unRAID Server Release 6.0-beta15-x86_64 Available

Here you go

 

I can't make out any of the text on either of those screenshots 

 

I got it working

 

check my bios and VT-D was disabled

1)  When you get to the efi shell, type the following commands:

 

fs0:
cd efi/boot
bootx64.efi

 

This will get you to the installer for your OS.

fs0:

returns 'fs0:' is not a valid mapping.

 

The Mapping Table is listing BLK0: - BLK5:, with 0/1 saying "Floppy" and 3 saying "CDROM"

 

What ISO are you trying to use for the install media?

I've got a Windows7 Pro disc that I slipstreamed for recent updates, added virtio drivers to, and added an autoinstall.xml to.  I works fine for installation to the VMs I've done or installed to a PC when burned to CDROM.

1)  When you get to the efi shell, type the following commands:

 

fs0:
cd efi/boot
bootx64.efi

 

This will get you to the installer for your OS.

fs0:

returns 'fs0:' is not a valid mapping.

 

The Mapping Table is listing BLK0: - BLK5:, with 0/1 saying "Floppy" and 3 saying "CDROM"

 

What ISO are you trying to use for the install media?

I've got a Windows7 Pro disc that I slipstreamed for recent updates, added virtio drivers to, and added an autoinstall.xml to.  I works fine for installation to the VMs I've done or installed to a PC when burned to CDROM.

Windows 7 does not support UEFI. That's why OVMF isn't working for you.  Please see the wiki in my signature for more information. Windows 8 or newer is required for OVMF.

is there anyway to give a VM a IP address on the same subnet as my local?

is there anyway to give a VM a IP address on the same subnet as my local?

 

Yes. Create a network bridge from th e network settings page and then type the name of that bridge in the alternate network bridge under the vm settings page. Note:  this will only impact VMs created going forward. To update an existing vm to use the unraid managed bridge, you will need to edit the vm and turn on advanced mode.  Type the unraid managed bridge name in the appropriate field, update the VM, then start it up.

Just upgraded to b15, went smoothly other than unRAID decided to start running a parity check once it booted into b15. I guess I'll just let it do a parity check. I noticed that the VM settings page has an option for "PCIe ACS Override: Yes"

 

I am assuming that I can now take "pcie_acs_override=downstream" out of my syslinux file?

append intel_iommu=on vfio_iommu_type1.allow_unsafe_interrupts=1 pcie_acs_override=downstream initrd=/bzroot

Just upgraded to b15, went smoothly other than unRAID decided to start running a parity check once it booted into b15. I guess I'll just let it do a parity check. I noticed that the VM settings page has an option for "PCIe ACS Override: Yes"

 

I am assuming that I can now take "pcie_acs_override=downstream" out of my syslinux file?

append intel_iommu=on vfio_iommu_type1.allow_unsafe_interrupts=1 pcie_acs_override=downstream initrd=/bzroot

Wonder if your flash is corrupt and couldn't update the run status of the array when you stopped the array to reboot and so "detected an unclean shutdown".

Wonder if your flash is corrupt and couldn't update the run status of the array when you stopped the array to reboot and so "detected an unclean shutdown".

I restarted the array with the "powerdown -r" command so I was thinking it has something to do with that plus removing the old VM plugin before the reboot because the VM's didn't shutdown like normal. I'll restart the array tomorrow and see if it tries to start the parity check again and then I will fix the flash drive if that's the case.

Just upgraded to b15, went smoothly other than unRAID decided to start running a parity check once it booted into b15. I guess I'll just let it do a parity check. I noticed that the VM settings page has an option for "PCIe ACS Override: Yes"

 

I am assuming that I can now take "pcie_acs_override=downstream" out of my syslinux file?

append intel_iommu=on vfio_iommu_type1.allow_unsafe_interrupts=1 pcie_acs_override=downstream initrd=/bzroot

 

If you do, it will change the setting on the vm settings page to "no".

 

Also, there is a very long blog post from Alex Williamson about this which may be hard to read through, but it covers some important information about this patch and what it actually does.  Using ACS overrides is at your own risk.

