VM Crashing

[eshchar]

On my system I have an M.2 SSD that I use with Unassigned Devices and store my 2 VM imgs (C and D).

M.2 is formatted to btrfs.

in the VM C.img bus is virtio and I also tried scsi with no lock.

And my problem is that when I play some games they cause the M.2 drive to disappear from the Unassigned Devices and the windows to crash.

 

[JorgeB]

Please post the diagnostics after a crash.

[xlucero1]

@JorgeB can you please link post/lead for me how to build Win 10 VM. I’m on 6.10.3 using 12900k with 690 mobo. I built old vm in 6.8 and it doesn’t work with new build. Kinda desperate here lol 🍔

[JonathanM]

1 hour ago, xlucero1 said:

Kinda desperate here

Posting the diagnostics after the crash and before you reboot Unraid should help figure out what is happening.

[xlucero1]

Thanks for responding. I mean, I can’t find out how to even build the VM. Sorry if I hijacked the thread but is there a video guide or anything to help with building a Win 10 VM?
[I seen @SpaceInvaderOne videos (which helped me when I built on 6.8) but nothing really works and I’m not sure where to begin]. Thank you 🙏🏽 

 

UPDATE: the documents pertaining to VM's on Unraid are here: https://wiki.unraid.net/Manual under VM Management & VM Guest Support

Edited July 28, 2022 by xlucero1
Update: Answer to my question

[JorgeB]

8 hours ago, xlucero1 said:

I’m not sure where to begin

You should create a new post in the KVM section, explain the problem and post the diagnostics and the VM XML.

[eshchar]

On 7/25/2022 at 10:31 AM, JorgeB said:

Please post the diagnostics after a crash.

tower-diagnostics-20220726-2123.zip

The name of the VM is Windows 10 new

the name of the despairing drive is Samsung_SSD_970_EVO_1TB_S467NX0M104640T   nvme0n1

 

[JorgeB]

Jul 26 21:22:50 Tower kernel: nvme nvme0: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10
Jul 26 21:22:50 Tower kernel: blk_update_request: I/O error, dev nvme0n1, sector 627533784 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 0
Jul 26 21:22:50 Tower kernel: nvme 0000:02:00.0: enabling device (0000 -> 0002)
Jul 26 21:22:50 Tower kernel: nvme nvme0: Removing after probe failure status: -19

 

NVMe device dropped offline, this can sometimes help:

 

Some NVMe devices have issues with power states on Linux, try this, on the main GUI page click on flash, scroll down to "Syslinux Configuration", make sure it's set to "menu view" (on the top right) and add this to your default boot option, after "append initrd=/bzroot"

nvme_core.default_ps_max_latency_us=0

e.g.:

append initrd=/bzroot nvme_core.default_ps_max_latency_us=0

Reboot and see if it makes a difference.

 

P.S. server is running out of RAM, you should limit resources.

[eshchar]

9 hours ago, JorgeB said:

Jul 26 21:22:50 Tower kernel: nvme nvme0: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10
Jul 26 21:22:50 Tower kernel: blk_update_request: I/O error, dev nvme0n1, sector 627533784 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 0
Jul 26 21:22:50 Tower kernel: nvme 0000:02:00.0: enabling device (0000 -> 0002)
Jul 26 21:22:50 Tower kernel: nvme nvme0: Removing after probe failure status: -19

 

NVMe device dropped offline, this can sometimes help:

 

Some NVMe devices have issues with power states on Linux, try this, on the main GUI page click on flash, scroll down to "Syslinux Configuration", make sure it's set to "menu view" (on the top right) and add this to your default boot option, after "append initrd=/bzroot"

nvme_core.default_ps_max_latency_us=0

e.g.:

append initrd=/bzroot nvme_core.default_ps_max_latency_us=0

Reboot and see if it makes a difference.

 

P.S. server is running out of RAM, you should limit resources.

i did add the nvme_core.default_ps_max_latency_us=0 and still crashed my vmtower-diagnostics-20220727-2232.zip

[JorgeB]

It dropped again, try a different M.2 slot if available, failing that try a different brand/model NVMe device (or a different board).

[eshchar]

I don't have lying around more nvme devices and my board have only 2 slots of M.2 conations the second one will take 2 sata conation off line.

it only happens when playing game or doing something that is ram intensive.

[JorgeB]

You can also try a BIOS update and/or update to Unraid v6.11.0-rc2, sometimes just a BIOS/kernel change can help, if it keeps dropping it's likely hardware related, but not necessarily bad hardware.

[eshchar]

i have updated to the latest bios and it did not help and i don't like the non stable version of unraid 

[eshchar]

I have moved the Windows vm  img to the cache pool form the M.2 and the issue completely stopped.

Just to add some info on M.2 i hade 2 img's C and D for the same VM.