Jump to content

hocky

Members
  • Posts

    53
  • Joined

  • Last visited

Posts posted by hocky

  1. 49 minutes ago, Aeloth said:

    I have set XMX_SIZE to 4G, XMS_SIZE to 2G and XMN_SIZE to 2G and it works relatively well with 3 players. I wouldn't worry about CPU pinning. I let it decide for itself how many cores to use. I have another 10 mods running on my server.
     

    Next, I have set:

    view-distance=10

    Mh, that´s strange.

    49 minutes ago, Aeloth said:

    Maybe you need more CPU cores. I guess

    Well, i´ve got 12 cores of which 5 are in use by VMs. One other docker (Ubiquiti Unifi server) is running.

    Overall load of the server is around or below 10% and not even half the RAM are currently utilized...

    Having said that, i think there must be something else...

  2. Hi,

    i´m trying to setup a Minecraft server for my son and his friends. Bear with me as i´m not really familiar with Minecraft and it´s terminology...

    The docker installed fine and we´re able to connect to the server and start playing. However, my son is reporting that the server is seriously lagging. We did a test with a friend and after a short time (2-3 minutes), the server crashed.

    At the server console, i´m getting messages like "can´t keep up. Is server overloaded?".

    I did CPU-pinning with 3 CPU cores. CPU is i7-8700 CPU @ 3.20GHz with 32GB RAM. I was thinking that should be enough for 1-2 players...

    Any ideas where to look at?

    Thanks,
    Hocky

  3. I managed to have the GTX1050 show up in the VM by adding these lines to the XLM-file:

        <hostdev mode='subsystem' type='pci' managed='yes'>
          <driver name='vfio'/>
          <source>
            <address domain='0x0000' bus='0x01' slot='0x00' function='0x0'/>
          </source>
          <address type='pci' domain='0x0000' bus='0x05' slot='0x00' function='0x0' multifunction='on'/>
        </hostdev>
        <hostdev mode='subsystem' type='pci' managed='yes'>
          <driver name='vfio'/>
          <source>
            <address domain='0x0000' bus='0x01' slot='0x00' function='0x1'/>
          </source>
          <address type='pci' domain='0x0000' bus='0x05' slot='0x00' function='0x1'/>
        </hostdev>

    Both the GPU and the audio part show up in device manager in Windows 10 and i´m also able to install the NVidia driver for the GPU.

    But after reboot, the NVidia driver is obviously not loaded and the GTX1050 show up with a yellow exclamation mark.

     

    Any idea what i´m missing?

  4. Hi,

    i´m running a Windows 10 VM on an UNRAID 6.9.2 (registered).

    The VM is running a light control software which is currently using an NVidia NVS295 GPU. To be able to use all of it´s features, i would like to change the NVS295 with a GTX1050Ti, which is already installed in the server and currently assigned (and working) to another VM.

    Unfortunately, i´m not able to change the GPU via the UNRAID-GUI. If i do it, the VM doesn´t boot (no boot drive)...

    I did quite some fiddling in the past to get the initial VM running with the NVS295, so there´s a high possibility i just screwed up the configuration.

    Please find attached the configuration files for the VM that is working well with the GTX1050 and the one that is currently working with the NVS295. Any help how to change the NVS295 to the GTX1050Ti is much appreciated.

    One constraint: since the license of the software running in the NVS295 VM is tied to some sort of machine id, i can´t just setup a new and fresh VM - which i think would solve my issue easily. I need to get it fixed with the current VM.

     

    Thanks,

    Thorsten

     

    ELM Nvidia NVS295 VM-Konfig .txt PC-Claudia working Config mit GTX1050.txt UNRAID-system devices.txt

  5. Hi,

    i´m running UNRAID 6.5.3 for quite some time now to my satisfaction. 

    However, i´d like to utilize my old GTX1050Ti for Plex hardware transcoding (Plex Pass is available) and learned, that i need to update my UNRAID server for this.

    I´ve read about stability issues with 6.7.x versions.

    So what would be a good choice for an UNRAID version between 6.5.3 and the current that is safe to use?

     