 

Alex Williamson post for more information:  http://vfio.blogspot.com/2014/08/iommu-groups-inside-and-out.html

is there anyway to give a VM a IP address on the same subnet as my local?

 

Yes. Create a network bridge from th e network settings page and then type the name of that bridge in the alternate network bridge under the vm settings page. Note:  this will only impact VMs created going forward. To update an existing vm to use the unraid managed bridge, you will need to edit the vm and turn on advanced mode.  Type the unraid managed bridge name in the appropriate field, update the VM, then start it up.

Great Thanks

 

Using ACS overrides is at your own risk.

 

I would assume by this quote that you recommend running the VM's with pcie acs override set to "no" and only turning it on if a pci-e device is not functioning properly?

Installed the latest version and everything seems to work OK - did uninstall the VM manager before the required reboot.

The only one thing I cannot figure out is how to pass a graphics card for a Win10 VM.

I used both SeaBIOS and OVMF w/ and w/o  "ACS overrides" but I cannot get anything out of my GeForce 210 PCIE card. The card has VGA, HDI and DVI outputs - is rather old and maybe here lies the issue.

The mouse and keyboard attached to the system do pass trough it's just the graphics card that doesn't seem to cooperate.

Before I go about and get a new card I need to make sure that I did try all the options.

 

Thanks.

 

Edit:

Did try again with the configuration shown in the picture attached and the system did reboot

Now the flash drive seems to be read only, will check the drive after ~ 8 hours when the parity check finishes.

syslog.zip

Using ACS overrides is at your own risk.

 

I would assume by this quote that you recommend running the VM's with pcie acs override set to "no" and only turning it on if a pci-e device is not functioning properly?

 

Even broader...  Don't mess with PCIe ACS Overrides unless you read and understand that post by Alex Williamson.  If you have multiple PCI devices in an IOMMU group for which you want to be able to assign to individual VMs, the better solution is to try and remount the devices to an alternate PCI slot in hopes that it would be bound to an alternet PCI root port, thereby allowing it to be in a separate IOMMU group.

I still am experiencing spin down issues with this beta. Any ideas to figure out what keeps making those discs spin up? Seems to be something in the Unraid environment itself since if I isolate it from the network I still get this problem, frustrating...especially with summer coming. Also no spin down issues pre-beta 13 as far as I remember.

I still am experiencing spin down issues with this beta. Any ideas to figure out what keeps making those discs spin up? Seems to be something in the Unraid environment itself since if I isolate it from the network I still get this problem, frustrating...especially with summer coming. Also no spin down issues pre-beta 13 as far as I remember.

Disabling your plugins / dockers / vm's will help to narrow it down and then bring them up one at a time

 

It completed at 11:35pm - the (today) part is the incorrect part

 

Myk

 

I did a parity check before upgrading last night to current version and the dashboard says:

 

Last checked on Friday, 04/17/2015, 11:35 PM (today), finding 0 errors.

 

That was last night - today is 4/18/15

 

Myk

 

Did it just complete today?

 

I can confirm, with the attached pic can't attach here so quoted below, a tiny bug in the determination of 'yesterday' and 'today'.  Looks like an 'off by one' bug.  Started a parity check this morning, ran almost 11 hours, finished a few minutes ago, today, Sunday the 19th.

 

Last checked on Sun 19 Apr 2015 07:17:14 PM EDT (yesterday), finding 0 errors

Duration: 10 hours, 47 minutes, 44 seconds. Average speed: 77.2 MB/sec

 

Otherwise, v6.0-beta15 has been great.  I particularly am happy to see my Cache drive displayed on the Main page, with User Shares not enabled; and to see the syslog Download button at the top of the syslog page.  Thank you.

also in the last version of VM manager after install and reboot it always went back to the installation routine from the iso again, none of that nonsense now.

 

got the login prompt straightaway.

So your initial impressions with OVMF are good then?  I realize you have more testing yet, but first impressions are important.

