, instance, image_meta, injected_files, admin_password, network_info=None, block_device_info=None, name_label=None, rescue=False)

suspend(instance)

Suspend the specified instance.

unfilter_instance(instance_ref, network_info)

Removes filters for each VIF of the specified instance.

unpause(instance)

Unpause VM instance.

unplug_vifs(instance, network_info)

unrescue(instance)

Unrescue the specified instance.

• unplug the instance VM’s disk from the rescue VM.
• teardown the rescue VM.
• release the bootlock to allow the instance VM to start.

cmp_version(a, b)

Compare two version strings (eg 0.0.1.10 > 0.0.1.9).

make_step_decorator(context, instance, instance_update)

Factory to create a decorator that records instance progress as a series of discrete steps.

Each time the decorator is invoked we bump the total-step-count, so after:

@step
def step1():
    ...

@step
def step2():
    ...

we have a total-step-count of 2.

Each time the step-function (not the step-decorator!) is invoked, we bump the current-step-count by 1, so after:

step1()

the current-step-count would be 1 giving a progress of 1 / 2 * 100 or 50%.