Warrentheo

Members
  • Posts

    311
  • Joined

  • Last visited

Everything posted by Warrentheo

  1. Possibly (probably)... But being a Windows VM I don't think it matters, since UnRaid just hands it to QEMU, and QEMU just hands the whole device to Windows... Linux VM's probably can optimize things in the passthrough, but I don't think Windows does... Bottom line I have been running it with those lines for over a year, no issue...
  2. To be honest, the auto-driver install should work fine, and the fact that it doesn't and that moonlight also has issues shows that you probably don't have a complete passthrough yet... It is normal for the card to work just fine on first boot, but if the passthrough is not working correctly than a reboot of the VM wont be able to correctly reset the PCIe slot for the card, and so it will still be stuck trying to talk to the now rebooted VM... The ROM files that I download from my card manufacturer are also around 251k, but the one I dumped from my GTX1070 is around 126K, and the ones download from the Manufacturer even when they are the ones currently applied to the card using the normal Windows (Baremetal) don't work... There are other places to get the ROM, but the ones from the card directly are best... Yours being a GTX1060 might be smaller than mine, but I would not expect it to be 50% the size... We would need the VM's XML file, and the diagnostics file to be able to help further...
  3. we would need the VM xml file and diagnostics file to be able to help, but most likely you have something mis-configured somewhere... Those specs should be more than enough for Dota 2
  4. I have been using a Windows 10 VM as my daily driver and gaming station for over a year now... Low latency, and no issues with connection stability... We would need to see the XML file from the VM and diagnostics file for the host to be able to help further... That said there is a better then 50% chance the issue is hardware related, not software...
  5. You probably want to figure out the Virt-Manager docker container using the community applications plugin, or just create the VM without it, then manually edit the VM's XML file to include a section similar to the one I posted above... The WebUI is expecting a vDisk image, not a real block device...
  6. To be honest, so do I... There is alot more you can do with it than what I am currently doing... If you happen to have an NVidia card, and want to pass it into the docker container, you might want to read up on this one:
  7. That sort of thing you run from inside the Ubuntu container, getting the container working in the first place is just a couple of "Docker" commands... Unfortunately no one has made a template that makes it link into UnRaid easily, so if you are not familiar with Docker then the VM way is probably best... You will get better performance and have a smaller footprint however if you run outside of the VM environment...
  8. You might want to look into and consider: https://hub.docker.com/_/ubuntu/ It bypasses a lot of the issues with running in a VM, and allows for better performance, but is not perfect for all use cases, so look into it...
  9. I have both my Windows 10 Image file stored on the UnRaid Cache (on my system a 2x 960 Evo in Raid-0 Striped pool), as well as a second image that just works as a scratch pad for temp storage, also backed by the UnRaid Cache... This gets the driver letter E:\ on my system... I also have an UnRaid share for Steam, and map steam to have most games installed on the share by default, about 70% of games run off of a network share just fine, and I use that for most small or older games where load times are negligible... Most everything else, I use the "Move Local Files" function of Steam to move them to the E:\ Steam share (again, on my system this is backed by the UnRaid Cache) speeding loading times up quite a bit... I use a separate image file because then I can do Windows backups by just copying the image file for Windows, and ignore the "Cached scratch drive" for backups...
  10. It is possible, I use the following section in mine to do that very thing: <disk type='file' device='disk'> <driver name='qemu' type='raw'/> <source file='/dev/disk/by-id/ata-Hitachi_HDS5C3030ALA630_MJ1321YNG06KNA'/> <target dev='sdc' bus='scsi'/> </disk> Would need the XML file for your VM to troubleshoot further... Edit: Your libvirt log file shows this error several times, it indicates that you are trying to get to files on the drive through an invalid file path... There is nothing in a block device driver after the device id, so the ".../cache/VMS/Windows 10/vdisk2.img" portion of that line just breaks things...
