> > > Just applied the fix to the first affected AMD machine and all is well > > again. > > > > Would this be worth putting a ticket into Vultr to get them to make > > appropriate updates on their side? > > Yes (but I see you already did) >
Here's the reply I got from Vultr about this: Thank you for contacting us. We are aware of OpenBSD instances dropping to ddb upon boot after the initial zenbleed patches went out. This was fixed as of the 015 patch that was released today: https://ftp.openbsd.org/pub/OpenBSD/patches/7.3/common/015_hvamdcpu.patch.sig If you have an OpenBSD instance that is still stuck at ddb and therefore cannot apply the 015 patch, please let us know, as there is a manual workaround we can do to get it to boot. After that, you can run syspatch again which should install that newest patch, and you should have no issues. Regarding the Zenbleed vulnerability itself, none of our AMD hosts are known to be vulnerable at this time as they are all running Milan and later CPUs. The CPU config provided to guests will say Rome, but that is provided for compatibility purposes and is not related to the actual host node CPU. We are continuing to monitor this vulnerability, and will be taking appropriate steps to mitigate it once the full impact has been determined.
