Unable to create PFSense VM following SpaceInvader One's Guide IOMMU Issues :-(

[mbc0]

Hi,

 

As the title suggests, I am trying to create a PFSense VM following Space Invader One's guide (which is excellent!) however, although I have a PCI  intel 4 port nic it's IOMMU group is bundled with other devices (not sure why that would be??)

 

so I enabled ACS and it split 2 ports into group 14 and another 2 ports into group 15

 

I am now unable to start the VM due to errors on the IOMMU groups 14 & 15

 

I added this line to my syslinux.cfg which made no difference so I am unsure where to go from here?

vfio_iommu_type1.allow_unsafe_interrupts=1

I have attached my diagnostics file, hope someone can help!

 

Thank you 

 

 

 

unraidserver2-diagnostics-20180703-0500.zip

Edited July 3, 2018 by mbc0

[1812]

this is your error:

 

Jul  3 04:59:53 unraidserver2 kernel: vfio-pci 0000:09:00.0: Device is ineligible for IOMMU domain attach due to platform RMRR requirement.  Contact your platform vendor.

 

Please see the proliant/hp link in my sig and find the section regarding rmrr errors. 

 

*note- I haven't seen this error occur with a network card before, normally GPUs. So the fix may just be bios update/settings.

Edited July 3, 2018 by 1812

[mbc0]

Hi,

 

4 hours ago, 1812 said:

this is your error:

 

Jul  3 04:59:53 unraidserver2 kernel: vfio-pci 0000:09:00.0: Device is ineligible for IOMMU domain attach due to platform RMRR requirement.  Contact your platform vendor.

 

Many thanks for your input! you say the above is my error but this is what I get when starting the VM

internal error: process exited while connecting to monitor: 2018-07-03T22:25:21.198374Z qemu-system-x86_64: -device vfio-pci,host=0a:00.1,id=hostdev0,bus=pci.2,addr=0x0: vfio error: 0000:0a:00.1: failed to setup container for group 15: failed to set iommu for container: Operation not permitted

 

I am not having much luck with that thread on rmrr errors ? maybe this card is just not going to work?

[mbc0]

I Just tried with another dual nic I had lying around and get the same issue.. after enabling ACS and stubbing the device to make it available to the VM I get this error on startup of the VM

 

internal error: process exited while connecting to monitor: 2018-07-03T23:01:20.098013Z qemu-system-x86_64: -device vfio-pci,host=07:00.0,id=hostdev0,bus=pci.2,addr=0x0: vfio error: 0000:07:00.0: failed to setup container for group 11: failed to set iommu for container: Operation not permitted

 

Is this just a Proliant Gen8 issue? I updated to the latest bios as advised but still no difference.

 

Just read through your proliant thread, sorry but I should have posted in there but to be honest it would not make a difference by the looks of thingsas others have simular unresolved issues it would seem that the Gen8 Microserver is not a good solution for the pfsense vm setup to I am going to purchase some new hardware and retire the HP back to the loft for another day!

 

Thanks for your  help all the same

Edited July 3, 2018 by mbc0

[1812]

2 hours ago, mbc0 said:

Hi,

 

internal error: process exited while connecting to monitor: 2018-07-03T22:25:21.198374Z qemu-system-x86_64: -device vfio-pci,host=0a:00.1,id=hostdev0,bus=pci.2,addr=0x0: vfio error: 0000:0a:00.1: failed to setup container for group 15: failed to set iommu for container: Operation not permitted

 

I am not having much luck with that thread on rmrr errors ? maybe this card is just not going to work?

 

That is the error that pops up on the screen. The RMRR error which occurs in the logs is the reason why it can't set the container.

 

1 hour ago, mbc0 said:

 

Is this just a Proliant Gen8 issue? I updated to prthe latest bios as advised but still no difference.

 

 

It's an HP proliant issue across several generations.  Hopefully in the very near future, the RMRR issue will be resolved. The proxmox users found a way to patch out the RMRR error and I'm hoping it will be patched in unRaid soon a well. Even if it takes a one-off version.

 

[mbc0]

Thank you for the feedback, I will keep notifications open on this thread in the hope that it happens! ?

[matto2494]

any news on this update? I am struggling with all the same issues :(

[mbc0]

yes mate, all working for me and others