OpenStack Yoga
Updated version from Victoria to Yoga
We’re excited to announce that we’ve updated the version of OpenStack installed with our new Private Clouds. Clouds will be installed the Yoga version of OpenStack.
A few notable changes introduced with the release:
- Cinder: Block Storage API microversion 3.68 adds the ability for users to request that a volume be re-imaged. It has always been possible in Block Storage API version 3 to create a bootable volume by requesting that image data from the Image service be written to the volume when the volume was created; now it’s possible to do this to an existing volume.
- Horizon: Horizon added experimental support of System Scope. There is a new entry System Scope added in context switcher menu which allows users to switch to a system scope token so operations that require this kind of token can be performed. It is disabled by default in the Yoga release.
- Neutron: Local IP – a virtual IP which can be shared across multiple ports or VMs is now available. Local IP is guaranteed to only be reachable within the same physical server or node boundaries.
- Octavia: Octavia load balancers now support deep observability by adding
PROMETHEUS
listeners that expose a Prometheus exporter endpoint. The Octavia amphora provider exposes over 150 unique metrics.
For more information, please see the Release Notes
Ceph Updates
Updated version from Octopus to Quincy
We provide Ceph with all Private Clouds to address block, file, and object storage. All new clouds will now be deployed with the Quincy release of Ceph.
A few notable changes from the release:
- Filestore has been deprecated in Quincy. BlueStore is Ceph’s default object store.
- LevelDB support has been removed.
WITH_LEVELDB
is no longer a supported build option. Users should migrate their monitors and OSDs to RocksDB before upgrading to Quincy. - A health warning is now reported if the
require-osd-release
flag is not set to the appropriate release after a cluster upgrade. - The
device_health_metrics
pool has been renamed.mgr
. It is now used as a common store for allceph-mgr
modules. After upgrading to Quincy, thedevice_health_metrics
pool will be renamed to.mgr
on existing clusters.
For more information, please see the Release Notes
Deployment System
Our primary focus with our recent release has been updating the deployment system we use to setup your infustructure and services. Our new system will allow us to deliver a wider range of services and cloud types. We plan to offer Bare Metal clusters within your OpenMetal Central interface within the coming months!
For those of you familiar with the initial deployment system, the fm-deploy
container has been replaced by a node-agent
container on each of your hardware nodes.
Expanded Monitoring
We’ve created a new Datadog dashboard & alerts to make it easier to monitor your critical services. We’ve also expanded on our internal monitoring of your hardware to continue providing you exceptional uptime!