    TIA

    Regards, Hocky

  6. root@UNRAID:~# find /mnt/user/docker/appdata/MakeMKV/
    /mnt/user/docker/appdata/MakeMKV/
    /mnt/user/docker/appdata/MakeMKV/log
    /mnt/user/docker/appdata/MakeMKV/log/nginx
    /mnt/user/docker/appdata/MakeMKV/log/nginx/error.log
    /mnt/user/docker/appdata/MakeMKV/log/nginx/access.log
    /mnt/user/docker/appdata/MakeMKV/xdg
    /mnt/user/docker/appdata/MakeMKV/xdg/config
    /mnt/user/docker/appdata/MakeMKV/xdg/config/QtProject.conf
    /mnt/user/docker/appdata/MakeMKV/xdg/cache
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/openbox
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/openbox/sessions
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/openbox/openbox.log
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/fontconfig
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/fontconfig/CACHEDIR.TAG
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/fontconfig/69c00635-d29d-4916-9d3d-e607e35c1646-le64.cache-7
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/fontconfig/d24edf35-e992-437d-8dd3-bcbf0ea30221-le64.cache-7
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/fontconfig/6f126b59-7288-461b-a110-e71ee549114e-le64.cache-7
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/fontconfig/4fbaf3a1-0150-4e33-9d96-6ca239a1262e-le64.cache-7
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/fontconfig/33181cbf-a687-43fe-ae2a-d2a18eed9e78-le64.cache-7
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/fontconfig/df311e82a1a24c41a75c2c930223552e-le64.cache-7
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/fontconfig/17090aa38d5c6f09fb8c5c354938f1d7-le64.cache-7
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/fontconfig/2d31a572ce6667f6a0da9c8dc611898b-le64.cache-7
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/fontconfig/d21dfb71-1bac-4663-948b-a39e28221c70-le64.cache-7
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/fontconfig/e32a6f1d-ec59-4217-bf06-f2d99b2f9b71-le64.cache-7
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/fontconfig/e8d985e1-a57d-4110-8033-c0ccefd70882-le64.cache-7
    /mnt/user/docker/appdata/MakeMKV/xdg/cache/fontconfig/5ca8086aeacc9c68e81a71e7ef846b3b-le64.cache-7
    /mnt/user/docker/appdata/MakeMKV/data
    /mnt/user/docker/appdata/MakeMKV/data/_private_data.tar
    /mnt/user/docker/appdata/MakeMKV/.MakeMKV
    /mnt/user/docker/appdata/MakeMKV/update.conf
    /mnt/user/docker/appdata/MakeMKV/_private_data.tar
    /mnt/user/docker/appdata/MakeMKV/screen.conf
    /mnt/user/docker/appdata/MakeMKV/dump_SD_46B4F294C1B39626D2E5.tgz
    /mnt/user/docker/appdata/MakeMKV/Videos
    /mnt/user/docker/appdata/MakeMKV/myProfile.mmcp.xml
    /mnt/user/docker/appdata/MakeMKV/settings.conf

    As requested.

  7. 19 hours ago, Djoss said:

    I don't have the issue on my side... Can you provide the output of:

    
    find /mnt/user/appdata/MakeMKV

     

    Mh, in the meantime, i reinstalled the container and now it´s not even starting. It says "sed: couldn't open temporary file /config/sedXXXXXX: No space left on device". Need to find out what the issue might be...

  8. So i changed the custom script command from

    rclone sync /mnt/user/Data/Media/Bilder /mnt/disks/HiDrive/Media/Bilder/
    to

    rclone sync /mnt/user/Data/Media/Bilder HiDrive:users/tkhidrive/Media/Bilder/ -v

     

    A couple of test-runs look promising - i´m now down to 1m15s for a sync without changes which took 4h before. 🙂

    Thanks a lot, @Stupifier, for your excellent support!
     

  9. Hi,

    i´ve installed the rclone plugin to synchronize a folder sitting on my UNRAID server to a cloud storage.

    I´m using rclone together with the user scripts plugin. The script consists of the following command:

    "rclone sync /mnt/user/Data/Media/Bilder /mnt/disks/HiDrive/Media/Bilder/"

    I´ve configured user scripts to run in every night at 2:00am.

    In general, everything is working (mounting, synching), both source and destination folder are in synch.

    However, the synch process takes almost takes 4 hours and saturates the internet connection, although nothing has changed in the source folder (so it seems to me that this is the time rclone uses just to check if there are any changes).

    We´re talking about ~150GB in ~40.000 files (photo archive).

     

    Any advice how this can be accelerated?

     

    Regards,

    Hocky

  10. 3 minutes ago, bonienl said:

    The load is calculated based on the nominal power (to convert percentage to Wattage), which isn't present.

    Ups, sorry. I was a bit confused (didn´t have my second coffee yet). Load percentage is displayed, UPS load is missing.

     

    6 minutes ago, bonienl said:

    If the key is not present, there is no way to show its value unless there is an equivalent key, which I don't know.

    OK, not a problem. I don´t need it in UNRAID anyway.

  11. 2 minutes ago, bonienl said:

    You are missing the key "NOMPOWER".

    Oh, true. That could be the reason why it´s not displayed ;-)

    Is it sth that can be fixed or is it just the way it is?

    However, the load percentage is available but also not displayed in the UNRAID status.

  12. Thanks a lot guys.

    I´m currently experimenting with setting up a virtual HTPC with a GT1050 for testing purposes and ran into exactly the same issue.

    Switching into MSI-Mode helped. :-)

    Actually, you can brake down the lengthy description in the links posted into:

    1) Check if MSI is enabled with the MSI-Util.

    2) Change Registry values for graphics card and related sound device

    3) Reboot

    4) repeat step 1)

    5) Be happy

    Thanks again!

     

    Cheers, Hocky

     

×
×
  • Create New...