h3llfire83

Thanks for pointing me in the right direction! I wasn't aware of the dedicated CVE reporting board. I thought the Security section of the forum would be the appropriate place to raise this concern. I'll submit the report there as suggested. Thanks again for the clarification JorgeB 👍

Hi Unraid Team, I'd like to raise awareness of the recently disclosed Linux kernel vulnerability CVE-2026-99999 (PwnEscape) and ask whether its potential impact on current Unraid OS releases has already been assessed. The vulnerability affects IPv4 checksum handling in the Linux kernel and may allow an attacker to escape from a KVM guest and potentially execute code on the host under certain conditions. According to the published release notes: Unraid OS 7.3.3 (stable, October 8) uses Linux kernel 6.18.54-Unraid. Unraid OS 7.4.0-rc1 also uses Linux kernel 6.18.54-Unraid. The upstream fix was integrated into the Linux 6.18 stable series with version 6.18.57, released on October 15. Since Unraid also relies on KVM for virtualization, this raises a few questions: Has the Unraid team already assessed CVE-2026-99999? Are current releases affected, or has the relevant fix already been backported into the Unraid kernel? If affected, is a security update planned for the current stable 7.3.x branch? Would a targeted kernel update or hotfix be possible without waiting for Unraid 7.4 or the upcoming Unraid 8? Are there any officially recommended temporary mitigations for users running virtual machines until a patched kernel becomes available? Could the team provide a security advisory or confirmation of the vulnerability status so users can assess the risk to their systems? I'm aware that Unraid 8 is expected to introduce a different underlying distribution and improvements to security update delivery. However, given the potential severity of this vulnerability, I believe it would be helpful to clarify the situation for current Unraid 7.x installations as well. This is not intended as criticism of the development or release schedule. I'm primarily interested in understanding the actual exposure and whether users need to take any immediate action. References: Research and technical analysis: https://www.pwn.ai/blog/pwnescape Upstream kernel fix: https://github.com/torvalds/linux/commit/6785011f8b16abf40b1e9d8b2e332232b3e68822 Linux 6.18.57 changelog: https://www.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.18.57 Red Hat vulnerability assessment: https://access.redhat.com/security/cve/cve-2026-99999 Thanks in advance for any clarification or guidance from the Unraid team

I had it running in earlier Versions (macOS 11/12) under a earlier Unraid Build, but i replaced it with the 6800XT and other HW

Can you give me some Details ? Do you have setup the Bootarguments ? agdpmod=pikera Maybe you can fix it : https://github.com/TylerLyczak/Unsupported-6900XT-Hackintosh-Fix keep my fingers crossed🤞 But still i dont know why it does not work with other QEMU Version above 6.0

I myself have had difficulties for months using macinabox with a graphics card after a motherboard change. I even recently updated to the latest Unraid version, although it's a beta, but without success. Apparently, the issue lies with the selected QEMU version. With all the new versions, despite correct IDs and multifunction=1 for the graphics card, I was never able to boot the VM successfully with the graphics card. The VM would get stuck at the Apple logo. In the meantime, I also had trouble passing through a dedicated USB-PCI card correctly. Over VNC, there were never any issues. I am using a 6800XT and not a 6900 like you (here, the card may need to be spoofed). What works for me now is QEMU version 6.0; unfortunately, anything above that doesn’t work and causes the VM to hang at the Apple logo again. Maybe this helps someone. Can someone explain why it works with QEMU v6.0 and why other QEMU versions cause this issue? I’m just glad to have GPU passthrough working again under Unraid and macOS, and I’m happy, but I'm still curious about the root cause. Of course, I have updated OpenCore. I was about to give up since this issue has been frustrating me for months, and without graphics acceleration, macOS wasn’t really usable for my purposes.