May 14, 201412 yr Last Update 05/30/2014 System "Name" - tower2 Xen capable? - Yes. PCI Passthrough capable? - Yes. Case - fractal design define R3 CPU - Intel i5 2500 - vt-d capable Motherboard- Asus P8H77-V LE GPU-XFX AMD HD 7750 Single Slot FX-775A-ZNP4 (8.1 Passthrough) PCI Rosewill NEC 4+1 internal USB RC-101 Chipset NEC 720101 (8.1 Passthrough) PCI Syba SD-SATA-4P Flashed to b5500 bios (IDE) Sata 4 port for ODD Chipset Sil3114 (8.1 Passthrough, Special Note Below) - using PCIe 1 StarTech Sata 2 port for ODD Chipset ASM1062 (8.1 Passthrough, Special Note Below) - tested and works, not using RAM- CORSAIR Vengeance LP 16 GB DDR3 PSU - SeaSonic X Series X650 Gold HBA / SAS Card - Supermicro AOC-SASLP-MV8 flashed to Firmware_3.1.0.21 NIC(s) Realtek 8169 based on-board NIC (bridge for all VMs) Hard drives 1x 4 TB HGST RAID NAS drive - On MOBO 2x 3 TB MISC - On SASLP 2x 2 TB MISC - On SASLP 1x 500 GB SSD EVO Cache (VM Folder) - On MOBO Brief description of VMs and use cases Dom0 - unRAID DomU1 - 2vcpus / 2gb ram - Windows 8.1 main desktop / gaming VM / DVD ISO Creation / GPU Passthrough DomU2 - 2vcpus / 2gb ram - UBUNTU 14.04 LTS with lite desktop / crashplan / machine maint services DomU3 - 2vcpus / 2gb ram - UBUNTU 12 LTS with full desktop / onling banking / play tests. Only started on request DomU4 - Not in use syslinux.cfg default /syslinux/menu.c32 menu title Lime Technology prompt 0 timeout 50 label unRAID OS kernel /bzimage append initrd=/bzroot label unRAID OS Safe Mode (no plugins) kernel /bzimage append initrd=/bzroot unraidsafemode label Memtest86+ kernel /memtest label Xen/unRAID OS menu default kernel /syslinux/mboot.c32 append /xen dom0_mem=2097152 --- /bzimage xen-pciback.hide=(01:00.0)(01:00.1)(05:00.0)(05:00.1)(05:00.2)(06:00.0)(05:02.0) --- /bzroot label Xen/unRAID OS Safe Mode (no plugins) kernel /syslinux/mboot.c32 append /xen dom0_mem=2097152 --- /bzimage --- /bzroot unraidsafemode windows.cfg ironicbadgers guide: http://lime-technology.com/forum/index.php?topic=31754.0 # Disable Passthrough for Driver updates and any Windows update requiring a restart. # Use TightVNC Java Viewer with RAW encoding to Update (port: 5901). IP check in unraid terminal: ifconfig. name = 'windows' builder = 'hvm' vcpus = '2' #can set this to 1-4 safely depending on your CPU/# of VMs. memory = '2048' #DO NOT EXCEED UNLESS BRAVE IF DOING GPU PASS THROUGH #memory = '8192' #----For Disc Installation and any Updates (VNC into Windows) #disk = ['phy:/mnt/cache/VM/windows/windows.img,hda,w','file:/mnt/cache/VM/windows/Windows81_DVD.iso,hdc:cdrom,r'] #boot = 'dc' #usbdevice='tablet' #----After Installation and for GPU Passthrough #disk = ['phy:/mnt/cache/VM/windows/windows.img,hda,w'] disk = ['phy:/mnt/cache/VM/windows/windows.img,hda,w','phy:/dev/sdd,hdb,w'] boot = 'c' usbdevice = ['tablet','host:413c:3016','host:413c:2005','host:05e3:0719'] #LOOK UP VIA lsusb COMMAND IN CONSOLE pci = [ '01:00.0','01:00.1','05:00.0','05:00.1','05:00.2' ] #LOOK UP VIA lspci COMMAND IN CONSOLE #Prior to Start VIA xl pci-assignable-list COMMAND IN CONSOLE to show list of excluded from unRaid PCI devices. #Note, after starting VM, Putty into Tower2 and issue "xl pci-attach windows 0000:05:02.0" to mount the Sata Card / Optical Disk Drive #---Other Settings xen_platform_pci='1' #UNCOMMENT POST INSTALL OF XEN PARAVIRTUALIZATION DRIVERS on_poweroff="destroy" on_reboot="restart" on_crash="destroy" vif = ['mac=00:16:3E:51:20:4C,bridge=br0,model=e1000'] #network