 

few teething troubles with the install, but so far so  good.

 

i killed the vm and will run another tomorrow with a graphics card and see what happens, just wanted to familiarise myself with the setup.

Ok there may be some snags with having multiple ovmf guests, but we are working those out.

 

by killed i mean obliterated, banished, deleted and inhumated, one could even say terminated with extreme prejudice.

Ah, ok then. You'll be fine.

 

 

i think my card manufacturer are lying at worst or misleading at best on their website.

the page for my card has a uefi compliant symbol on it, although nothing on the box to that effect.

 

http://www.club-3d.com/index.php/products/reader.en/product/radeon-hd-6450-pci-express-x1.html

 

it was a cheapy card bought because it was pcie 1x and that's the only slot left on my mobo.

 

it's a club 3d radeon 6450 , and whatever i try with ovmf enabled and this card as either the primary or secondary graphics card ends up with a blank screen when starting the vm.

 

my go file

 

#!/bin/bash

# Start the Management Utility
/usr/local/sbin/emhttp &
cp /boot/scripts-etc/myscripts/* /usr/bin/

# set syslog file for remote logging
killall rsyslogd
mv /etc/rsyslog.conf /etc/rsyslog.orig
cp /boot/config/rsyslog.conf /etc/rsyslog.conf
rsyslogd

#bindpci
# /usr/local/sbin/vfio-bind 0000:07:00.0 0000:07:00.1

 

NOTE, tried it both with and without the vfio-bind line commented out, same result for both.

 

syslinux.cfg

 

default /syslinux/menu.c32
menu title Lime Technology
prompt 0
timeout 50
label unRAID OS
  menu default
  kernel /bzimage
  append intel_pstate=disable initrd=/bzroot
label unRAID OS Safe Mode (no plugins)
  kernel /bzimage
  append initrd=/bzroot unraidsafemode
label Memtest86+
  kernel /memtest
label Xen/unRAID OS
  kernel /syslinux/mboot.c32
  append /xen --- /bzimage --- /bzroot
  label Xen/unRAID OS Safe Mode (no plugins)
  kernel /syslinux/mboot.c32
  append /xen --- /bzimage --- /bzroot unraidsafemode

 

tried all possible combos of qemu emulated processor/xeon q35/i440x to no avail.

 

 

Ok, a few things you can try here.  First, let's try using the pciback hide mechanism to bind the devices you want to use with passthrough prior to the OS itself booting.  This will involve a little command line work and a modification to your syslinux.cfg file under /boot/config/syslinux/.

 

First, go to the command line and type this:

 

lspci -n

 

Should print out a list something like this:

 

00:00.0 0600: 8086:0c00 (rev 06)
00:01.0 0604: 8086:0c01 (rev 06)
00:01.1 0604: 8086:0c05 (rev 06)
00:02.0 0300: 8086:0412 (rev 06)
00:03.0 0403: 8086:0c0c (rev 06)
00:14.0 0c03: 8086:8c31 (rev 04)
00:16.0 0780: 8086:8c3a (rev 04)
00:19.0 0200: 8086:153b (rev 04)
00:1b.0 0403: 8086:8c20 (rev 04)
00:1c.0 0604: 8086:8c10 (rev d4)
00:1c.3 0604: 8086:8c16 (rev d4)
00:1f.0 0601: 8086:8c44 (rev 04)
00:1f.2 0106: 8086:8c02 (rev 04)
00:1f.3 0c05: 8086:8c22 (rev 04)
01:00.0 0300: 1002:68f9
01:00.1 0403: 1002:aa68
02:00.0 0300: 10de:1004 (rev a1)
02:00.1 0403: 10de:0e1a (rev a1)
04:00.0 0400: 1a0a:6202 (rev 01)

 

Locate your PCI devices by their ID (the first column of numbers in this format:  ##:##:#.  Your GPU device id is listed in the VM manager edit page when you select the GPU.  Once you've found it and it's audio counterpart (the ##:##.1), note the vendor/product ids for each and put them in your syslinux config file.  For example, in mine, I have this:

 

label unRAID OS
  kernel /bzimage
  append pci-stub.ids=1002:68f9,1002:aa68,10de:1004,10de:0e1a initrd=/bzroot

 

After applying that, reboot your server and give it another whirl.  Also, remove any VFIO bind scripts you use.  We take care of all that automatically for you.

