Jump to content

randomninjaatk

Members
  • Posts

    70
  • Joined

  • Last visited

  • Days Won

    1

Posts posted by randomninjaatk

  1. +1

     

    I want docker-compose to automate setting up a new system... I just started messing with it, and it would be way easier to have a yaml file with all the configurations to completely setup a new machine automatically....

    • Like 2
  2. So a big change, mostly to make my life easier. Is all the settings are now in a config file in: /config/sma (sonarr-sma/radarr-sma) containers, or /config/scripts/configs  for sabnzbd-extended container.

     

    It removes some of the ease of use that was previously setup. however its a lot to maintain when the script can change at anytime, requiring lots of re-tweaking. This should hopefully mean updates can be seamless and I don't have to actively maintain it.

  3. @Sandspike

     

    I don't know if you have tried my latest updates to these containers. But I decided to give downloading subs a try in my Sonarr-SMA container, and it did successfully download them. 

     

    So it should be working in all containers....

     

    If you want to chat more about this, feel free to PM me and we can possibly chat on Discord...

  4. So everything were discussing in here, only applies to the SAB container. (However, SMA is installed using the exact same method for all the containers I have, so that does apply to both...)

     

    The radarr/sonarr-sma containers that I put out, don't do this. They strictly use SMA if you enable it as a sonarr/radarr post processing script.

     

    Whereas this SAB container runs up to 4 functions for video files:

    1. Deobfuscate SAB Script
    2. Video File audio/subtitle language check
      1. MKV Cleaner (If enabled)
      2. SMA (MP4 automator, if enabled)

    I could look into adding an option to skip the language checks, I'd personally never use it because I like to have standards, and if a release doesn't meet those requirements then I don't trust it and I'm okay with just grabbing an alternative release until one passes...

  5. @Sandspike

     

    TLDR:

    Your likely downloading files that are tagged poorly. They should have been marked as the correct language from the uploader, but they are not. So the script is detecting that and marking the files bad as a result... Everything is working as intended. I've added some better error messaging if you have updates enabled and reboot your container to get the latest script updates....

     

     

    Detailed information:

     

    So pushed a couple more updates, to give better error messaging, you should see something like this now:

     

    Begin processing 1 of 1: File Name.mkv
    Checking for audio/subtitle tracks
    1 video track found!
    1 audio tracks found!
    1 subtitle tracks found!
    Checking for "eng" video/audio/subtitle tracks
    INFO: "cos" audio track found!
    INFO: "und" subtitle track found!
    ERROR: no "eng" audio/subtitle tracks found!

     

    As you can see now, it gives a better understanding to why it failed. (In my case: eng is the set language).

     

    So I manually checked this file, after the recent changes will no longer delete the file, just mark it as bad, and this is what I found.... So this file's audio is actually in English, and the subtitle is also in English, but because the uploader hasn't tagged anything properly, the script cannot fix that and marks it as bad... So everything is working as intended in this case, the script can detect poorly tagged files, but cannot fix them.... in this case....

     

    Some more background on the logic behind passing language verification requirements.

     

    The basic logic for passing the language requirement test is this:

     

    1. If a file has Foreign Audio (IE: not matching the set language), then a subtitle track must be included matching the set language language. If the file has foreign audio only and no matching subtitle language track, then the download is marked as failed. This is set that way because I personally wouldn't want a file with possibly JAP audio with JAP subtitles only, because it would be useless in my case because my primary language is English.

     

    2. If a file has UND/NULL audio language track, this is assumed to be the primary set language based on your settings. MKV Cleaner will re-tag those tracks to the correct language. This will not cause the script to remove/mark these files as bad. SMA will also still be able to process these files even with MKV Cleaner disabled, and SMA also has the ability to re-tag these tracks. However, if you do not have MKV Cleaner or SMA enabled using the "TRUE" setting. These downloads will be marked as failed with a script exit 1 based on your language setting in the container.

     

    MKV Cleaner will not re-tag or assume UND/NULL language subtitle tracks are the same language as the primary set language. I could make a change, so that it could re-tag those, just like it does for audio/video tracks. However, since they are text based tracks, I'd rather not incorrectly tag these as a different language. So that is why UND/NULL audio language tracks are considered bad and would not be re-tagged. These tracks are removed from the file if using MKV Cleaner, and SMA will handle them however you have it configured.

     

    Hope this gives you a better understanding of what's actually happening. I think the added error messaging makes it clearer. 

