planetwilson

Ah ok didn't know that.

You don't need any port forwarding either if your router supports upnp. Good thread on it here:- https://forums.plex.tv/discussion/203697/how-does-the-plex-web-app-access-the-server

I have exactly the same card as @gridrunner as I bought it after he recommended it. Mine works fine but I only have devices plugged directly into the card, no hub used.

Is iMessage working for you? I had to follow some guide on TonyMac86 about generating a new valid serial number first. Apple does some checking in the back end for things like iMessage and maybe that also applies to the App Store? by default the serial number created by clover might not be a valid one, you have to regenerate then check with Apple to make sure it is genuine.

No it is common, I ignore it. Doesn't seem to affect anything.

If it was badly formed XML then I don't think it would save. There were a bunch of subtle differences but nothing which stood out to me. Oh well, I'll just keep this Ubuntu VM about to edit every now and again.

Yes I have done this. I have a bunch of subdomains all coming in over https and getting routed to different dockers or VMs. In addition I have an Open VPN server that is serving over 443 as many places block the standard VPN ports. So the question is how do you handle both incoming VPN traffic as well as a bunch of sub-domains? Some of the incoming traffic is for a Windows server setup and federation with ADFS so might be similar to SBS? There are two ways this can be achieved. The first is to have OpenVPN in front of nginx and use the ability for it to pass on non-VPN traffic. I had this setup for a few months but the performance was erratic and unstable. It worked but I would get timeouts sometimes on various websites I was hosting The second way is to do the routing in nginx using an nginx module called stream and that is what I have settled on and it works rather well although it sounds like your situation might be a little easier and the other thread more helpful Attached my config.... nginx_sample.conf

So in the end I created a new VM and edited it with Virt Mgr form an Ubuntu VM as suggested slightly earlier. That seemed to resolve all my issues, passthough now working again. I can't quite work out what the changes in the XML were though.

That happened to me a few months ago. I assumed I had just done something daft and not realised. To access remotely I have OpenVPN installed in a docker to get onto my local network then access unRaid as normal

Anyone know how I make use of the previous folder on the tab drive to try and boot the previous version to try that out? Update: didn’t make any difference. Something else is going on bit frustrating

Moving to another slot with a different card in the first did nothing. I honestly cannot see what has gone wrong. Just to be clear, I have had Windows and OSX VMs working fine for with GPU passthrough months on 6.3 and the 6.4 RCs. Only when I moved to official 6.4 did it all stop working.

I am getting no display out of it at all, either OSX or Windows booting. I am passing a GTX 950 in both cases. I'll check my bios (ASRock Extreme 6 X99 - I don't remember seeing anything like that in there but will check again) The card is sitting in the first slot currently but will try a little later moving it to another one.

in fact it isn't working on Windows either. I just have a general GPU passthrough issue I think ever since moving to 6.4 full version

Well that didn't work tried dumping the ROM again as well. 2018-01-19 10:50:53.254+0000: Domain id=1 is tainted: high-privileges 2018-01-19 10:50:53.254+0000: Domain id=1 is tainted: custom-argv 2018-01-19 10:50:53.254+0000: Domain id=1 is tainted: host-cpu 2018-01-19T10:50:53.306628Z qemu-system-x86_64: -chardev pty,id=charserial0: char device redirected to /dev/pts/1 (label charserial0) 2018-01-19T10:50:55.302315Z qemu-system-x86_64: -device vfio-pci,host=03:00.0,id=hostdev0,bus=pci.3,addr=0x0,romfile=/mnt/cache/downloads/gtx950mw.dump: Failed to mmap 0000:03:00.0 BAR 3. Performance may be slow

yes I am coming to the same conclusion. I have ordered a super cheapo passive card to stick in so hopefully unRAID will use that and I can safely assign the main card to the VM. Does anyone know how unRAID decides? does it matter as it will switch to the other if I try to assign one to a VM?

