No boot after latest updates... #10312
-
I'm not sure if it's the chicken or the egg causing this, but none of our SO VM's will boot this week. Other VM's are not effected. SOUP was run last week and had some updates, proxmox, which hosts all of our SO VM's, was updated last Friday. Today I went to give the SO VM's a reboot because.... At boot it does this: Then after sitting on the boot sequence for a really long time it does this: Some investigation reveals that the SO installs have lost VIRTIO support. I have tried booting older QEMU versions, but this does not help. All "virtIO" based virtualized hardware fails to map/work at boot. If I switch the boot disk to SATA then it boots, but it leaves me without networking. Was virtio support removed recently from SO? Or is SO installed on such an old linux distro that proxmox no longer supports SO? Not sure... I seem to recall discussing this configuration with SO devs awhile back on the phone and I think yall mentioned that you have a proxmox cluster in-house used for development/testing. Have you recently seen this issue? Any fix? Thanks, |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments
-
I kept experimenting with things and believe I found the culprit. I believe there's a bug in the latest Proxmox Kernel causing a problem on the particular platform our cluster is built from that is impacting some VM's. I also created a thread on Proxmox forums and answered my own question after finding a "solution." Ultimately, it's more of a patch but it's easy and gets the system up and running again. The particular error exposed when the CPU architecture in-place was selected indicates that the Kernel developers for linux have some work to do. https://forum.proxmox.com/threads/centos-7-based-vms-wont-boot-anymore.127117/ |
Beta Was this translation helpful? Give feedback.
-
Closing discussion as I don't believe this issue is an SO issue... |
Beta Was this translation helpful? Give feedback.
I kept experimenting with things and believe I found the culprit.
I believe there's a bug in the latest Proxmox Kernel causing a problem on the particular platform our cluster is built from that is impacting some VM's.
I also created a thread on Proxmox forums and answered my own question after finding a "solution." Ultimately, it's more of a patch but it's easy and gets the system up and running again. The particular error exposed when the CPU architecture in-place was selected indicates that the Kernel developers for linux have some work to do.
https://forum.proxmox.com/threads/centos-7-based-vms-wont-boot-anymore.127117/