bridge device_model_version="qemu-xen-traditional" acpi = '1' apic = '1' sdl = '0' stdvga = '0' viridian = '1' usb=1 #VNC SETTINGS vnc = '1' vnclisten = '0.0.0.0' #vncpasswd = '' vncdisplay = 1 Windows Post Install (and Install) Notes: Using VNC to remote access: Windows Installed Various .NET frameworks installed: 3.5 SP1, 4.0, and 4.5.1 : http://msdn.microsoft.com/en-us/vstudio/aa496123 Numerous windows updates with with VM shutdown within windows followed by full machine powerdown. Repeat until no more updates All drivers updated (AMD Catalyst Control Center NOT installed) fast boot turned off: http://www.eightforums.com/tutorials/6320-fast-startup-turn-off-windows-8-a.html GPLPV Drivers installed: http://wiki.univention.de/index.php?title=Installing-signed-GPLPV-drivers Command Prompt Opened as Admin:cd C:\Program Files (x86)\XEN PV Drivers\bin [enter] shutdownmon.exe -i [enter] [*]passthrough setting enabled in windows.cfg [*]Ironic Badgers GPU Eject script used: http://blog.ktz.me/?p=219 Windows Optical Burner Passthrough (Original) http://lime-technology.com/forum/index.php?topic=32344.0 SATA Card passthrough in Windows.cfg appears to be broken After booting the Windows 8.1 VM (After installing the Sata card driver in windows (ASM1062) the first time, then reboot) In Unraid Putty/Terminal I issue: xl pci-attach windows 0000:06:00.0 This allows normal Optical disk use IF I place ['06:00.0'] into my Windows.cfg PCI = ['06:00.0'] All that happens is that I am shown a similar boot lock screen showing "No bootable device" I have confirmed this with 3 different Sata cards / chipsets. Windows Optical Burner Passthrough Automation: http://lime-technology.com/forum/index.php?topic=33563.0 ubuntu14.cfg tomf's Guide: http://lime-technology.com/forum/index.php?topic=32733.0 name = "ubuntu14" bootloader = "pygrub" #kernel = "/mnt/cache/VM/ubuntu14/vmlinuz" #ramdisk = "/mnt/cache/VM/ubuntu14/initrd.gz" memory = 2048 vcpus = '2' disk = ['phy:/mnt/cache/VM/ubuntu14/ubuntu14.img,xvda,w',] vif = [ 'mac=00:16:3e:aa:bb:cc,bridge=br0' ] #extra = "debian-installer/exit/always_halt=true -- console=hvc0" ubuntu12.cfg name = "ubuntu12" bootloader = "pygrub" #kernel = "/mnt/cache/VM/ubuntu12/vmlinuz" #ramdisk = "/mnt/cache/VM/ubuntu12/initrd.gz" memory = 2048 vcpus = '2' disk = ['phy:/mnt/cache/VM/ubuntu12/ubuntu12.img,xvda,w',] vif = [ 'mac=00:16:3e:aa:bb:cc,bridge=br0' ] #extra = "debian-installer/exit/always_halt=true -- console=hvc0" Ubuntu LTS 12.04 has the ability to display RDP so long as 1- basic ubuntu server is installed 2 - ubuntu desktop is installed 3 - sudo apt-get update 4 - sudo apt-get install xrdp Ubuntu LTS 14.04 has these RDP and 2D desktop components REMOVED, but they may be added back. I was unable to get VNC working on 14.04 1- basic ubuntu server is installed 2 - ubuntu desktop is installed 3 - sudo apt-get install xrdp 4 - sudo apt-get install xubuntu-desktop 5 - echo "xfce4-session" > ~/.xsession 6 - sudo /etc/init.d/xrdp restart This take a little longer for the screen to come up - and it is not as pretty as the Ubuntu desktop, but it WORKS. Unraid Packages Powerdown script - needed APC UPS script - needed Email notification - needed Physical Note 1 Front panel USB port attached to NEC PCI card, 1 Front panel USB port attached to Shared Internal MOBO port with Unraid Flash card. Adapter Cable, USB 2.0, IDC 5 Male (single row) to USB A Male (Used to connect devices