Image service overview¶
The Image service (glance) enables users to discover, register, and retrieve virtual machine images. It offers a REST API that enables you to query virtual machine image metadata and retrieve an actual image. You can store virtual machine images made available through the Image service in a variety of locations, from simple file systems to object-storage systems like OpenStack Object Storage.
Important
For simplicity, this guide describes configuring the Image service to
use the file back end, which uploads and stores in a
directory on the controller node hosting the Image service. By
default, this directory is /var/lib/glance/images/.
Before you proceed, ensure that the controller node has at least
several gigabytes of space available in this directory. Keep in
mind that since the file back end is often local to a controller
node, it is not typically suitable for a multi-node glance deployment.
For information on requirements for other back ends, see Configuration Reference.
The OpenStack Image service is central to Infrastructure-as-a-Service (IaaS). It accepts API requests for disk or server images, and metadata definitions from end users or OpenStack Compute components. It also supports the storage of disk or server images on various repository types, including OpenStack Object Storage.
A number of periodic processes run on the OpenStack Image service to support caching. Replication services ensure consistency and availability through the cluster. Other periodic processes include auditors, updaters, and reapers.
The OpenStack Image service includes the following components:
- glance-api
Accepts Image API calls for image discovery, retrieval, and storage.
Note
An OpenStack Community Goal in the Pike release was Control Plane API endpoints deployment via WSGI. As currently constituted, however, glance-api is not suitable to be run in such a configuration. Instead we recommend that Glance be run in the traditional manner as a standalone server. See the “Known Issues” section of the Glance Release Notes for the Pike and Queens releases for more information.
- glance-registry
Stores, processes, and retrieves metadata about images. Metadata includes items such as size and type.
Warning
The registry is a private internal service meant for use by OpenStack Image service. Do not expose this service to users.
Note
The Glance Registry Service and its APIs have been DEPRECATED in the Queens release and are subject to removal at the beginning of the ‘S’ development cycle, following the OpenStack standard deprecation policy.
For more information, see the Glance specification document Actually Deprecate the Glance Registry.
- Database
Stores image metadata and you can choose your database depending on your preference. Most deployments use MySQL or SQLite.
- Storage repository for image files
Various repository types are supported including normal file systems (or any filesystem mounted on the glance-api controller node), Object Storage, RADOS block devices, VMware datastore, and HTTP. Note that some repositories will only support read-only usage.
- Metadata definition service
A common API for vendors, admins, services, and users to meaningfully define their own custom metadata. This metadata can be used on different types of resources like images, artifacts, volumes, flavors, and aggregates. A definition includes the new property’s key, description, constraints, and the resource types which it can be associated with.
Running Glance Under Python3¶
You should always run Glance under whatever version of Python your distribution of OpenStack specifies.
If you are building OpenStack yourself from source, Glance is currently supported to run under Python2 (specifically, Python 2.7 or later).
Some deployment configuration is required if you wish to run Glance under Python3. Glance is tested with unit and functional tests running Python 3.5. The eventlet-based server that Glance runs, however, is currently affected by a bug that prevents SSL handshakes from completing (see Bug #1482633). Thus if you wish to run Glance under Python 3.5, you must deploy Glance in such a way that SSL termination is handled by something like HAProxy before calls reach Glance.
