You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Virtualization.framework VM execution ownership and crash reclamation on Intel macOS Monterey
#772
I'm investigating a lifecycle/security property of Apple's Virtualization.framework on Intel macOS Monterey 12.7.x, and VirtualBuddy seems particularly relevant because it directly manages VM lifecycle on top of Apple's virtualization stack.
The specific question is:
If a process creates and owns a running VZVirtualMachine, and that process is terminated abruptly — for example by crash or SIGKILL — is the underlying VM/vCPU execution guaranteed to terminate as part of that process's reclamation?
I'm trying to understand the relationship between:
the process holding the VZVirtualMachine object,
the com.apple.Virtualization.VirtualMachine / XPC backend process,
and the actual Hypervisor-backed guest execution.
I'm not asking about graceful shutdown through stop(). The security property I need is narrower and stronger: after the userspace execution owner is destroyed, guest computation must not be able to continue indefinitely as an independent execution domain.
Target environment:
Intel x86_64
macOS Monterey 12.7.x
Virtualization.framework
abrupt owner-process termination / crash reclamation
Since VirtualBuddy has direct experience with VM lifecycle management, I was wondering whether anyone here has observed or tested what happens when the process owning a running VZVirtualMachine is killed unexpectedly.
In particular:
Does the com.apple.Virtualization.VirtualMachine backend survive the client process?
Can it become orphaned or adopted by launchd?
Can guest vCPU execution continue after the original client has died?
Is there any reliable diagnostic technique to map a VM instance to the process actually owning its Hypervisor execution?
Apple Developer Forums thread with the full context: https://developer.apple.com/forums/thread/848542
Any firsthand observations, implementation notes, reproducible tests, or pointers to relevant VirtualBuddy lifecycle code would be extremely helpful.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
I'm investigating a lifecycle/security property of Apple's Virtualization.framework on Intel macOS Monterey 12.7.x, and VirtualBuddy seems particularly relevant because it directly manages VM lifecycle on top of Apple's virtualization stack.
The specific question is:
If a process creates and owns a running VZVirtualMachine, and that process is terminated abruptly — for example by crash or SIGKILL — is the underlying VM/vCPU execution guaranteed to terminate as part of that process's reclamation?
I'm trying to understand the relationship between:
I'm not asking about graceful shutdown through stop(). The security property I need is narrower and stronger: after the userspace execution owner is destroyed, guest computation must not be able to continue indefinitely as an independent execution domain.
Target environment:
Since VirtualBuddy has direct experience with VM lifecycle management, I was wondering whether anyone here has observed or tested what happens when the process owning a running VZVirtualMachine is killed unexpectedly.
In particular:
Apple Developer Forums thread with the full context:
https://developer.apple.com/forums/thread/848542
Any firsthand observations, implementation notes, reproducible tests, or pointers to relevant VirtualBuddy lifecycle code would be extremely helpful.
All reactions