designed to plug into USB motherboard header pins to an external USB 2.0 connector) http://www.amazon.com/gp/product/B000V6WD8A/ref=oh_details_o00_s00_i00?ie=UTF8&psc=1 StarTech.com 2 Port USB Motherboard Header Adapter (USBMBADAPT2) http://www.amazon.com/gp/product/B002GNU2V6/ref=oh_details_o08_s00_i01?ie=UTF8&psc=1 Slim DVD Optical: Panasonic UJ-265 UJ265 6X 3D Blu-ray Burner StarTech.com Slimline SATA to SATA Adapter with Power(SLSATAADAP) Single 5.25 slot allows Preclear hot-swap & Bluray burner iStarUSA T-5K35T-SA 5.25" to Slim ODD and 3.5" SATA 6Gb/s
May 15, 201412 yr Good compilation. The Windows notes may be the most complete single list of post install tasks I've seen on the forums. Those need to be typed up in a nice guide format and tacked onto Ironic's Windows 8.1 VM guide.
May 15, 201412 yr Author Thank you both. The windows section took a TON of trial and error, and a great deal of time scouring these forums and others. Now following it I have a STABLE USABLE version of 8.1!
May 15, 201412 yr Any luck increasing the RAM on the 8.1 VM after you got everything stable? Sent from my Nexus 4 using Tapatalk
May 15, 201412 yr Author Any luck increasing the RAM on the 8.1 VM after you got everything stable? Boot times appear normal with 4 vcpus but not 2 when 8 GB of ram is used. Shutdown scripts and GPU ejects also have worked (twice so far). I have set it back to 2 vcpus and 2 GB of ram until I have a block of time to more thoroughly test stability and performance again. I HOPE that it stays stable during intensive GPU use like this ---> I suspect that the RAM on the GPU may impact the stability too - and that greater GPU RAM allows greater System Ram. I currently have a 7750 single slot, 1 GB card. If Steam hosting for streaming with Splashstop/Kainy to my android phone works I will likely swaps thsi out for a HD 7970.
May 16, 201412 yr Author Any luck increasing the RAM on the 8.1 VM after you got everything stable? Boot times appear normal with 4 vcpus but not 2 when 8 GB of ram is used. Shutdown scripts and GPU ejects also have worked (twice so far). I had a boot this way last night that took 40+ min (normally takes about 10 seconds). Restarted VM and it appeared to work ok Started up Skyrim and the system ground to a halt Restarted with 2 & 2 - Skyrim was fine. Lower memory to 3000 or less. Video pass does NOT work with more than that amount (approximately). This is a known issue and we hope to get around this in the future.
May 18, 201412 yr Question about running a windows gaming VM. Can I passthrough my cache drive and still have it available to other VM & unRAID or does it get allocated an locked? Same applies to USB as well. Reason being my VM are on my cache drive, I'd need a USB port for a game controller and naturally I am aware that the GPU WOULD be allocated to that VM solely. Hope I am clear. Kryspy
May 18, 201412 yr With the windows VM you would create a windows.img and let's say you told it to be 30 GB, then you install windows on the .img file. When you turn on the windows VM it will using that 30 GB .img file as the C: drive. You could then map a network drive back to your unRAID array and edit/view the cache drive like you are doing now. You have to passthrough any devices to the windows VM that you want it to use and once they are passed through only the windows VM has access to that device. So in other words unRAID will not be able to use the same USB as the windows VM.
