glance/doc/source/admin/useful-image-properties.rst
Danylo Vodopianov 66566c3286 Add Packed Virtqueue extra spec and image properties
Virtqueue feature was added to the metadata properties for flavor and
images

- Add the hw:virtio_packed_ring flavor extra spec and corresponding
  hw_virtio_packed_ring image property.

Nova spec: https://review.opendev.org/c/openstack/nova-specs/+/868377

Depends-On: https://review.opendev.org/c/openstack/nova/+/876075

Change-Id: Ia8caf39467a91431ca85b2a56ec93a60c41a8e7d
2023-12-05 10:59:18 +02:00

26 KiB
Raw Blame History

Useful image properties

You can set image properties that can be consumed by other services to affect the behavior of those other services. For example:

  • Image properties can be used to override specific behaviors defined for Nova flavors
  • Image properties can be used to affect the behavior of the Nova scheduler
  • Image properties can be used to affect the behavior of particular Nova hypervisors
  • Image properties can be used to provide additional information to Ironic (even when Nova is not used)

Using image properties

Some important points to keep in mind:

  • In order to allow custom image properties, Glance must be configured with the glance-api.conf setting allow_additional_image_properties set to True. (This is the default setting.)

  • The glance-api.conf setting image_property_quota should be sufficiently high to allow any additional desired properties. (The default is 128.)

  • You can use Glance property protections to control access to specific image properties, should that be desirable. See the property-protections section of this Guide for more information.

  • Glance reserves properties namespaced with the os_glance prefix for its own use and will refuse attempts by API users to set or change them.

  • You can use a plugin to the interoperable image import process to set specific properties on non-admin images imported into Glance. See iir_plugins for more information. See the original spec, Inject metadata properties automatically to non-admin images for a discussion of the use case addressed by this plugin.

  • The Nova ImagePropertiesFilter, enabled by default in the Compute Service, consumes image properties to determine proper scheduling of builds to compute hosts. See the Compute schedulers section of the Nova Configuration Guide for more information.

  • Nova has a setting, non_inheritable_image_properties, that allows you to specify which image properties from the image a virtual machine was booted from will not be propagated to a snapshot image of that virtual machine. See the Configuration Options section of the Nova Configuration Guide for more information.

  • Some properties recognized by Nova may have no effect unless a corresponding property is enabled in the server flavor. For example, the hw_rng_model image property has no effect unless the Nova flavor has been configured to have hw_rng:allowed set to True in the flavor's extra_specs.

  • In a mixed hypervisor environment, the Compute Service uses the hypervisor_type image property to match images to the correct hypervisor type.

    Depending upon what hypervisors are in use in your Nova installation, there may be other image properties that these hypervisors can consume to affect their behavior. Read through the configuration information for your hypervisors in the Hypervisors section of the Nova Configuration Guide for more information.

    In particular, the VMware hypervisor driver requires that particular image properties be set for optimal functioning. See the VMware vSphere section of the Nova Configuration Guide for more information.

Image property keys and values

Here is a list of useful image properties and the values they expect.

architecture
Type

str

The CPU architecture that must be supported by the hypervisor. For example, x86_64, arm, or ppc64. Run uname -m to get the architecture of a machine. We strongly recommend using the architecture data vocabulary defined by the libosinfo project for this purpose.

One of:

hypervisor_type
Type

str

The hypervisor type. Note that qemu is used for both QEMU and KVM hypervisor types.

One of:

  • hyperv
  • ironic
  • lxc
  • qemu
  • uml
  • vmware
  • xen.
instance_uuid
Type

str

For snapshot images, this is the UUID of the server used to create this image. The value must be a valid server UUID.

img_config_drive
Type

str

Specifies whether the image needs a config drive.

One of:

  • mandatory
  • optional (default if property is not used)
img_type
Type

str

Specifies the partitioning type of the image. The default value is partition if the kernel_id/ramdisk_id properties are present, otherwise whole-disk.