2018-01-17T08:25:01.051885Z qemu-system-x86_64: -device vfio-pci,host=02:00.0,id=hostdev0,bus=pci.4,addr=0x0,romfile=/mnt/disks/VMNVM/highsierra/gtx950.rom: Failed to mmap 0000:02:00.0 BAR 3. Performance may be slow usb_desc_get_descriptor: 2 unknown type 33 (len 10) usb_desc_get_descriptor: 1 unknown type 33 (len 10) usb_desc_get_descriptor: 1 unknown type 33 (len 10) 2018-01-17T08:26:15.112664Z qemu-system-x86_64: vfio_region_write(0000:02:00.0:region3+0x1088, 0x7ffe11,8) failed: Device or resource busy 2018-01-17T08:26:15.128268Z qemu-system-x86_64: vfio_err_notifier_handler(0000:02:00.1) Unrecoverable error detected. Please collect any data possible and then kill the guest 2018-01-17T08:26:15.221718Z qemu-system-x86_64: vfio_err_notifier_handler(0000:02:00.0) Unrecoverable error detected. Please collect any data possible and then kill the guest 2018-01-17T08:26:15.221765Z qemu-system-x86_64: vfio_err_notifier_handler(0000:02:00.1) Unrecoverable error detected. Please collect any data possible and then kill the guest yeah it is just an SSD unassigned device that I run VMs from. The High Sierra VM is running off another unassigned NVMe drive. I copied the ROM file to the same location as the disk images on the NVMe just to check and still get the same issue. I am thinking this might be to do with whether it is the only graphics card. I will need to test when I get back home by throwing in another old card I have.

Do you have any other graphics in your system like built in Intel or another card?

Has anyone got PCI graphics passthrough working on OSX with an NVidia card and did you have to do a BIOS dump for the card and include in the XML? I can see on @gridrunner's latest video for Windows he hasn't dumped a bios to get the card working in Windows...but it still doesn't work for me when I try that (no onboard graphics in my setup) UPDATE: followed GridRunner's video on NVidia as primary card and edited a ROM from techpowerup but still same result with that or my original dumped one. This was working at the weekend when I was on the RC21 release.