May 18, 201412 yr Ok got it. Cache drive would be like any other shared network folder and the USB port I use and GPU I use will be tied and locked to the Windows VM. Kryspy
May 18, 201412 yr Thanks for the thorough post, and great notes. I'm specifically interested in the Windows install, but for Windows 7 in my case. I've modified my windows.cfg script to more closely match yours, and my windows VM still runs fine (which is good!) In reviewing your Windows post install notes, I'm unsure about this part... Command Prompt Opened as Admin: cd C:\Program Files (x86)\XEN PV Drivers\bin [enter] shutdownmon.exe -i [enter] What is the purpose of this? I've installed all the Windows updates, and rebooted several times, all from within the Windows VM (while using TightVNC to access the VM), and it has rebooted fine every time. Do I need to do this, and if so, what for? Also, any thoughts on passing thru the whole PCI bridge vs. passing thru a single PCI device? How about USB, same question, better to passthru the whole hub, or just individual devices? I've re-ordered the results of lsusb, to organize by bus then Device. My results are below... Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 001 Device 002: ID 8087:8008 Intel Corp. Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 002 Device 002: ID 8087:8000 Intel Corp. Bus 002 Device 003: ID 13fe:3d00 Kingston Technology Company Inc. Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 003 Device 002: ID 174c:2074 ASMedia Technology Inc. Bus 003 Device 003: ID 045e:00dd Microsoft Corp. Comfort Curve Keyboard 2000 V1.0 Bus 003 Device 004: ID 051d:0002 American Power Conversion Uninterruptible Power Supply Bus 003 Device 005: ID 0a12:0001 Cambridge Silicon Radio, Ltd Bluetooth Dongle (HCI mode) Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub Bus 004 Device 002: ID 174c:3074 ASMedia Technology Inc. I expect I need to do a bunch of trail and error to figure out which things I need to passthru. I've read a lot about this on these forums already, but it's all a bit of a hodge podge I think I also need to exclude these devices from unRAID, but I'm not sure if I have to exclude everything or not. More trial and error I suspect Anyway, thanks for what you have provided, it's been a great help.
May 19, 201412 yr In reviewing your Windows post install notes, I'm unsure about this part... Command Prompt Opened as Admin: cd C:\Program Files (x86)\XEN PV Drivers\bin [enter] shutdownmon.exe -i [enter] What is the purpose of this? I've installed all the Windows updates, and rebooted several times, all from within the Windows VM (while using TightVNC to access the VM), and it has rebooted fine every time. Do I need to do this, and if so, what for? While it's entirely possible to shutdown / restart your VM from inside the VM (via VNC), you can also manage VMs from the Xen / unRAID command line as follows; xl shutdown VMName For this to work for windows VMs, you need the gplpv drivers installed in the VM. The above lines configure a service inside the VM to listen for these commands and act on them. You might think you can happily do this from inside the VM so what's the point? Sometimes it's convenient to shutdown several VMs from one spot (the command line) or perhaps to use suspend or shutdown commands from inside a script (such as the VM backup script currently being discussed on the forums). Also, any thoughts on passing thru the whole PCI bridge vs. passing thru a single PCI device? How about USB, same question, better to passthru the whole hub, or just individual devices? It seems this is hardware dependent. With my setup, I've had great success passing through discrete PCI devices including GPUs, TV Tuner Cards and Sound Cards. (not so much on-board or add-in firewire). However, with USB, I've had much better success passing through discrete USB devices rather than entire controllers. Though from reading here, some people have had the opposite experience. (I have quite a bit on USB passthrough on my blog, linked in footer). I expect I need to do a bunch of trail and error to figure out which things I need to passthru. I've read a lot about this on these forums already, but it's all a bit of a hodge podge I think I also need to exclude these devices from unRAID, but I'm not sure if I have to exclude everything or not. You only need to exclude PCI devices and/or controllers that you are planning on passing through. This is done via the pciback.hide command in the syslinux.cfg file. If you're passing through discrete USB devices, you don't need to exclude those. You do this with a command similar to the following in your VM config; usbdevice = ['host:abcd:1234' , 'host:qwer:9876'] In terms of trial & error, if trying to identify discrete devices, issue lsusb, unplug the device you wish to passthrough and do another lsusb. It's pretty easy to spot the missing device when comparing the two lists. Peter
May 19, 201412 yr Author Cache drive would be like any other shared network folder and the USB port I use and GPU I use will be tied and locked to the Windows VM. Confirmed true. Your USB port and GPU assumes that the devices have been excluded from unRaid in your syslinux.cfg and passed through in your <vm name>.cfg Thanks for taking the time to post this. Thanks for the thorough post, and great notes You are both very welcome I'm specifically interested in the Windows install, but for Windows 7 in my case. Not with passthrough I hope? I was unsuccessful in getting 7 to work. I believe Iconicbadger has also had issues with 7. In reviewing your Windows post install notes, I'm unsure about this part... Command Prompt Opened as Admin: cd C:\Program Files (x86)\XEN PV Drivers\bin [enter] shutdownmon.exe -i [enter] What is the purpose of this? I've installed all the Windows updates, and rebooted several times, all from within the Windows VM (while using TightVNC to access the VM), and it has rebooted fine every time. The XEN/GPLPV drivers do, to my knowledge, 3 things: 1 – emulate baremetal hardware (i.e. act as generic device drivers) 2 – give you a low latency 10 gbps internal network between your VMs 3 – provide the ability for your Windows OS to talk to the XEN host (in this case unRaid). By installing shutdownmon as a service unRaid can talk BACK to your Windows OS. This means that not only can the VM be shutdown within windows, but via an unRaid console (xl shutdown <vm name>) or from the unRaid Web Gui (extensions, XEN, <vm name>, Shutdown). Though I have yet to get the following integrated, I believe it will be a critical component of the powerdown scripts and APCUPS scripts ... I almost sound like I know what I am talking about, Luddite though I be.
