News for hackers. Stuff that matters.
Re:CVE-2026-99312: KVM Guest-to-Host Escape by gbrindisi on Wednesday November 18, 2026 @11:50AM (#50548619)
Thinking about agents sandboxes... it truly seems like the time is ripe for unikernels
Re:CVE-2026-99312: KVM Guest-to-Host Escape by wmf on Wednesday November 18, 2026 @01:14PM (#50549890)
Unikernels that run on KVM aren't going to protect against KVM escapes.
Re:CVE-2026-99312: KVM Guest-to-Host Escape by rvz on Wednesday November 18, 2026 @01:38PM (#50550248)
Maybe the problem...is Linux itself.
Re:CVE-2026-99312: KVM Guest-to-Host Escape by imhoguy on Wednesday November 18, 2026 @03:34PM (#50551767)
Would FreeBSD be better here?
Re:CVE-2026-99312: KVM Guest-to-Host Escape by wmf on Wednesday November 18, 2026 @04:16PM (#50552230)
bhyve escapes just haven't been discovered yet.
Re:CVE-2026-99312: KVM Guest-to-Host Escape by ptx on Wednesday November 18, 2026 @04:41PM (#50552516)
There was a DoS vulnerability in FreeBSD's vmm announced earlier today (letting the guest panic the host on Intel CPUs): https://www.freebsd.org/security/advisories/FreeBSD-SA-26:84...
Re:CVE-2026-99312: KVM Guest-to-Host Escape by eyberg on Wednesday November 18, 2026 @02:04PM (#50550590)
As others have stated this is a nested issue. Most unikernel users I know deploy straight to ec2 or gcp. This is going to affect "sandbox" providers more than anything else.
Re:CVE-2026-99312: KVM Guest-to-Host Escape by fcarraldo on Wednesday November 18, 2026 @02:50PM (#50551192)
Nested virt on EC2/GCE are vulnerable. Metal instances wouldn’t be, though using nested virt on Nitro Sandbox capable EC2 instances wouldn’t be either.

So this is a tricky one. You can (and should) deploy unikernel workloads on metal and avoid nested virt but this is hard to make economically viable in some cases.

Re:CVE-2026-99312: KVM Guest-to-Host Escape by wmf on Wednesday November 18, 2026 @04:17PM (#50552236)
I've always been under the impression that bare metal is cheaper.
Re:CVE-2026-99312: KVM Guest-to-Host Escape by cryptonector on Wednesday November 18, 2026 @03:36PM (#50551794)
> This is going to affect "sandbox" providers more than anything else.

This is going to affect "sandbox" providers and their customers more than anything else.

FTFY

Re:CVE-2026-99312: KVM Guest-to-Host Escape by Kozmik1 on Wednesday November 18, 2026 @03:26PM (#50551679)
This was some of my first thinking when I posted the original link - the importance of hardening VMs for agent isolation, and the risk that bugs like this pose in legacy code - especially C codebases.