One of:

  • whole-disk - an image with a partition table embedded.
  • partition - an image with only the root partition without a partition table.

Note

This property is currently only recognized by Ironic.

kernel_id
Type

str

The ID of an image stored in the Image service that should be used as the kernel when booting an AMI-style image. The value must be a valid image ID

os_admin_user
Type

str

The name of the user with admin privileges. The value must be a valid username (defaults to root for Linux guests and Administrator for Windows guests).

os_distro
Type

str

The common name of the operating system distribution in lowercase (uses the same data vocabulary as the libosinfo project). Specify only a recognized value for this field. Deprecated values are listed to assist you in searching for the recognized value.

One of:

  • arch - Arch Linux. Do not use archlinux or org.archlinux.
  • centos - Community Enterprise Operating System. Do not use org.centos or CentOS.
  • debian - Debian. Do not use Debian` ororg.debian``.
  • fedora - Fedora. Do not use Fedora, org.fedora, or org.fedoraproject.
  • freebsd - FreeBSD. Do not use org.freebsd, freeBSD, or FreeBSD.
  • gentoo - Gentoo Linux. Do not use Gentoo or org.gentoo.
  • mandrake - Mandrakelinux (MandrakeSoft) distribution. Do not use mandrakelinux or MandrakeLinux.
  • mandriva - Mandriva Linux. Do not use mandrivalinux.
  • mes - Mandriva Enterprise Server. Do not use mandrivaent or mandrivaES.
  • msdos - Microsoft Disc Operating System. Do not use ms-dos.
  • netbsd - NetBSD. Do not use NetBSD or org.netbsd.
  • netware - Novell NetWare. Do not use novell or NetWare.
  • openbsd - OpenBSD. Do not use OpenBSD or org.openbsd.
  • opensolaris - OpenSolaris. Do not use OpenSolaris or org.opensolaris.
  • opensuse - openSUSE. Do not use suse, SuSE, or org.opensuse.
  • rocky - Rocky Linux. Do not use Rocky or rockylinux.
  • rhel - Red Hat Enterprise Linux. Do not use redhat, RedHat, or com.redhat.
  • sled - SUSE Linux Enterprise Desktop. Do not use com.suse.
  • ubuntu - Ubuntu. Do not use Ubuntu, com.ubuntu, org.ubuntu, or canonical.
  • windows - Microsoft Windows. Do not use com.microsoft.server or windoze.
os_version
Type

str

The operating system version as specified by the distributor.

The value must be a valid version number (for example, 11.10).

os_secure_boot
Type

str

Secure Boot is a security standard. When the instance starts, Secure Boot first examines software such as firmware and OS by their signature and only allows them to run if the signatures are valid.

For Hyper-V: Images must be prepared as Generation 2 VMs. Instance must also contain hw_machine_type=hyperv-gen2 image property. Linux guests will also require bootloader's digital signature provided as os_secure_boot_signature and hypervisor_version_requires'>=10.0' image properties.

One of:

  • required - Enable the Secure Boot feature.
  • disabled or optional - (default if property not used) Disable the Secure Boot feature.
os_shutdown_timeout
Type

int

By default, guests will be given 60 seconds to perform a graceful shutdown. After that, the VM is powered off. This property allows overriding the amount of time (unit: seconds) to allow a guest OS to cleanly shut down before power off. A value of 0 (zero) means the guest will be powered off immediately with no opportunity for guest OS clean-up.

ramdisk_id

The ID of image stored in the Image service that should be used as the ramdisk when booting an AMI-style image.

The value must be a valid image ID.

rootfs_uuid

For whole-disk images (see img_type above), the UUID of the root partition.

This property is used by Ironic when configuring software RAID.

trait:<trait_name>
Type

str

Added in the Rocky release. Functionality is similar to traits specified in flavor extra specs.

