xccrev

Members
  • Posts

    19
  • Joined

  • Last visited

Converted

  • Gender
    Undisclosed

xccrev's Achievements

Newbie

Newbie (1/14)

0

Reputation

  1. Can also confirm this. Sent from my iPad using Tapatalk
  2. Are you passing any other pcie devices through? Like a USB card? I had a very similar experience to yours. Happens on reboot, shutdown; takes down the router and network. Vm and unraid eventually go unresponsive.My USB pcie card wasn't playing nicely when it got the signal to shut down. Took the USB card out it went away. Sent from my iPad using Tapatalk
  3. we are working on it. that's the next step for us too. Just to chime in, I'd also find a docker instance of this to be incredibly useful. Interested in seeing how things develop with this.
  4. Tried in a Windows 8.1 VM today. Stability is doing ok now with my changed settings, so I'm pleased about that, but still having trouble passing the card after a VM reboot. Have to hard reset the system to get it through. Could be just my card/mobo combo if no one else is seeing similar problems. Annoying, but I can put up with it.
  5. Hmm. Tried OVMF without much success. Didn't get system hangs really, but also not much passthrough success either Ok, so as of now, I have replaced the card with a different board (same controller), but I'm still seeing similar (if not identical) issues. Still getting a couple of weird system hangs, but they're much less frequent. I've been able to improve things by adjusting a couple things, but I'm not sure what exactly is helping or hurting, as I've sort of just been shooting from the hip. I've turned acs override on. Usb pci card is stubed in syslinux. Also I turned off Windows fastboot feature, which I noticed had been turned on. I definitely saw improvements after turning ACS override on. Not sure about the other two. Seems inconsistent, so hard to test. I'm suspecting that something about the new build of windows 10 1511 is perhaps causing my usb card to enter a bad power state when I shut down the VM. They've made some additions with regards to power management, and thats the only thing I can think of that could be going so terribly wrong. I may try in a Windows 7 VM later... I can get the card to work properly in Seabios if I start the VM from a fresh system boot (on Unraid), but after I shut down the VM the first time, if I restart it, I can't get the card to pass in anymore. If anyone has any brilliant ideas let me know.
  6. Good to know. I'm curious, are you using Seabios or OVMF? I've ordered the 2 port version of the card, so I'm likely going to replace this one with that. I'll report back if I continue to see issue.
  7. After some further testing, it looks like stubbing the card didn't fix the issue, like I thought. So, I went ahead and tried out the 6.2-23 beta, and it looks like I'm still getting this issue. I'm starting to think it may be something about the card itself that is causing this ugliness. Especially since it looks like I'm not the only one that has had some issues with this card either. The unfortunate part though is that the primary purpose of the card was for the oculus rift, as it it is *incredibly* picky about Usb 3.0 controllers, as the reddit thread here documents Reading the thread, this controller is the only one they seem to support for the pci-e cards being manufactured currently....so that's frustrating. Passing the rift usb devices individually didn't work either, as oculus complained at the controller being used I may end up buying a different version of the board with the same controller to see if that may help at all. I'll post back if it does, I imagine other people using oculus may run into issues as well if this card is unstable across the board.
  8. I've had some pretty odd experiences with this card as well both on 6.1.9 and 6.2-23 beta. It behaved fine when I passed it through to a Windows VM for a long time. However, I made some system changes (I upgraded my GPU), and then it was causing massive system instability. System hangs, etc. I've been able to reliably pass in the card when there are no usb devices plugged in. But If I try to start the VM with usb devices plugged in, however, I have pretty disastrous results. There are times too when the card seems to be the cause of very high CPU utilization. I've tried stubbing the card out in the syslinux.cfg so that unraid doesn't address it...which didn't really work either. Kinda frustrating because there aren't any other recommended PCI usb cards for oculus.... and currently passing the oculus usb devices through the vm gui individually isn't working for me (I'm running the 6.2-23). I'd be interested in hearing your further experience with this card, and if it has been able to work consistently for you.
  9. Good to know. I'll let you know if I decide to try out the beta.