     

     

    Github Commit fix: https://github.com/RandomNinjaAtk/sabnzbd-scripts/commit/8982babfd173047297936ca5e5bd80ab6e0dfd21#diff-3d12cc7ce365dada5f5ca61464ce06c8

  6. @Sandspike

     

    So after some more debugging, it does not look like a bug at all. Especially with the latest changes. It seems that the incoming files are likely poorly marked with the wrong language and because of that, the script is smart enough to detect it and mark it as bad. 

     

    In one instance, the language of a file was actually English audio after testing the file, but was marked bad because the actual audio track was tagged with a different language.

     

    Basically my script is working as designed, but can't fix crappy uploaders.... In any case, I'd rather it personally mark these as bad, because they are not correct to begin with... So this is not a bug and everything is working as intended...

     

    If you reboot your container now, you should get the debugging version of the script. It will mark the download as failed if you have that enabled in SABnzbd, but it will no longer delete the file. 

     

    I can probably work on including some more descriptive error messaging so it's less confusing going forward. But everything appears to be working as intended...

     

  7. 9 hours ago, Sandspike said:

    Thanks I had to shut off the post processing script, as it keeps deleting files, even after cleaner appears to run.

     

    Video Post Processor: Process with MKV Cleaner: True

     

    You need to set that to: TRUE

    Just like the description says to enable it. Capitalization matters...

     

    I was seeing some of the same issues today with the script, and made a bugfix change that I think resolves it.

     

    The pre-reqs are installed via SMA's installation file:

    https://github.com/mdhiggins/sickbeard_mp4_automator/blob/master/setup/requirements.txt

     

    EDIT: Sample output

     

    Example output using MKV Cleaner with no changes needed:

    ====================================================
    Begin processing 1 of 1: File Name.mkv
    Checking for audio/subtitle tracks
    1 video track found!
    1 audio tracks found!
    2 subtitle tracks found!
    Checking for "eng" video/audio/subtitle tracks
    1 audio tracks found!
    2 subtitle tracks found!
    INFO: Video passed all checks, no processing needed
    Post Processing Complete!

    Example output using MKV Cleaner when changes are needed:

    ====================================================
    Begin processing 1 of 1: File Name.mkv
    Checking for audio/subtitle tracks
    1 video track found!
    1 audio tracks found!
    1 subtitle tracks found!
    Checking for "eng" video/audio/subtitle tracks
    1 "unknown" audio tracks found, re-tagging as "eng"
    1 subtitle tracks found!
    1 "unknown" video language track found, re-tagging as "eng"
    mkvmerge v19.0.0 ('Brave Captain') 64-bit
    '/mnt/disks/SSD01/downloads/sabnzbd/complete/sonarr/Folder/File Name.mkv': Using the demultiplexer for the format 'Matroska'.
    '/mnt/disks/SSD01/downloads/sabnzbd/complete/sonarr/Folder/File Name.mkv' track 0: Using the output module for the format 'AAC'.
    '/mnt/disks/SSD01/downloads/sabnzbd/complete/sonarr/Folder/File Name.mkv' track 1: Using the output module for the format 'AVC/h.264'.
    '/mnt/disks/SSD01/downloads/sabnzbd/complete/sonarr/Folder/File Name.mkv' track 2: Using the output module for the format 'text subtitles'.
    The file '/mnt/disks/SSD01/downloads/sabnzbd/complete/sonarr/Folder/File Name.merged.mkv' has been opened for writing.
    Progress: 0%
    Progress: 34%
    Progress: 67%
    Progress: 82%
    Progress: 99%
    Progress: 100%
    The cue entries (the index) are being written...
    Multiplexing took 2 seconds.
    SUCCESS: mkvmerge complete
    INFO: Options used: -d 1 --language 1:eng -a 0 --language 0:eng
    INFO: Renamed source file
    INFO: Renamed temp file
    INFO: Deleted source file
    ====================================================
    Post Processing Complete!

     

  8. MKV Cleaner is strictly a repackager, so it will not convert or anything else, but it will clean up the file. So it would be safe to enable it, and then have radarr/sonarr use SMA to further convert and process the file.

     

    To turn on MKV CLeaner, set "Video Post Processor: Process with MKV Cleaner:" to TRUE

     

    When I was using SMA, my preferred setup was:

     

    Sabnzbd with SMA conversion -> Radarr/Sonarr and SMA Tagging only.

     

    So the the file was converted before sonarr/radarr imported it and all radarr/sonarr was doing at that point was adding tags to the file when using MP4 as the format.

     

    I stopped using SMA because I was having trouble with forced subs (disposition) when using MKV. And it's easier to use sonarr/radarr to just grab your target quality and I simply want to cleanup the file (re-tag unknown audio to correct language and remove unwanted streams), not re-encode the video/audio. 