 

Also, some vendors may have updates to the BIOS firmware for their cards.  Check to see if there is one for yours.  If the card supports UEFI, it really should work, but in some scenarios, we've had challenges.

 

Other things you can try:

 

1)  Adjust to Q35 or i440fx chipset.

2)  Load your ROM BIOS for the GPU manually.

 

To do the latter, you will first need to find your BIOS from this site:  http://www.techpowerup.com/vgabios/

 

Download the bin for your BIOS and save it to somewhere on your cache or in the array.

 

Edit the XML for your VM through the webGui.  Locate the section for your GPU pass through which will look like this:

 

    <hostdev mode='subsystem' type='pci' managed='yes'>
      <driver name='vfio'/>
      <source>
        <address domain='0x0000' bus='0x01' slot='0x00' function='0x0'/>
      </source>
    </hostdev>

 

Add the following after the </source> but before the </hostdev>:

 

<rom bar='on' file='/mnt/path/to/rom.bin'/>

 

Replace the file path to the path of your bin file.

 

I've had much easier success with NVIDIA devices that have UEFI BIOS than AMD.  I actually don't have any AMD cards that have UEFI capability.  Even a modern R9-290 I got from ASUS about a year ago doesn't have UEFI support.  With NVIDIA, most devices as of the 650 series and up seem to have it, and I've had particularly good luck with EVGA.

 

Give these things a go and let me know how it goes.  There are a few other things we can try as well.  I'd be inclined to have you try that GPU without the OVMF setting (bet it would work), but let's keep trying OVMF first, as it is definitely going to be the preferred method.

sparklyballs, you can also try NOT passing through the sound device from the video card to see if that works.  I remember having luck with one of the AMD video cards I have when I didn't pass though its HDMI audio.

Just thought I would post an update: the webgui is much snappier. It pretty much instantly loads from tab to tab, whereas before there was usually a significant delay between some tabs. Also, the VM XML editor is fantastic! Just spent the afternoon recreating a few of my vm's with the newer templates. Cleaned up a lot of code that wasn't  needed and it was a breeze to do. Way to go!

Ok, a few things you can try here.  First, let's try using the pciback hide mechanism to bind the devices you want to use with passthrough prior to the OS itself booting.  This will involve a little command line work and a modification to your syslinux.cfg file under /boot/config/syslinux/.

 

First, go to the command line and type this:

 

lspci -n

 

Should print out a list something like this:

 

00:00.0 0600: 8086:0c00 (rev 06)
00:01.0 0604: 8086:0c01 (rev 06)
00:01.1 0604: 8086:0c05 (rev 06)
00:02.0 0300: 8086:0412 (rev 06)
00:03.0 0403: 8086:0c0c (rev 06)
00:14.0 0c03: 8086:8c31 (rev 04)
00:16.0 0780: 8086:8c3a (rev 04)
00:19.0 0200: 8086:153b (rev 04)
00:1b.0 0403: 8086:8c20 (rev 04)
00:1c.0 0604: 8086:8c10 (rev d4)
00:1c.3 0604: 8086:8c16 (rev d4)
00:1f.0 0601: 8086:8c44 (rev 04)
00:1f.2 0106: 8086:8c02 (rev 04)
00:1f.3 0c05: 8086:8c22 (rev 04)
01:00.0 0300: 1002:68f9
01:00.1 0403: 1002:aa68
02:00.0 0300: 10de:1004 (rev a1)
02:00.1 0403: 10de:0e1a (rev a1)
04:00.0 0400: 1a0a:6202 (rev 01)

 

Locate your PCI devices by their ID (the first column of numbers in this format:  ##:##:#.  Your GPU device id is listed in the VM manager edit page when you select the GPU.  Once you've found it and it's audio counterpart (the ##:##.1), note the vendor/product ids for each and put them in your syslinux config file.  For example, in mine, I have this:

 

label unRAID OS
  kernel /bzimage
  append pci-stub.ids=1002:68f9,1002:aa68,10de:1004,10de:0e1a initrd=/bzroot

 

After applying that, reboot your server and give it another whirl.  Also, remove any VFIO bind scripts you use.  We take care of all that automatically for you.