May 19, 201412 yr While it's entirely possible to shutdown / restart your VM from inside the VM (via VNC), you can also manage VMs from the Xen / unRAID command line... For this to work for windows VMs, you need the gplpv drivers installed in the VM. The above lines configure a service inside the VM to listen for these commands and act on them. You might think you can happily do this from inside the VM so what's the point? Sometimes it's convenient to shutdown several VMs from one spot (the command line) or perhaps to use suspend or shutdown commands from inside a script (such as the VM backup script currently being discussed on the forums). Okay, that makes sense. I had already installed the GPLPV drivers (and have confirmed they are still installed), but when I ran (as Administrator) the command: shutdownmon.exe -i I got an error {CreateService Failed (1072)} I tried to run shutdownmon.exe -u to uninstall it, and that seemed to go fine, but running shutdownmon.exe -i again resulted in the same error message. I'll have to check the /bin folder to see if the exe is actually in there, but otherwise, I'm stuck on that one. I rebooted unRAID since then, but have not tried this again yet, due to my testing USB slots (see below). Also, any thoughts on passing thru the whole PCI bridge vs. passing thru a single PCI device? How about USB, same question, better to passthru the whole hub, or just individual devices? It seems this is hardware dependent. With my setup, I've had great success passing through discrete PCI devices including GPUs, TV Tuner Cards and Sound Cards. (not so much on-board or add-in firewire). However, with USB, I've had much better success passing through discrete USB devices rather than entire controllers. Though from reading here, some people have had the opposite experience. (I have quite a bit on USB passthrough on my blog, linked in footer). It looks like you've had good luck passing thru discrete devices, whether PCI or USB, so that's where I'll start too. When I finish testing/mapping the USB slots, I'll start testing passthru of devices. I'll check out your blog before I continue; thanks. In terms of trial & error, if trying to identify discrete devices, issue lsusb, unplug the device you wish to passthrough and do another lsusb. I decided to just plug in an external USB drive to all the empty USB slots, one at a time, and do lsusb after each change, and started tracking which bus/device I'd just plugged into. When I got to one of my slots, I realize that my Bluetooth dongle was in a USB 3 slot, and in the lower slot, which meant I was wasting a space and it was hard to remove, so I moved it to another spot, and plugged the drive into that slot I removed it from, as a quick test. When I ran lsusb, it registered a different bus/device for the slot that had the BT, but now has the external drive. How is that possible? I wouldn't expect the bus and device numbers to be different based on what kind of device is connected, but it seems that's what's happening. Does that sound right? (the BT is USB 1.0, and the drive is USB 3.0; if it matters). {Sorry for the thread hijack - I'll start a new thread if further discussion is needed}
May 19, 201412 yr Author Okay, that makes sense. I had already installed the GPLPV drivers (and have confirmed they are still installed), but when I ran (as Administrator) the command: shutdownmon.exe -i I got an error {CreateService Failed (1072)} I tried to run shutdownmon.exe -u to uninstall it, and that seemed to go fine, but running shutdownmon.exe -i again resulted in the same error message. I'll have to check the /bin folder to see if the exe is actually in there, but otherwise, I'm stuck on that one. You need to run the CMD prompt as Administrator. In the Windows Search bar Type CMD, Right Click on CMD.EXE, Select Run As Administrator. Ensure that shutdownmon.exe exists in the file location referenced before changing the directory in the CMD prompt via CD prior to doing this. lsusb When I got to one of my slots, I realize that my Bluetooth dongle was in a USB 3 slot, and in the lower slot, which meant I was wasting a space and it was hard to remove, so I moved it to another spot, and plugged the drive into that slot I removed it from, as a quick test. When I ran lsusb, it registered a different bus/device for the slot that had the BT, but now has the external drive. How is that possible? I wouldn't expect the bus and device numbers to be different based on what kind of device is connected, but it seems that's what's happening. Does that sound right? (the BT is USB 1.0, and the drive is USB 3.0; if it matters). I am having a little