Jump to content

Naiqus

Members
 View Profile  See their activity

  • Posts

    10

  • Joined

  • Last visited

 Content Type 

Posts posted by Naiqus

    NVMe device assignment messed up

    in Prerelease Support [DEPRECATED]

    Posted

    1 hour ago, bonienl said:

    @Naiqus and @HellDiverUK

     

    Can you also post the output of this command:

     

    udevadm -x /dev/nvme0n1

     

    Ps. I've made an update for the Dynamix SCSI devices plugin to address changes between old and new persistent-rules files, but I doubt it will correct the nvme issue, hence my question above.

     

     

    Thanks for your effort and reply. I already updated you plugin. It turned out that the driver is back in the list, but it is not renamed properly. The same with before the plugin installing.

    The command you gave seems not valid on my machine. I suppose you wanna see the info so here is the output:

     

    Quote

    root@Tower:~# udevadm info /dev/nvme0n1
    P: /devices/pci0000:00/0000:00:01.1/0000:01:00.0/nvme/nvme0/nvme0n1
    N: nvme0n1
    S: 500GB_S3EUNX0HC10760J
    S: 960
    S: EVO
    S: SSD
    S: disk/by-id/nvme-Samsung
    S: disk/by-id/nvme-eui.0025385c61b0578f
    E: DEVLINKS=/dev/500GB_S3EUNX0HC10760J /dev/960 /dev/EVO /dev/SSD /dev/disk/by-id/nvme-Samsung /dev/disk/by-id/nvme-eui.0025385c61b0578f
    E: DEVNAME=/dev/nvme0n1
    E: DEVPATH=/devices/pci0000:00/0000:00:01.1/0000:01:00.0/nvme/nvme0/nvme0n1
    E: DEVTYPE=disk
    E: ID_PART_TABLE_TYPE=dos
    E: ID_SERIAL=Samsung SSD 960 EVO 500GB_S3EUNX0HC10760J
    E: ID_SERIAL_SHORT=S3EUNX0HC10760J
    E: MAJOR=259
    E: MINOR=0
    E: SUBSYSTEM=block
    E: USEC_INITIALIZED=13362312


     

     

     

    NVMe device assignment messed up

    in Prerelease Support [DEPRECATED]

    Posted

    5 minutes ago, bonienl said:

    Can you show the output of 

    
ls -l /dev/disk/by-id

    (with and without the scsi plugin)

     

    Before install the plugin:

    Quote

    Jun  5 11:26 nvme-Samsung -> ../../nvme0n1p1
    Jun  5 11:26 nvme-eui.0025385c61b0578f -> ../../nvme0n1
    Jun  5 11:26 nvme-eui.0025385c61b0578f-part1 -> ../../nvme0n1p1

     

    After install the plugin, the nvme disk is also disappeared when I run this command.

    Anybody planning a Ryzen build?

    in Motherboards and CPUs

    Posted

    5 hours ago, Frag-O-Byte said:

     

     

    A Drill, with a 2.5CM diameter auger bit down in the corner of the skirting board,

     

    Im trying to make 3 systems.

     

    1 used by me for gaming, 1 used by my daughter for youtube/web surfing and 1 more for gaming for my son.

     

    Also i see your point with the audio thru HDMI, that will work, but for Keyboard and mice ect.. can i use a hub?

     

    8 hours ago, tsouza11 said:

    and usb controller?

     

    I've only managed to pass one usb controller to a VM. My workaround is to buy a PCIe x1 USB controller and pass it to the other VM. HDMI audio works fine by the way.

    [6.3.2] Linux VM crashes Unraid host system on a ryzen build (CPU stalls)

    in General Support

    Posted

    On 3/12/2017 at 3:27 PM, RobJ said:

    Nice, a Ryzen syslog!  And it looks very impressive!

     

    You were able to boot at least, but I'm afraid you're running too close to the bleeding edge!  This kernel is the first to have any support for Ryzen, but just guessing, it doesn't look like kvm (and related modules) have been updated for it yet.  You have turned on ACS override, but I understand that's not working with Ryzen, not updated yet.  For correct and stable operation, I think you'll have to wait for kernel and kvm subsystem updates.  However there is one other user who seems to be fully operational.

     

    Look for the Ryzen thread (don't remember where it is).  There are some very knowledgeable people there.

    Yeah, concerning what I've got I am totally lucky. Now I found a workaround which is to use "host_model" instead of "host_passthrough" in CPU config. Now the Linux VM is working fine and nothing weird happened yet. Hopefully the passthrough will work in the following unraid version with new kernels.

    [6.3.2] Linux VM crashes Unraid host system on a ryzen build (CPU stalls)

    in General Support

    Posted

    Hi, I've got a ryzen build recently with two graphic cards. It works fine with Windows VMs. But when I migrate my Linux system to the build, it started to crash the whole system. 

     

    When the Linux vm come back to idle from fully load state, it hang after a few seconds. At the same time the web interface was also down. The Linux VM could not be killed by virsh command. Following that, other VMs crashed too. 

     

    The VM hosted is Linux Mint 18.1. I've tried with kernel 4.4.0, 4.8.0, and 4.10.1, all crashed in the same way with “rcu_preempt detected stalls on CPUs/task” error message. I though it may be the problem of CPU scaling in the guest OS, but it seems that with CPU host-passthrough mode, the CPU scaling function is disable already in guest OS.

     

    Further information:  I used a single nvme SSD in the array without parity drive. With all VMs running, I left 4GB ram for Unraid. Memory test over night passed. No OC. CPU C6 mode disabled in BIOS.

     

    I am new to Unraid and qemu-kvm, any help will be appreciated. 

     

     

    syslog_crash.txt

    tower-diagnostics-20170312-1125.zip

    Anybody planning a Ryzen build?

    in Motherboards and CPUs

    Posted

    12 hours ago, mikeyosm said:

    If anyone manages to get 2 x GPUs passed through to seperate VMs let me know. I plan to get an MSI x370 titanium and 1800x with 32GB g.skill.

    IOMMU grouping issues is putting me off at the moment.

    I've managed it, with a Gigabyte X370 Gaming 5 mainboard. IOMMU groupping issue is resolved by AC override. Everything works fine up to now. 

    • Upvote 1
×
×
  • Create New...