Traits allow specifying a server to build on a compute node with the set of traits specified in the image. The traits are associated with the resource provider that represents the compute node in the Placement API.

The syntax of specifying traits is trait:<trait_name>=value, for example:

  • trait:HW_CPU_X86_AVX2=required
  • trait:STORAGE_DISK_SSD=required

The nova scheduler will pass required traits specified on the image to the Placement API to include only resource providers that can satisfy the required traits. Traits for the resource providers can be managed using the osc-placement plugin.

Image traits are used by the nova scheduler even in cases of volume backed instances, if the volume source is an image with traits.

The only valid value is required. Any other value is invalid.

One of:

  • required - <trait_name> is required on the resource provider that represents the compute node on which the image is launched.
vm_mode
Type

str

The virtual machine mode. This represents the host/guest ABI (application binary interface) used for the virtual machine.

One of:

  • hvm - Fully virtualized. This is the mode used by QEMU and KVM.
  • xen - Xen 3.0 paravirtualized.
  • uml - User Mode Linux paravirtualized.
  • exe - Executables in containers. This is the mode used by LXC.
hw_cpu_sockets
Type

int

The preferred number of sockets to expose to the guest.

Only supported by the libvirt driver.

hw_cpu_cores
Type

int

The preferred number of cores to expose to the guest.

Only supported by the libvirt driver.

hw_cpu_threads
Type

int

The preferred number of threads to expose to the guest.

Only supported by the libvirt driver.

hw_cpu_policy
Type

str

Used to pin the virtual CPUs (vCPUs) of instances to the hosts physical CPU cores (pCPUs). Host aggregates should be used to separate these pinned instances from unpinned instances as the latter will not respect the resourcing requirements of the former.

Only supported by the libvirt driver.

One of:

  • shared - (default if property not specified) The guest vCPUs will be allowed to freely float across host pCPUs, albeit potentially constrained by NUMA policy.
  • dedicated - The guest vCPUs will be strictly pinned to a set of host pCPUs. In the absence of an explicit vCPU topology request, the drivers typically expose all vCPUs as sockets with one core and one thread. When strict CPU pinning is in effect the guest CPU topology will be setup to match the topology of the CPUs to which it is pinned. This option implies an overcommit ratio of 1.0. For example, if a two vCPU guest is pinned to a single host core with two threads, then the guest will get a topology of one socket, one core, two threads.
hw_cpu_thread_policy
Type

str

Further refine hw_cpu_policy=dedicated by stating how hardware CPU threads in a simultaneous multithreading-based (SMT) architecture be used. SMT-based architectures include Intel processors with Hyper-Threading technology. In these architectures, processor cores share a number of components with one or more other cores. Cores in such architectures are commonly referred to as hardware threads, while the cores that a given core share components with are known as thread siblings.

Only supported by the libvirt driver.

One of:

  • prefer - (default if property not specified) The host may or may not have an SMT architecture. Where an SMT architecture is present, thread siblings are preferred.
  • isolate - The host must not have an SMT architecture or must emulate a non-SMT architecture. If the host does not have an SMT architecture, each vCPU is placed on a different core as expected. If the host does have an SMT architecture - that is, one or more cores have thread siblings - then each vCPU is placed on a different physical core. No vCPUs from other guests are placed on the same core. All but one thread sibling on each utilized core is therefore guaranteed to be unusable.
  • require - The host must have an SMT architecture. Each vCPU is allocated on thread siblings. If the host does not have an SMT architecture, then it is not used. If the host has an SMT architecture, but not enough cores with free thread siblings are available, then scheduling fails.
hw_cdrom_bus
Type

str

Specifies the type of disk controller to attach CD-ROM devices to. As for hw_disk_bus.

Only supported by the libvirt driver.

hw_disk_bus
Type

str

Specifies the type of disk controller to attach disk devices to.

Only supported by the libvirt driver.