  11. If it really is a vdisk image file, than the install is just a single file, copy it where you will... Then just edit the VM you have and point it to the new location for the file... Easy... Running: fallocate -d <filenamegoeshere> on the file will "Dig" the zero's out of the image, and make it take both less space and less time to copy, but up to you... And as always, I recommend doing a backup of the file first, because it is hard to go wrong with that one...
  12. Linux is better, nvflash is not needed... Let UnRaid boot the card with the default drivers (not vfio-pci, or the NVidia unraid plugin) then watch this video, it has the commands you need... P.S. I love Space Invader One, his videos are awesome!!
  13. I have a GTX 1070, all I had to do was dump with the command line before I messed with switching UnRaid over to the vfio-pci driver for the card... it dumped just fine, no modification needed...
  14. depends on what else is using the cache, if it is getting thrashed already, then separate drive... I currently don't use much else on it, so I do host mine on the cache drive... But it depends on your setup and preference...
  15. Please also post the diagnostics file from the system...
  16. You can just edit the VM's XML directly, change this section: <clock offset='localtime'> <timer name='hypervclock' present='yes'/> <timer name='hpet' present='no'/> </clock> And just set it to 'no' instead... That said, I have a GTX1070 passed in to a Win10 VM with the ROM file from the card passed in... Hyper-V is working fine for me with Fully updated Windows and GeForce drivers version 419.17... So this sounds like you need to check your system...
  17. Added that to my go file (Reversed the order, because when I tried to type them manually, they said that bluetooth required some of the sub modules, so I tried to unload the sub modules first) : ...... rmmod btusb rmmod btrtl rmmod btintel rmmod btbcm rmmod bluetooth # Start the Management Utility /usr/local/sbin/emhttp & and it causes these lines right after the winbindd -D part of the boot process: rmmod: ERROR: Module btusb is not currently loaded rmmod: ERROR: Module btrtl is not currently loaded rmmod: ERROR: Module btintel is not currently loaded rmmod: ERROR: Module btbcm is not currently loaded /var/tmp/go: line 15: $'r\357\273\277mmod': command not found Edited the go file with nano from the web cli, so I don't think it is a Windows text file issue... According to the internet, I added this to my syslinux config as my next attempt: ... bluetooth.blacklist=yes ... That showed up as this error in my syslog: bluetooth: unknown parameter 'blacklist' ignored Can't seem to find a way to kill this for now... Not a huge issue for me for now, but currently will require me to rollback to 6.6.7 to be able to use Bluetooth in my VM for now (nothing I seem to find will let me attach it without it being (Code 10) for now...)
  18. I actually do pass through a 5-port USB card to the VM as well, but every port is in use... My motherboard has a lot of ports, but none of them can be passed through for various reasons... I used the passthrough method for less important devices that were always attached so that I could use some of the motherboard ports... One question, does this mean that using a USB Bluetooth in docker versus vm's are mutually exclusive? Are the command line commands to switch between them? And bottom line, how much of this can be automated in WebUI? For now, what do I add to my syslinux to kill this driver for now?
  19. Changed Priority to Minor (It only locks up when I attempt to attach the device)
  20. I have a USB bluetooth key (BCM20702A0 Bluetooth 4.0) passed into my VM by USB ID... Under 6.6.6 and prior, this worked fine, but starting in 6.7.0-rc3 through 6.7.0-rc5 this now seems to get a driver from the host kernel, which causes aberrant behavior on both the host and VM... Symptoms include: Full system reboot then boot into Win10 VM with device passed through by ID: Device manager boots with the USB device listed, but now shows "This device can not start (Code 10), profiling not started" First time I got this, I attempted to use WebUI to detach the device, this acted like it worked, but Windows now showed the device attached and working(???)