trouble following you here. I believe you only need the ID, such as 413c:2005 (see below) for passing through a single device – it will retain this ID in different USB ports. Also note, I have found that 2 of the same devices can result in a repeated ID. What do you mean by ‘wasting space’ and ‘hard to remove’? Do you physically restart the machine between plugging USB devices in and out? Bus 001 Device 002: ID 8087:0024 Intel Corp. Integrated Rate Matching Hub Bus 002 Device 002: ID 8087:0024 Intel Corp. Integrated Rate Matching Hub Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub Bus 001 Device 003: ID 0781:5581 SanDisk Corp. Bus 002 Device 003: ID 413c:3016 Dell Computer Corp. Optical 5-Button Wheel Mouse Bus 002 Device 004: ID 413c:2005 Dell Computer Corp. RT7D50 Keyboard
May 19, 201412 yr Okay, that makes sense. I had already installed the GPLPV drivers (and have confirmed they are still installed), but when I ran (as Administrator) the command: shutdownmon.exe -i I got an error {CreateService Failed (1072)} I tried to run shutdownmon.exe -u to uninstall it, and that seemed to go fine, but running shutdownmon.exe -i again resulted in the same error message. I'll have to check the /bin folder to see if the exe is actually in there, but otherwise, I'm stuck on that one. You need to run the CMD prompt as Administrator. In the Windows Search bar Type CMD, Right Click on CMD.EXE, Select Run As Administrator. Ensure that shutdownmon.exe exists in the file location referenced before changing the directory in the CMD prompt via CD I did run as administrator. I have yet to check if the file is in the /bin folder, but will confirm that next. When I got to one of my slots, I realize that my Bluetooth dongle was in a USB 3 slot, and in the lower slot, which meant I was wasting a space and it was hard to remove, so I moved it to another spot, and plugged the drive into that slot I removed it from, as a quick test. When I ran lsusb, it registered a different bus/device for the slot that had the BT, but now has the external drive. How is that possible? I wouldn't expect the bus and device numbers to be different based on what kind of device is connected, but it seems that's what's happening. Does that sound right? (the BT is USB 1.0, and the drive is USB 3.0; if it matters). I am having a little trouble following you here. I believe you only need the ID, such as 413c:2005 (see below) for passing through a single device – it will retain this ID in different USB ports. Also note, I have found that 2 of the same devices can result in a repeated ID. What do you mean by ‘wasting space’ and ‘hard to remove’? Do you physically restart the machine between plugging USB devices in and out? Bus 001 Device 002: ID 8087:0024 Intel Corp. Integrated Rate Matching Hub Bus 002 Device 002: ID 8087:0024 Intel Corp. Integrated Rate Matching Hub Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub Bus 001 Device 003: ID 0781:5581 SanDisk Corp. Bus 002 Device 003: ID 413c:3016 Dell Computer Corp. Optical 5-Button Wheel Mouse Bus 002 Device 004: ID 413c:2005 Dell Computer Corp. RT7D50 Keyboard Sorry, I meant that having a USB1 dongle in a USB3 slot was wasting space (that could be used fora USB3 device instead) As for being "hard to remove", I meant that the dongle is small and rounded and being in the bottom slot meant I couldn't get my fingers to grip the curved dongle, so it's too hard to remove unless I have space below it, so I could grab it on the top and bottom. It's mostly irrelevant, but that's what I meant I'll try using the device ID instead of the bus and device ID to pass thru the individual devices. I was hoping to be able to passthru a whole USB bus, so that I could know which USB slots I could plug into and know that the attached device would pass thru to the VM, but I guess this might not be possible. I started a new thread here to discuss... http://lime-technology.com/forum/index.php?topic=33465.0
May 20, 201412 yr I've rebooted now, and successfully passed thru several devices to my Windows VM Ensure that shutdownmon.exe exists in the file location referenced before changing the directory in the CMD prompt via CD prior to doing this. I did verify that shutdownmon.exe does exist in the /bin folder, and this time, when I ran shutdownmon.exe -i it said it successfully installed. Yay However, when I tried to shutdown the VM from the unRAID GUI, it was unsuccessful. I don't know if that's because it was not installed/running when I created the VM, or some other problem. I'll investigate more later. For now, I've gotta run, but will be back to this later today, this evening. Thanks again!
