VM messed up after removing physical drive, now IO error reported

[thejessman321]

I was trying to remove a drive from the physical host. I pulled it from the server. It didn't reflect the changes in Windows Server 2016. I tried refreshing the server manager and it still showed up despite being physically removed. So I thought let me try creating a folder on the disk in windows explorer and surely it'll realize it can't write to it and just remove the disk. WRONG! So now the VM shows up as io error with a yellow triangle. I put the disk back in but it still shows up like that.

Is there any way to recover from this without having to force close the VM and restart? I was hoping to not have to do that. I did a "qm resume (vm #)" but that didn't work. Any solutions aside from losing any open data and restarting? Thanks!

 

[thejessman321]

Below is what it says after running the restore command.....

Dec 07 12:14:38 deathstar QEMU[3605]: virtio-blk failed to set guest notifier (-24), ensure -accel kvm is set.
Dec 07 12:14:38 deathstar QEMU[3605]: virtio_bus_start_ioeventfd: failed. Fallback to userspace (slower).
Dec 07 12:14:38 deathstar QEMU[3605]: virtio-blk failed to set guest notifier (-24), ensure -accel kvm is set.
Dec 07 12:14:38 deathstar QEMU[3605]: virtio_bus_start_ioeventfd: failed. Fallback to userspace (slower).
Dec 07 12:14:38 deathstar QEMU[3605]: virtio-blk failed to set guest notifier (-24), ensure -accel kvm is set.
Dec 07 12:14:38 deathstar QEMU[3605]: virtio_bus_start_ioeventfd: failed. Fallback to userspace (slower).

 

[thejessman321]

Is there any way to fix this to allow the VM to continue or am I forced to just do a stop and whatever data is lost is the cost? It seems like it's still going, but idk. The console won't open up. So am I just screwed?

 

[bbgeek17]

Its unclear how this disk is/was connected to the VM, That said, there are situations that you just cant recover from. Usually those are exacerbated by the environment where they happen. Physically removing a disk that was passed through to a VM (?), which was running Windows - bad combination all around.
Its very unlikely that you can recover from it without reboot or even full shutdown of the VM. Even then, depending on what that disk was used for, there is a risk the VM(windows) wont boot without some sort of recovery.

Good luck.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[thejessman321]

Its unclear how this disk is/was connected to the VM, That said, there are situations that you just cant recover from. Usually those are exacerbated by the environment where they happen. Physically removing a disk that was passed through to a VM (?), which was running Windows - bad combination all around.
Its very unlikely that you can recover from it without reboot or even full shutdown of the VM. Even then, depending on what that disk was used for, there is a risk the VM(windows) wont boot without some sort of recovery.

Good luck.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

Yeah. That's what I figured. It was done using passthrough. I pulled the drives physically without thinking to offline them in windows and then detach from the vm in proxmox. So I guess I'll just have to stop the vm and restart it. Sucks to lose anything that was open, but I guess there's no choice now. It's still running but shows the io error. It seems like there's nothing left to try. Thank you though.