Jump to content

pupmeister

Members
  • Posts

    10
  • Joined

  • Last visited

Posts posted by pupmeister

  1. It is a network drive that I have mapped and mounted using the unassigned devices that I use for backups. I have just now unmounted as I now actually do the backups through a method that does not need it to be mounted locally.

     

    Thanks

  2. Hi I am getting a constant warning every few weeks from fix common problems stating that my system logs are filling up. This last time I was able to capture the diagnostic file before I had to reboot the system to clear the logs.

     

    I am running the latest stable release of Unraid 6.9.2 and use a AMD Ryzen 7 3700X on a ASRock X470 Taichi Ultimate.  I run multiple dockers including Emby, Tdarr, Radarr, Sonarr, sabnzbd and qbittorrent.

     

    The system will run for a few weeks with no problem and then suddenly I will get the notification that the log files are filling up and the system becomes unstable shortly thereafter and I normally forced to do a hard restart because I cant get any response from the system.

     

    Any help would be appreciated.

     

    blueduck-diagnostics-20210622-0600.zip

  3. I have got the following error from Fix Common Problems.

    "Your server has detected hardware errors. You should install mcelog via the NerdPack plugin, post your diagnostics and ask for assistance on the unRaid forums. The output of mcelog (if installed) has been logged"

     

    Luckily I had mcelog installed so I have attached my diagnostics, so hopefully somebody can help me with this.

     

    Furthermore I sometimes get errors that my log is to full. Last time this happened I was unable to download my diagnostic files and could only restart my machine. Hopefully somebody could tell me what is causing this error.

     

    Thanks

    Stephen

     

     

    blueduck-diagnostics-20210525-2038.zip

  4. On 8/26/2020 at 1:40 PM, druck21 said:

    Has anything changed recently? I had this all set up correctly and working great using dns validation through cloudflare, but lately whenever I try to check my SAB docker by using the letsencrypt domain, I can get to the login page, but once I login I just get stuck on a  "Lost connection to SABnzbd.." error screen. I can view SAB just fine when I go directly to the docker's internal address, just not when going through letsencrypt. Any ideas?

     

    Did you ever find out the reason and a solution for this problem?

  5. 31 minutes ago, Energen said:

    I won't speak for housewrecker but I can probably take a stab in the dark and say that scanning a folder will never happen with this tool.  It retrieves information (movie IDs) from Plex in order to build the related information -- Plex already does the hard work -- GAPS would have no way of knowing any type of movie id or where to scrape info from without Plexs' already-existing ID.  Make sense?

    I am just going from the description in the Github Page

    "Gaps searches through your Plex Server or local folders for all movies, then queries for known movies in the same collection. If those movies don't exist in your library, Gaps will recommend getting those movies, legally of course."

     

    I did try and get the windows version from the Github page working, but could not figure it out.

  6. On 12/5/2020 at 12:15 PM, FreeMan said:

    @housewrecker any progress on either an Emby/JellyFin version or the scan directories version? This looks fantastic, but as an Emby user, the Plex integration is, well, useless (to me).

    To quote the previous question, any update on this or the ability to just scan my movies folders.

  7. 1 hour ago, nicksphone said:

    in options 
    Linux FFmpeg NVENC binary (3.4.5 for unRAID compatibility) set to on?

     

    try this plugin its the one i use the most seems to fail on less files

    Tdarr_Plugin_d5d3_iiDrakeii_FFMPEG_NVENC_Tiered_MKV

    Yes Linux FFmpeg NVENC binary (3.4.5 for unRAID compatibility) is on.

     

    I actually found a solution that seems to work in thread on handbrake on Reddit, and was not mentioned in the posts that I read or needed for the other containers that use the GPU.

    Add a variable NVIDIA_DRIVER_CAPABILITIES and set the value to all .

     

    Seems to be working now.

     

    Thanks for trying to help me.

  8. On 6/26/2020 at 7:07 PM, nicksphone said:

    which plugin are you using to transcode?

    Tdarr_Plugin_MC93_Migz1FFMPEG Migz-Transcode Using Nvidia GPU & FFMPEG

    can you transcode on cpu only plugin?
    Yes can transcode on CPU, only has a problem with GPU (I used the CPU plugin of the above to test)
    do you have the all nvidia card info in the config some people forget to click on the advanced tab to add the --runtime=nvidia as well as the other 2 lines?

    Yes all are passed through and working as you can see from a check with nvidia-smi that I didtdarr_docker.png.fc2524585d9874f4abae13a664ccff41.png

    I also stopped any other dockers that use the GPU and I a do not have any VM's currently installed or running

    have you tried removing tdarr and the folder and doing a fresh install using the comunity plugins version as they are now all rolled into one?

    Tried a reinstall. I run the Tdarr aio version of the docker

    tdarr_aio.png.ea14bb2c81d0c70925a41c15867dcf01.png

     

     

    I have answered your questions in blue above to the best of abilities. I know that the GPU is working, because I can get it to pass through to Emby and

    unmanic. I did stop both these dockers during my testing to ensure that I did not interefere.

     

    Thanks

    Stephen

     

  9. On 5/16/2020 at 7:30 PM, aron7676 said:

    @HaveAGitGat I think I have troubleshot as much as I can. Do you have any ideas? 

    Not mentioned above, but I do have this checked on the options page too:

    image.png.b3e723900b00bf11589ee3aff58feb4d.png

     

    I rolled back to an older version even, to get an older driver (6.8.0-rc5).

     

    For whatever reason it is not seeing or passing through the driver info.



    I WAS able to get it partially working by creating a read only path from host /usr/lib64 to container /usr/lib64. I am sure this is wrong and will cause things to catch on fire later, but got a few transcodes going. a few are still erroring out though. 

     

     


     

    Has anybody been able to sort out this problem with the conversion not been able to load libnvcuvid.so.1? I have tried all the options quote above, and am running the latest 6.8.3 Nvidia Build, the AIO version of the docker.

×
×
  • Create New...