May 20, 201412 yr However, when I tried to shutdown the VM from the unRAID GUI, it was unsuccessful. I don't know if that's because it was not installed/running when I created the VM, or some other problem. I'll investigate more later. For now, I've gotta run, but will be back to this later today, this evening. I took 3 minutes to rebuild the VM, then tried shutting it down wit the GUI, and this time it worked!! Now that's progress! thanks again
May 20, 201412 yr Author I did run as administrator. I have yet to check if the file is in the /bin folder, but will confirm that next. The default locations are: C:\Program Files (x86)\XEN PV Drivers\bin C:\Program Files\XEN PV Drivers\bin You may have installed in a NON default location too. Sorry, I meant that having a USB1 dongle in a USB3 slot was wasting space (that could be used fora USB3 device instead) As for being "hard to remove", I meant that the dongle is small and rounded and being in the bottom slot meant I couldn't get my fingers to grip the curved dongle, so it's too hard to remove unless I have space below it, so I could grab it on the top and bottom. It's mostly irrelevant, but that's what I meant Understood I've rebooted now, and successfully passed thru several devices to my Windows VM Fantástico! I did verify that shutdownmon.exe does exist in the /bin folder, and this time, when I ran shutdownmon.exe -i it said it successfully installed. However, when I tried to shutdown the VM from the unRAID GUI, it was unsuccessful. Have you shutdown the VM, then powerdown the unRaid machine? This has solved SO many setup issues for me Note the distinction between powerdown and restart I took 3 minutes to rebuild the VM, then tried shutting it down wit the GUI, and this time it worked!! What do you mean by this? Sorry to hammer away – I just want the notes in place should my first post require updates – and so that others walking up the learning curve need not trip over the same hurdles.
May 20, 201412 yr I did verify that shutdownmon.exe does exist in the /bin folder, and this time, when I ran shutdownmon.exe -i it said it successfully installed. However, when I tried to shutdown the VM from the unRAID GUI, it was unsuccessful. Have you shutdown the VM, then powerdown the unRaid machine? This has solved SO many setup issues for me Note the distinction between powerdown and restart I powered down unRAID last night (we lost power), and had started it this morning, before I ran shutdownmon.exe -i, but not after. The first time I tried to shutdown the windows VM with the unRAID GUI was right after installing shutdownmon. I did not reboot anything. It didn't work. So, I shutdown windows from within windows (start orb, shutdown). I took 3 minutes to rebuild the VM, then tried shutting it down wit the GUI, and this time it worked!! What do you mean by this? Sorry, I was unclear again, my bad I meant I restarted the Windows VM (not rebuilt). I did not actually open the VM with tightVNC, I just waited until unRAID GUI showed that it was running, and that it was using 2 cores (it only uses one while it's starting up). Once I saw it was running (after about 3 minutes), I just hit the shutdown button in the unRAID GUI, and it did shutdown windows, just like it should. Sorry to hammer away – I just want the notes in place should my first post require updates – and so that others walking up the learning curve need not trip over the same hurdles. No need to apologize! I'm putting all my notes in here for exactly that reason, so that others can learn from our efforts. I'm sorry that I'm not being as clear as I should be. At this point, I think the only thing I'm still missing is getting unRAID to recognize my Radeon X600 video card, so that I can pass it thru to windows. Hopefully I can figure that out soon, so I can actually use the VM without tightVNC. Thanks again for your guidance.
