This article lists important new features for Red Hat Enterprise Linux OpenStack Platform 5. For a complete listing of enhancements included in this version, see the Release Notes at https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux_OpenStack_Platform/5.
1. Red Hat Enterprise Linux OpenStack Platform Installer
The new Red Hat Enterprise Linux OpenStack Platform Installer is a wizard-based tool for the easy deployment of Red Hat Enterprise Linux OpenStack Platform across a set of hardware. The Installer builds upon the Foreman deployment tool's capabilities and makes enterprise-grade installations much easier.
The following features are included in the Installer:
- Simplified interface, offering a tool that is easier to use than Foreman and more robust than PackStack.
- Enables the orchestration and ordering of complex hardware environments, and supports multi-node deployment.
- Fully automates stage deployments; provides the automatic discovery of hardware, including bare-metal hosts.
The following modes are supported:
- Distributed, Compute networking (nova-network)
- Distributed, OpenStack Networking (neutron)
- Distributed with HA, Compute networking (nova-network)
- Distributed with HA, OpenStack Networking (neutron)
For more information about the Installer, refer to the Deploying OpenStack: Enterprise Environments (Red Hat Enterprise Linux OpenStack Platform Installer)
2. High Availability
Red Hat Enterprise Linux OpenStack Platform is now fully integrated with the Red Hat Enterprise Linux High Availability Add-On, to support highly available environments for customer deployments. This means that a cloud infrastructure can now be set up so that if one of its controller nodes fails (for example, because its host goes down), the machine can be brought back up and no data is lost (protection against single point of failure).
The following modes are supported:
Active / Active
In an active/active configuration, all system components are kept online; if a component fails, its load is passed to the next active component.
The majority of OpenStack services are configured to run in active/active configuration to provide better performance through load-balancing for more demanding deployments. To guarantee that services are constantly kept online, all OpenStack services are managed by pacemaker (RHEL-HA resource manager). For example, if a service fails, it is immediately restarted on a more suitable node to avoid any downtime.
Active / Passive
In this configuration, an offline backup is maintained for critical components. A small number of OpenStack services use an active/passive configuration for high availability.
The Red Hat Enterprise Linux OpenStack Platform Installer has built-in support for highly available environments by including the following default host groups (configuration sets):
- HA MySQL Node - Sets up an HA MariaDB Galera cluster, which can be used as the database back end for various service endpoints.
- Load Balancer - Sets up a private network used for load balancing, including a central administrative node.
- HA All In One Controller - Sets up the controller node for the highly available environment.
- Swift Storage Node - Sets up an Object Storage service host cluster, and ensures all relevant firewall rules are in place.
The first three groups all require Pacemaker and fencing techniques.
For more information about the High Availability Add-On, refer to the High Availability Add-On Overview.
3. Live Upgrades
In past releases, administrators have had to 'turn off' the cloud, in order to do an upgrade from one version to the next. Virtual machines could stay up, but all OpenStack services had to remain shut down. Furthermore, all components of Compute had to be on the same version of OpenStack.
With Red Hat Enterprise Linux OpenStack Platform 5, you can now use a combination of control-service upgrades and the live migration of instances to keep your cloud alive during upgrade:
- You first upgrade the control services atomically. For example, you would upgrade nova-api, nova-conductor, nova-scheduler, and the database schema. RHELOSP 4 compute nodes can now coexist with RHELOSP 5 nodes during the upgrade. Note: Your control-plane upgrade requires downtime, but this will be shorter because you will not have to upgrade compute nodes at the same time.
- You can then use live or cold migration to move instances over to newer RHELOSP 5 nodes.
- Finally, you upgrade all the compute nodes, and rehost the instances where necessary.
For more information, refer to the Upgrading OpenStack by Updating All Services Individually, with Live Compute article.
4. Dashboard Interface
The Dashboard has a number of changes to make it more user friendly, including:
The dashboard interface now works with horizontal tabs rather than the old vertical-accordion style paneling, providing an easy overview for the user.
OpenStack supports the migration of one virtual machine instance from one Compute host to another. Previously, the Dashboard only supported migration which stopped and then rebooted the instance. With Red Hat Enterprise Linux OpenStack Platform 5, users with an administrative login can now start a Live Migration from the Dashboard, allowing your instance to keep running during the process.
Support Access Option
The updated Red Hat Access tab gives you the option of signing in to the Red Hat Customer Portal, and searching for articles and solutions. Users can enter error codes, lines from a log file, or any set of keywords to produce a relevant query.
- The Log option displays a list of the active instances. Click on the View Local logs button to see logs from the Dashboard, or an instance's View Log button to view the console log for that instance.
- The Support option allows you to view and update your support cases. You can also use this option to open new support cases using the Open a New Support Case button. For more details, refer to Red Hat Access: RHOS OpenStack Dashboard Plug-in.
New Host Aggregates Tab
You can now add or modify host aggregates and availability zones using the Dashboard.
5. Default Software
SkySQL and MariaDB are now officially Certified as an OpenStack Partner, and MariaDB is the database of choice that is shipped with Red Hat Enterprise Linux. Red Hat Enterprise Linux OpenStack Platform now uses MariaDB instead of MySQL, enabling Red Hat to fully support community-developed software that is completely open source.
RabbitMQ is a robust open-source messaging system based on the AMQP standard; RabbitMQ is well known for its high performance in enterprise systems and having widespread commercial support. In Red Hat Enterprise Linux OpenStack Platform 5, RabbitMQ replaces Qpid as the default (and recommended) message broker.