i have HVM: Enabled and IOMMU: Enabled  when i manage my wm i cant choose my nic  i have 4 network  ports in unraid dashbord so i know unraid finds the card

Because you are using one of the 4 for unRAID you are not going to be able to stub them (most likely) as their device IDs will likely be the same.  But without the information provided in a Diagnostic I can't be positive about that.  You can look in the System Devices display and find your nics.  Once you do you can edit the XML and pass them through to the VM without stubbing them.  I ran my VMs that way for a while as unRAID will disconnect the affected nic from itself (unRaid) to make it available to the VM automatically when you start the VM.  You can also use the virtual nic that is automatically created when you create the VM.  The existing nics could be bonded to provide redundancy and increase in speed.  Then the virtual nic would share a bigger pipe to the outside world.  Just don't ask ME about bonding as my networking savvy is not that great.

here is the diagnostic info

tower-diagnostics-20171022-2036.zip

Making assumption that the following are your MB nics:

01:00.0 Ethernet controller [0200]: Intel Corporation 82575EB Gigabit Network Connection [8086:10a7] (rev 02)
	Subsystem: Intel Corporation 82575EB Gigabit Network Connection [8086:34dc]
	Kernel driver in use: igb
	Kernel modules: igb
01:00.1 Ethernet controller [0200]: Intel Corporation 82575EB Gigabit Network Connection [8086:10a7] (rev 02)
	Subsystem: Intel Corporation 82575EB Gigabit Network Connection [8086:34dc]
	Kernel driver in use: igb
	Kernel modules: igb
04:00.0 Ethernet controller [0200]: Intel Corporation 82571EB Gigabit Ethernet Controller [8086:105e] (rev 06)
	Subsystem: Hewlett-Packard Company NC360T PCI Express Dual Port Gigabit Server Adapter [103c:7044]
	Kernel driver in use: e1000e
	Kernel modules: e1000e
04:00.1 Ethernet controller [0200]: Intel Corporation 82571EB Gigabit Ethernet Controller [8086:105e] (rev 06)
	Subsystem: Hewlett-Packard Company NC360T PCI Express Dual Port Gigabit Server Adapter [103c:7044]
	Kernel driver in use: e1000e
	Kernel modules: e1000e

Based on your diagnostic info it looks like eth0 is device 01:00.0 above.  So you could pass through any of the other 3 to a VM since each one looks to be in a unique IOMMU group also based on the diagnostics.  I would suggest you use one of the E1000 nics above as that is going to be most compatible with VMs.

 

So you could add the following XML to the VM's xml and pass through the first E1000 nic to the VM:

    <hostdev mode='subsystem' type='pci' managed='yes'>
      <driver name='vfio'/>
      <source>
        <address domain='0x0000' bus='0x04' slot='0x00' function='0x0'/>
      </source>
    </hostdev>

Note when you paste it into the VMs XML make sure you do it after the "</video>" and before the "<memballoon model".  After it is pasted into the XML unRAID will automatically add an "<address type" line.  If you want to pass through the 2nd E1000 nic you would change the "function=" to "0x1" above.  I would use the 2nd igb nic last when passing through to VMs that is a newer nic than the E1000 so it would like work fine with newer versions of Windows/Linux but Windows 7 (my favorite) would need a driver to work with it but the E1000 may work immediately with just windows drivers.

 

Yep checked with my box and I'm fairly sure I got it correct for your MB.

Edited October 22, 20178 yr by BobPhoenix

this worked can i do this to add both e1000 ?

 

    <hostdev mode='subsystem' type='pci' managed='yes'>
      <driver name='vfio'/>
      <source>
        <address domain='0x0000' bus='0x04' slot='0x00' function='0x0'/>
      </source>
    </hostdev>'

    <hostdev mode='subsystem' type='pci' managed='yes'>
      <driver name='vfio'/>
      <source>
        <address domain='0x0000' bus='0x04' slot='0x00' function='0x1'/>
      </source>
    </hostdev>

 

 

Yep that should work.

that worked just fine thanks for your help

Edited October 22, 20178 yr by nico1995doom