I spoke too soon, mine was working fine but now when I start my VM I get an issue in mapping the NVidia rom file (I have a GTX 950) 2018-01-15 09:11:09.940+0000: starting up libvirt version: 3.8.0, qemu version: 2.10.2, hostname: pwnas LC_ALL=C PATH=/bin:/sbin:/usr/bin:/usr/sbin HOME=/ QEMU_AUDIO_DRV=none /usr/local/sbin/qemu -name 'guest=High Sierra Gfx,debug-threads=on' -S -object 'secret,id=masterKey0,format=raw,file=/var/lib/libvirt/qemu/domain-2-High Sierra Gfx/master-key.aes' -machine pc-q35-2.10,accel=kvm,usb=off,dump-guest-core=off,mem-merge=off -cpu host -drive file=/usr/share/qemu/ovmf-x64/OVMF_CODE-pure-efi.fd,if=pflash,format=raw,unit=0,readonly=on -drive file=/etc/libvirt/qemu/nvram/a2f6f8c2-8b33-11d9-6a41-87fa8f45543f_VARS-pure-efi.fd,if=pflash,format=raw,unit=1 -m 16384 -realtime mlock=off -smp 8,sockets=1,cores=4,threads=2 -uuid 920636ba-12a2-ee7c-eff6-a6575d058a7c -display none -no-user-config -nodefaults -chardev 'socket,id=charmonitor,path=/var/lib/libvirt/qemu/domain-2-High Sierra Gfx/monitor.sock,server,nowait' -mon chardev=charmonitor,id=monitor,mode=control -rtc base=utc,driftfix=slew -global kvm-pit.lost_tick_policy=delay -no-hpet -no-shutdown -boot strict=on 000-82545em,netdev=hostnet0,id=net0,mac=52:54:00:8b:8f:ec,bus=pci.2,addr=0x0 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -chardev 'socket,id=charchannel0,path=/var/lib/libvirt/qemu/channel/target/domain-2-High Sierra Gfx/org.qemu.guest_agent.0,server,nowait' -device virtserialport,bus=virtio-serial0.0,nr=1,chardev=charchannel0,id=channel0,name=org.qemu.guest_agent.0 -device usb-tablet,id=input0,bus=usb.0,port=1 -device vfio-pci,host=02:00.0,id=hostdev0,bus=pci.4,addr=0x0,romfile=/mnt/disks/VMSSD/vbiosbound.rom -device vfio-pci,host=02:00.1,id=hostdev1,bus=pci.5,addr=0x0 -device vfio-pci,host=07:00.0,id=hostdev2,bus=pci.6,addr=0x0 -device virtio-balloon-pci,id=balloon0,bus=pci.3,addr=0x0 -usb -device usb-mouse,bus=usb-bus.0 -device usb-kbd,bus=usb-bus.0 -smbios type=2 -cpu Penryn,vendor=GenuineIntel,kvm=on,+invtsc,vmware-cpuid-freq=on, -msg timestamp=on 2018-01-15 09:11:09.941+0000: Domain id=2 is tainted: high-privileges 2018-01-15 09:11:09.941+0000: Domain id=2 is tainted: custom-argv 2018-01-15 09:11:09.941+0000: Domain id=2 is tainted: host-cpu 2018-01-15T09:11:09.987833Z qemu-system-x86_64: -chardev pty,id=charserial0: char device redirected to /dev/pts/0 (label charserial0) 2018-01-15T09:11:11.798850Z qemu-system-x86_64: -device vfio-pci,host=02:00.0,id=hostdev0,bus=pci.4,addr=0x0,romfile=/mnt/disks/VMSSD/vbiosbound.rom: Failed to mmap 0000:02:00.0 BAR 3. Performance may be slow 2018-01-15T09:12:33.812985Z qemu-system-x86_64: vfio_region_write(0000:02:00.0:region3+0x1088, 0x7ffe11,8) failed: Device or resource busy 2018-01-15T09:12:33.830419Z qemu-system-x86_64: vfio_err_notifier_handler(0000:02:00.1) Unrecoverable error detected. Please collect any data possible and then kill the guest I used to always get the red line about tainted domains, the yellow part about the ROM is new and stopping VM from booting. part where I passthrough (video and sound PCI devices from the graphics card) :- <hostdev mode='subsystem' type='pci' managed='yes'> <driver name='vfio'/> <source> <address domain='0x0000' bus='0x02' slot='0x00' function='0x0'/> </source> <rom file='/mnt/disks/VMSSD/vbiosbound.rom'/> <address type='pci' domain='0x0000' bus='0x04' slot='0x00' function='0x0'/> </hostdev> <hostdev mode='subsystem' type='pci' managed='yes'> <driver name='vfio'/> <source> <address domain='0x0000' bus='0x02' slot='0x00' function='0x1'/> </source> <address type='pci' domain='0x0000' bus='0x05' slot='0x00' function='0x0'/> </hostdev>

Ah sorry I thought you said earlier that OSX was working okay? Your VM is no longer working following the update to 6.4? Mine are on Q35-2.10 now am on unRAID 6.4. I thought that was the version you had to be on in order to use a non-special patch clover?

I've got the smae but ignoring it as there don't appear to be any issues...

But this doesn't allow you to apply your own microcode updates does it? That is what is being asked here. The V3 Xeons can be made to run all their cores at max turbo with a microcode update but you need to be able to apply the update.

Aaah this is bad news I want to do the same

Does it matter that the CPU appears as a Core 2 Duo? Are we missing out on anything because OSX thinks it is that?

Well I have no idea why but when I downloaded it instead of doing it through the app store it worked all of a sudden.