Options depend on the value of nova's virt_type config option:

  • For qemu and kvm: one of scsi, virtio, uml, xen, ide, usb, or lxc.
  • For xen: one of xen or ide.
  • For uml: must be uml.
  • For lxc: must be lxc.
  • For parallels: one of ide or scsi.
hw_firmware_type

Specifies the type of firmware with which to boot the guest.

Only supported by the libvirt driver.

One of:

  • bios
  • uefi
hw_mem_encryption
Type

bool

Enables encryption of guest memory at the hardware level, if there are compute hosts available which support this. See nova's documentation on configuration of the KVM hypervisor for more details.

Only supported by the libvirt driver.

hw_virtio_packed_ring
Type

bool

Enables Packed VIRT-IO Queue feature. When set to true, instance will be scheduled to hosts that support negotiating the packed virt queue format. This feature may or may not be enabled depending on the guest driver. When used it will improve the small packet performance of network io. Only supported by the libvirt driver.

hw_pointer_model
Type

str

Input devices that allow interaction with a graphical framebuffer, for example to provide a graphic tablet for absolute cursor movement. Currently only supported by the KVM/QEMU hypervisor configuration and VNC or SPICE consoles must be enabled.

Only supported by the libvirt driver.

One of:

  • usbtablet
hw_rng_model
Type

str

Adds a random-number generator device to the image's instances. This image property by itself does not guarantee that a hardware RNG will be used; it expresses a preference that may or may not be satisfied depending upon Nova configuration.

The cloud administrator can enable and control device behavior by configuring the instance's flavor. By default:

  • The generator device is disabled.

  • /dev/urandom is used as the default entropy source. To specify a physical hardwre RNG device, use the following option in the nova.conf file:

    rng_dev_path=/dev/hwrng
  • The use of a hardware random number generator must be configured in a flavor's extra_specs by setting hw_rng:allowed to True in the flavor definition.

Only supported by the libvirt driver.

One of:

  • virtio
  • Other supported device.
hw_time_hpet
Type

bool

Adds support for the High Precision Event Timer (HPET) for x86 guests in the libvirt driver when hypervisor_type=qemu and architecture=i686 or architecture=x86_64. The timer can be enabled by setting hw_time_hpet=true. By default HPET remains disabled.

Only supported by the libvirt driver.

hw_machine_type
Type

str

For libvirt: Enables booting an ARM system using the specified machine type. If an ARM image is used and its machine type is not explicitly specified, then Compute uses the virt machine type as the default for ARMv7 and AArch64.

For Hyper-V: Specifies whether the Hyper-V instance will be a generation 1 or generation 2 VM. By default, if the property is not provided, the instances will be generation 1 VMs. If the image is specific for generation 2 VMs but the property is not provided accordingly, the instance will fail to boot.

For libvirt: Valid types can be viewed by using the virsh capabilities command (machine types are displayed in the machine tag).

For hyper-V: Acceptable values are either hyperv-gen1 or hyperv-gen2.

Only supported by the libvirt and Hyper-V drivers.

os_type
Type

str

The operating system installed on the image. The libvirt API driver contains logic that takes different actions depending on the value of the os_type parameter of the image. For example, for os_type=windows images, it creates a FAT32-based swap partition instead of a Linux swap partition, and it limits the injected host name to less than 16 characters.

Only supported by the libvirt driver.

One of:

  • linux
  • windows
hw_scsi_model
Type

str

Enables the use of VirtIO SCSI (virtio-scsi) to provide block device access for compute instances; by default, instances use VirtIO Block (virtio-blk). VirtIO SCSI is a para-virtualized SCSI controller device that provides improved scalability and performance, and supports advanced SCSI hardware.

Only supported by the libvirt driver.

One of:

  • virtio-scsi
hw_serial_port_count
Type

int

Specifies the count of serial ports that should be provided. If hw:serial_port_count is not set in the flavor's extra_specs, then any count is permitted. If hw:serial_port_count is set, then this provides the default serial port count. It is permitted to override the default serial port count, but only with a lower value.