  10. EDIT: Just realized I hadn't followed defect post guidelines as carefully as I should, so changed that. Its also possible this post belongs in a different section of the forum, as a cautionary tale, if for instance, this behavior is actually expected for unraid, and I'm just a dummy Unraid 6.1.9 Description For a long time, I had no problem at all passing through my usb pci card into my Windows 10 VM. A few days ago, I made a number of system changes (new cpu, new graphics card, new peripheral usb devices). I got GPU passthrough up and running, and then proceeded to pass through my usb card (which had a few new usb devices plugged in) using the host-dev method. Immediately upon starting the VM, the entire webgui hangs. SSH access yes, but Shutdown script/cmd doesn't work. So I reluctantly hard-reset. Here's where things get nasty. My VMs were set to autostart (bad idea in retrospect), so immediately after booting up, the entire gui system hangs again, because--of course--my VM starts automatically. I think the recommended way of fixing this would be to edit my xml using virsh, but that doesn't work because, guess what, the virsh cmd is completely unresponsive. So with some digging, found out the xmls are located in /etc/libvirt/qemu. Despite the warnings about autogenerated code at the top of the files, by editing these I was able to get the system back in working order. So, after turning autostart off (thank god), I looked into this more, and it seems this problem was only caused when I would boot the system with the usb devices inserted. Removing the usb devices before boot fixed things. So, I'm guessing there's a problem with unraid failing to isolate that pci device before passing it to the VM? Adding a pci-stub line with my usb card's id to my syslinux.cfg file, and seems like everything works great. Boom. Problem avoided. I'm a bit curious as to what might be going on though. I'm late to the unraid game, so I'm coming behind all the hard work that has been done to get these virtualization technologies simple and convenient to use, so what went wrong might be obvious to someone else. When I first passed through my pci card, I followed this guide to get it to work, but I didn't see any mention of needing to add a pci-stub line. I wouldn't have even known there was such a thing as pci-stub if it wasn't mentioned as an upcoming feature in 6.2 How to reproduce Pass the above usb controller through to a Windows VM using host-dev method with a USB 3.0 device attached. Having Microsoft's Xbox One Wireless dongle plugged in seems to successfully trigger this. Actual Results The webgui hangs, and other parts of the system, become unstable (virsh doesn't respond. Clean shutdown seemingly isn't possible) Expected Results USB controller can be passed into VM with devices plugged in. Other information Diagnostics posted for those interested, though they're from my system which is working now. I imagine I wouldn't have any trouble reproducing the problem if you can't find what you need in there. tower-diagnostics-20160608-0038.zip
  11. To be frank, I haven't read the wiki, so I headed over to "lime-technology.com/wiki/index.php/Setting_up_CPU_and_board_temperature_sensing". The steps in the wiki after based on an older version. Nowadays the installation is more simplified. A short overview Step 1 - preparation Install the perl package, this is required to run the script "sensors-detect" in the background. Step 2 - detection Press the detect button to search and automatically fill in the required drivers or alternatively - if you know the name of the driver(s) - fill them in manually. Step 3 - saving and activation Press the save button to save and activate (load) the driver(s). This will create the fille /config/plugins/dynamix.system.temp/drivers.conf on your flash device. Step 4 - sensor assignment and display Use the dropdown menus under sensors to assign the appropriate sensor for CPU and motherboard readings. You may need to consult the user guide of your motherboard to find out which sensor needs to be selected here. Once a sensor selection is done, the corresponding item will be displayed in the footer at the right of the screen. Press apply to confirm your selection. This will create the fille /config/plugins/dynamix.system.temp/sensors.conf on your flash device. Done Note: to unassign or remove a sensor just unselect it from the dropdown menu. This will also allow to make new assignment (for example when the wrong sensor was chosen). Last but not least: see also the online help of system temp! Worked Beautifully . Thanks for the summary and help.
  12. So, I figured it out. I was following the instructions on the wiki religiously, so I put sensors.conf in /boot/config/plugins/dynamix/ After no dice trying a bunch of things, I took an educated guess or two. Seems, at least for my box, it wouldn't work until I put it in /boot/config/plugins/dynamix.system.temp/ Did I miss the docs on that? I admit I didn't read this entire thread. Or should the wiki be updated? Thanks bonenl and RobJ for the help.
  13. 1. 6.1.9 2. I did have an ad-blocker running. I've whitelisted the GUI, but haven't noticed anything different.
  14. Yep. I installed perl. Didn't use nerdpack though. Just downloaded it and installed using the installer described in the wiki. That & the detection script seemed to work just fine..
  15. Hmm. No dice. I'm set at real time updates. All the sensor fields are grayed out in the plugin...Is this expected?