Oracle VM Servers can be rebooted from Oracle VM Manager, using the web user interface or the command line interface. See Restart Server in the Oracle VM Manager User's Guide. A reboot has different consequences depending on the hardware architecture that you the Oracle VM Server is running on.

On x86 hardware, a reboot causes the entire system to restart. This means that depending on the configuration of your server pool and whether the server has been put into maintenance mode first, all virtual machines that are running on the server are either migrated to an alternative server or stopped, before the server is rebooted.

On SPARC hardware, a reboot is only applied to the control domain. This mean that any virtual machines running on the SPARC-based Oracle VM Server continue to run while the control domain is rebooted. However, network access and disk I/O are blocked while the control domain is rebooting, unless a second service domain or shadow domain is available to facilitate this activity. See Configuring a Secondary Service Domain on SPARC in the Oracle VM Administrator's Guide. Once the control domain has finished rebooting, network access and disk I/O are restored for the running virtual machines and activity resumes as normal. If you are using server pool clustering, virtual machines may be migrated to an alternative server if the reboot of the control domain takes too long or does not complete for some reason.

It is worthwhile noting that a power cycle has the same effect on both x86 and SPARC servers. The entire system is restarted. All virtual machines are either stopped or migrated, depending on whether or not clustering is enabled for the server pool that the server belongs to. Killing an Oracle VM Server from within Oracle VM Manager has an equivalent effect to a power cycle. See Kill Server in the Oracle VM Manager User's Guide for more information on this.