Only supported by the libvirt driver.

hw_video_model
Type

str

The graphic device model presented to the guest. none disables the graphics device in the guest and should generally be used when using GPU passthrough.

One of:

  • vga
  • cirrus
  • vmvga
  • xen
  • qxl
  • virtio
  • gop
  • none
  • bochs

Only supported by the libvirt driver.

hw_video_ram
Type

int

Maximum RAM in MB for the video image. Used only if a hw_video:ram_max_mb value has been set in the flavor's extra_specs and that value is higher than the value set in hw_video_ram.

Only supported by the libvirt driver.

hw_watchdog_action
Type

str

Enables a virtual hardware watchdog device that carries out the specified action if the server hangs. The watchdog uses the i6300esb device (emulating a PCI Intel 6300ESB). If hw_watchdog_action is not specified, the watchdog is disabled.

Only supported by the libvirt driver.

One of:

  • disabled - (default) The device is not attached. Allows the user to disable the watchdog for the image, even if it has been enabled using the image's flavor.
  • reset - Forcefully reset the guest.
  • poweroff - Forcefully power off the guest.
  • pause - Pause the guest.
  • none - Only enable the watchdog; do nothing if the server hangs.
os_command_line
Type

str

The kernel command line to be used by the libvirt driver, instead of the default. For Linux Containers (LXC), the value is used as arguments for initialization. This key is valid only for Amazon kernel, ramdisk, or machine images (aki, ari, or ami).

Only supported by the libvirt driver.

hw_vif_model
Type

str

Specifies the model of virtual network interface device to use.

Only supported by the libvirt driver and VMware API drivers.

The valid options depend on the configured hypervisor.

  • KVM and QEMU: e1000, e1000e, ne2k_pci, pcnet, rtl8139, virtio and vmxnet3.
  • VMware: e1000, e1000e, VirtualE1000, VirtualE1000e, VirtualPCNet32, VirtualVmxnet and VirtualVmxnet3.
  • Xen: e1000, netfront, ne2k_pci, pcnet, and rtl8139.
hw_vif_multiqueue_enabled
Type

bool

If true, this enables the virtio-net multiqueue feature. In this case, the driver sets the number of queues equal to the number of guest vCPUs. This makes the network performance scale across a number of vCPUs.

Only supported by the libvirt driver.

hw_boot_menu
Type

bool

If true, enables the BIOS bootmenu. In cases where both the image metadata and Extra Spec are set, the Extra Spec setting is used. This allows for flexibility in setting/overriding the default behavior as needed.

Only supported by the libvirt driver.

hw_pmu
Type

bool

Controls emulation of a virtual performance monitoring unit (vPMU) in the guest. To reduce latency in realtime workloads disable the vPMU by setting hw_pmu=false.

Only supported by the libvirt driver.

img_hide_hypervisor_id
Type

bool

Some hypervisors add a signature to their guests. While the presence of the signature can enable some paravirtualization features on the guest, it can also have the effect of preventing some drivers from loading. Hiding the signature by setting this property to true may allow such drivers to load and work.

Only supported by the libvirt driver.

vmware_adaptertype
Type

str

The virtual SCSI or IDE controller used by the hypervisor.

Only supported by the VMWare API driver.

One of:

  • lsiLogic
  • lsiLogicsas
  • busLogic
  • ide
  • paraVirtual
vmware_ostype

A VMware GuestID which describes the operating system installed in the image. This value is passed to the hypervisor when creating a virtual machine. If not specified, the key defaults to otherGuest. See thinkvirt.com for supported values.

Only supported by the VMWare API driver.

vmware_image_version
Type

int

Currently unused.

instance_type_rxtx_factor
Type

float

Deprecated and currently unused.

auto_disk_config
Type

bool

Deprecated and currently unused.