 

Also, some vendors may have updates to the BIOS firmware for their cards.  Check to see if there is one for yours.  If the card supports UEFI, it really should work, but in some scenarios, we've had challenges.

 

Other things you can try:

 

1)  Adjust to Q35 or i440fx chipset.

2)  Load your ROM BIOS for the GPU manually.

 

To do the latter, you will first need to find your BIOS from this site:  http://www.techpowerup.com/vgabios/

 

Download the bin for your BIOS and save it to somewhere on your cache or in the array.

 

Edit the XML for your VM through the webGui.  Locate the section for your GPU pass through which will look like this:

 

    <hostdev mode='subsystem' type='pci' managed='yes'>
      <driver name='vfio'/>
      <source>
        <address domain='0x0000' bus='0x01' slot='0x00' function='0x0'/>
      </source>
    </hostdev>

 

Add the following after the </source> but before the </hostdev>:

 

<rom bar='on' file='/mnt/path/to/rom.bin'/>

 

Replace the file path to the path of your bin file.

 

I've had much easier success with NVIDIA devices that have UEFI BIOS than AMD.  I actually don't have any AMD cards that have UEFI capability.  Even a modern R9-290 I got from ASUS about a year ago doesn't have UEFI support.  With NVIDIA, most devices as of the 650 series and up seem to have it, and I've had particularly good luck with EVGA.

 

Give these things a go and let me know how it goes.  There are a few other things we can try as well.  I'd be inclined to have you try that GPU without the OVMF setting (bet it would work), but let's keep trying OVMF first, as it is definitely going to be the preferred method.

 

it worked in the last version of VM manager but i couldn't get any sound with any of the myriad of alsa config setups i found online.

 

I'll give this a go tomorrow, i do occasionally sleep, lol.

sparklyballs, you can also try NOT passing through the sound device from the video card to see if that works.  I remember having luck with one of the AMD video cards I have when I didn't pass though its HDMI audio.

 

LOL, when my 4 yr old niece is kicking the crap out of my shins because there's no sound on her barbie movies I am going to curse your name.

Just thought I would post an update: the webgui is much snappier. It pretty much instantly loads from tab to tab, whereas before there was usually a significant delay between some tabs. Also, the VM XML editor is fantastic! Just spent the afternoon recreating a few of my vm's with the newer templates. Cleaned up a lot of code that wasn't  needed and it was a breeze to do. Way to go!

 

You can thank Tom for that.  He put in lots of little optimization improvements for a snappier more responsive experience.  These were under the hood tweaks that have seemed to have a big impact.

sparklyballs, you can also try NOT passing through the sound device from the video card to see if that works.  I remember having luck with one of the AMD video cards I have when I didn't pass though its HDMI audio.

 

LOL, when my 4 yr old niece is kicking the crap out of my shins because there's no sound on her barbie movies I am going to curse your name.

Haha, don't want you to get bruised shins.  Does your motherboard have analog audio outputs? If so, use that device for the sound portion.  That's actually how Jon rolls.

sparklyballs, you can also try NOT passing through the sound device from the video card to see if that works.  I remember having luck with one of the AMD video cards I have when I didn't pass though its HDMI audio.

 

LOL, when my 4 yr old niece is kicking the crap out of my shins because there's no sound on her barbie movies I am going to curse your name.

 

Well you could go buy an NVIDIA GPU instead ;-).  Or you can pass through on-board audio (like analog or optical maybe even).  If you could copy/paste your PCI devices here (from the Tools -> System Devices page), I could tell you wish are which.