Power control support for both Master and Slave NexVisors.
Complete storage Fast path support for both local and remote READs and WRITEs.
DHCP gateway service isolation and IP address allocation.
CPU overcommit support in the UI.
Use vendor assigned Hardware MAC addresses.
SSL connection support for the UI.
Improved Resource group creation page.
Improved Infrastructure index page.
Initial support for JWT auth.
General stability improvements.
General API/CLI optimisations.
In Development - targetted for an upcoming release
Backup and restore of VM data to a remote storage target.
Automated failover of the Sunlight controller service in an event of failure.
Disaster recovery procedure.
NexVisor automatic failover.
NexVisor discovery mode and auto addition to a cluster.
Manual placement and motion of VMs between different NexVisor nodes from the UI
Windows guest support. This is actually planned for the next verison.
Graphical mode console support.
Multi-user support with Role Based Access Control.
Striping of volumes across multiple physical NVMe drives to enable larger volume capacity and for enhancing single drive performance to 1 Million IOPs and above.
Increased NIC speed support to 25/50/100gbit support. Supported Hardware revisions are planned in 2019 to accommodate faster NIC speeds. Sunlight will be releasing a custom 100gbit/s acceleration card that will significantly enhance virtualised NIC performance.
Online Upgrade Support. This has taken priority in our dev cycle.
Multi-LUN aggregation support (creating super LUNs made up of many smaller LUNs that can span physical NVMe drives for increased capacity)
When power cycling the Sunlight cluster, even with a Fault Tolerance policy configured, automatic VM start-up is not currently supported when the cluster comes online. Fault Tolerance policy is only currently executed when the controller node stays online.
Failover of network traffic between physical ethernet ports is not currently supported in this release. All 10 GBit ethernet ports must be connected for proper product behaviour.
Multi-user and RBAC support is not supported in this version.
During this phase, all upgrades will require controller and slave nodes to be powered down. Upgrade/install is achieved via the boot configuration option.
We suggest that the maximum number of instances created in a cluster (in one go) is limited to less or equal than 8. We are investigating an issue relating to large scale clusters.