... Re-attaching the device Win10 went back to "(Code 10)"... Detaching the device a second time removed the device from Windows entirely, but when Re-attaching a second time, the VM locked up for 2 minutes... Performed a second full reboot and boot into Win10 VM has different behavior: Device again shows as (Code 10) in Windows, but the host WebUI no longer shows either the USB adapter, and now seems to have removed the other USB device I was passing through (a Synaptics USB Finger Print Reader), and the only device now showing is a pair of headphones hooked to the PC to charge, even though the Windows VM does show the device attached but in (Code 10)... I assume to fix this I would need to go down the road of disabling the USB driver for this device in the Syslinux config, but do we really want to go down the road of having devices like these having to have special rules that have to be remembered? Perhaps we can add the driver, but disable it by default? Or is there a driver like VFIO-PCI for usb devices? I am willing to test solutions since I rarely use Bluetooth (Only noticed this problem when I went into device manager for other reasons), and will be remaining on 6.7.0-rc5 for now... qw-diagnostics-20190223-1603.zip lsusb -v for BCM20702A0 Bluetooth 4.0.txt Win10VM.XML
  21. In one sense, it doesn't really matter if all you use it for is for the host to notify of system shutdown/reboot/hibernate... The 32-bit should work fine for that... That said, the 64-bit version not installing points to some other issue that should not be happening, and might be a symptom of something else that might bite you in the butt later... This would bother the hell out of me 😛 But I suppose it is up to you 😉 Edit: Silly question, but are you sure you installed the 64-bit version of windows?
  22. Mine also is: <memballoon model='none'/> And NVidia GTX1070 passthrough is working fine for me for months now... Basically the Balloon devices should be ignored for Windows, it just is not designed for it to work that well without major headaches on the part of the user... Edit: and the Sub Answer to that is that "Min Memory" and "Max Memory" should always be the same on Windows VM's...
  23. There are some possibilities, but most of them have to do with preventing you from booting, not crashing... You might want to download the latest VIRTIO drivers: https://fedorapeople.org/groups/virt/virtio-win/direct-downloads/latest-virtio/virtio-win.iso and then manually reinstall each-one one at a time including the guest agent... That is my best guess... Anything other than that would be host issues (Hardware), or mis-configured VM XML files, or broken install of Windows... What I do with my system is install Windows on one image, then add another for all my data that needs to be "Local" to the VM, then everything else is on the UnRaid SMB... This lets me back them up separately, and therefor restore them separately, preventing issues like this...
  24. I believe it also has to do with the "Balloon" driver stuff that no one uses on Windows because it is buggy and doesn't really work... this shouldn't really help, but there is a new version of this, you can get it here: https://fedorapeople.org/groups/virt/virtio-win/direct-downloads/latest-virtio/virtio-win.iso Also, need the XML files from the VM in question... Your libvirt logs has some things I have never seen that seems to pertain... Some of this is normal (at least they show up on mine), but there is also stuff about Qemu monitors and Guest agents being unavailable... This seems like something to do with the VM XML file... ...... 2019-01-23 10:18:40.187+0000: 7155: error : qemuMonitorIO:718 : internal error: End of file from qemu monitor 2019-01-23 10:21:01.170+0000: 7158: error : virDomainDefRejectDuplicateControllers:3966 : XML error: Multiple 'scsi' controllers with index '0' 2019-01-23 10:21:19.120+0000: 7158: error : virDomainDefRejectDuplicateControllers:3966 : XML error: Multiple 'scsi' controllers with index '0' 2019-01-23 10:26:18.213+0000: 7160: warning : qemuDomainObjTaint:7831 : Domain id=3 name='Handyman' uuid=913f649f-9692-6e90-629e-94f6e067c0c0 is tainted: high-privileges 2019-01-23 10:26:18.213+0000: 7160: warning : qemuDomainObjTaint:7831 : Domain id=3 name='Handyman' uuid=913f649f-9692-6e90-629e-94f6e067c0c0 is tainted: host-cpu 2019-01-23 10:30:15.849+0000: 7157: warning : qemuDomainObjTaint:7831 : Domain id=4 name='Handyman' uuid=913f649f-9692-6e90-629e-94f6e067c0c0 is tainted: high-privileges 2019-01-23 10:30:15.849+0000: 7157: warning : qemuDomainObjTaint:7831 : Domain id=4 name='Handyman' uuid=913f649f-9692-6e90-629e-94f6e067c0c0 is tainted: host-cpu 2019-01-23 10:31:50.426+0000: 7159: error : qemuAgentSend:866 : Guest agent is not responding: Guest agent not available for now ......
  25. Yah, IP is best, but this is a bit of a known issue for Windows Lan Manager... If you don't want to read up more than you ever wanted to know about Windows networking, then try this out: Or just paste this in the install page on the host plugin: https://raw.github.com/bergware/dynamix/master/unRAIDv6/dynamix.local.master.plg This will make the host the local master, and remove most name issues with networking...