May 23, 201412 yr Author Windows Optical Burner Passthrough Update Not yet AUTOMATED mount upon boot http://lime-technology.com/forum/index.php?topic=32344.0 SATA Card passthrough in Windows.cfg appears to be broken After booting the Windows 8.1 VM (After installing the Sata card driver in windows (ASM1062) the first time, then reboot) In Unraid Putty/Terminal I issue: xl pci-attach windows 0000:06:00.0 This allows normal Optical disk use IF I place ['06:00.0'] into my Windows.cfg PCI = ['06:00.0'] All that happens is that I am shown a similar boot lock screen showing "No bootable device" I have confirmed this with 3 different Sata cards / chipsets.
May 23, 201412 yr Windows Optical Burner Passthrough Update Not yet AUTOMATED mount upon boot http://lime-technology.com/forum/index.php?topic=32344.0 SATA Card passthrough in Windows.cfg appears to be broken After booting the Windows 8.1 VM (After installing the Sata card driver in windows (ASM1062) the first time, then reboot) In Unraid Putty/Terminal I issue: xl pci-attach windows 0000:06:00.0 This allows normal Optical disk use IF I place ['06:00.0'] into my Windows.cfg PCI = ['06:00.0'] All that happens is that I am shown a similar boot lock screen showing "No bootable device" I have confirmed this with 3 different Sata cards / chipsets. When I passed through my M1015 to my WHS2011 I had to get into the card bios and turn off the boot functionality of the card. Once I did that my WHS2011 VM booted from my .img file. So if your card has a bios you can access try turning it off. It might just be turning off interrupt 13 for example but might be more involved.
May 24, 201412 yr So, I've installed windows 8 now, since 7 just isn't working for me with gpu passthru. I got the ATI HD5550 card passed thru to the Windows machine, and everything is up to date, and working fine, other than the latest Catalyst drivers. I tried installing the drivers last night, and the machine hung at "Detecting graphics hardware" I've left it all night, and it's still hung. i cannot get it to accept any mouse of keyboard commands, so ctl-alt-del won't work. I tried issuing that from VNC, but still nothing. I cannot find any way to restart or shut down the windows VM. If I click shut down from unRAID GUI, it doesn't work either (I have the shutdownmon installed, and have rebooted everything since installing it). So, now I'm trying to just stop the array, to just reboot unRAID, but I cannot stop the array either. It's stuck for about the last 5 minutes in syncing file systems. Meaning, I'm about to hard-boot unRAID, which I'm never a big fan of. Anyway, I don't want to just run into this same issue with installing the ATI drivers next time, so I'm wondering if there are any 'gotchas' in the GPU driver install I'm not aware of; or if this was just a one-time bad luck issue?
May 24, 201412 yr So, I've installed windows 8 now, since 7 just isn't working for me with gpu passthru. I got the ATI HD5550 card passed thru to the Windows machine, and everything is up to date, and working fine, other than the latest Catalyst drivers. I tried installing the drivers last night, and the machine hung at "Detecting graphics hardware" I've left it all night, and it's still hung. i cannot get it to accept any mouse of keyboard commands, so ctl-alt-del won't work. I tried issuing that from VNC, but still nothing. I cannot find any way to restart or shut down the windows VM. If I click shut down from unRAID GUI, it doesn't work either (I have the shutdownmon installed, and have rebooted everything since installing it). So, now I'm trying to just stop the array, to just reboot unRAID, but I cannot stop the array either. It's stuck for about the last 5 minutes in syncing file systems. Meaning, I'm about to hard-boot unRAID, which I'm never a big fan of. Anyway, I don't want to just run into this same issue with installing the ATI drivers next time, so I'm wondering if there are any 'gotchas' in the GPU driver install I'm not aware of; or if this was just a one-time bad luck issue? If a VM hangs like that, you can issue XL destroy VMName from the unRAID command line. (obviously replace 'VMName' with your own vm name. That's the equivalent of pulling the plug on a physical machine. I've found that if I need to do that on a VM with GPU passthrough and then start it up again, hald the time the video will be scrambled necessitating a full system reboot in any case.
Archived
This topic is now archived and is closed to further replies.