     

     

     

  9. It's probably a problem with your configuration settings, if you have it working in an instance already, you need to make sure your settings match your other one to get the same result.

     

    I think I'm going to remove all these settings from the docker config and move to using a external configuration file because it's a bit too much to maintain for a script I'm not actively writing/creating myself.

     

    The reason I added all the settings is to make using it more convenient.

  10. So quick update that should fix it, I've been primarily using MKV Cleaner only, so this hasn't been an issue for me.

     

    Fixed for SMA with: https://github.com/RandomNinjaAtk/sabnzbd-scripts/commit/adcbf92b926d9f13e394ce250c3e3b0d04894125

     

    To get the update, you need to have the "Auto Update SABnzbd Extended Scripts:" set to: TRUE and then reboot the container.

     

     

    8 minutes ago, Sandspike said:

    So I use SMA in the docker container for Radarr and Sonarr, but it isn’t downloading or encoding subtitles, so that is another issue. I did put the same options in the SAB container, but it doesn’t appear to be executing anything. I actually had to add the deobfuscate script to the script folders as well. Maybe a recommended add? MKV Cleaner, where would I enable that...I don’t remember seeing it. How much extra processing are we talking?

     

    MKV Cleaner uses MKVToolsNix to repackage the files into mkv based on your language preference. Its primary function is to mark UND/NULL audio streams to the specified language, and to remove unwanted audio/subtitle streams. Example: If you have a file with 10 audio streams, 1 is english, and your setting is english, all the other audio streams willl be removed via mkvmerge application. This also works the same for subtitles. If no modifications are needed, the script simply checks the file and moves on.

     

    The sonarr-sma and radarr-sma dockers I have need some updates probably. The main MP4 automator script was updated a bunch, and I'm not sure I've incorporated the config option changes. Those changes were added to this container. I may move the configuration to outside of the container settings, because its a lot to maintain and track when that script is updated. Also doesn't help that I'm not actively using it, and just using MKV Cleaner right now.

     

     

     

  11. It depends on what options your using. By default if you don't have SMA or MKV Cleaner enabled, if it does not detect the specified language track for either subtitle or audio, it says its a bad release.

     

    However in practice, sometimes I find what would be considered english is not marked as english in the original file (Typically UND or Null language). MKV Cleaner fixes that by ignoring that check and fixing the file, but I'm not sure the same workaround is in place when SMA enabled to run in that same scenario.

     

    Enabling MKV Cleaner would resolve that issue, but know that you'll be effectively processing the file twice, once with MKV Cleaner, a script I wrote, and then with SMA if your using that. 

  12. 1 hour ago, d3fault said:

    Hi i have tested the Container works with intel quick sync nicely. 1 thing I found ist the config file is overwritten every time the container starts.

    i have opend a issue in GitHub.

    Should be fixed in latest update...

     

    Edit:

    Also if you have any suggestions for the configuration file for quicksync encoding settings, I would appreciate it. I successfully get it to work with quicksync, but the quality seems to be terrible in comparison, but maybe I'm not using the right options for bitrate/profile/etc....

  13. The latest script implementation uses a caching engine, so if no changes need to be made to the existing cache, your technically only re-searching the existing cache.

     

    It's not likely to be possible to just skip checking them entirely, but caching does make that faster...

     

    Edit: In the config settings, you can control how many it attempts to process at a time, but you cannot change the order, so that means the first 100 in your case will always be re-checked. The setting to adjust from the default is  "LAD: Missing/Cutoff Amount", by default its set to the maximum

    • Like 1
  14. Docker: Lidarr (Nightly) with Lidarr Automated Downloader (LAD) script included!

     

    Script: https://github.com/RandomNinjaAtk/lidarr-automated-downloader

     

    Documentation can be found here: https://github.com/RandomNinjaAtk/docker-lidarr-lad

     

     

     

    IMPORTANT: I will be discontinuing this container sometime soon, its already removed from the app store.

     

    Please use: AMD and AMVD in the app store as replacements

     

    Documentation found here:

    https://github.com/RandomNinjaAtk/docker-amvd

    https://github.com/RandomNinjaAtk/docker-amd

     

     

    • Like 3
  15. 4 hours ago, knex666 said:

    Thanks. 

     

    Since I will not change the image because I dont want any fork you have 2 options:

    1. build your own with cron: https://github.com/nextcloud/docker/tree/master/.examples/dockerfiles/cron/apache

    2. use the cron from unraid (Webinterface) and use docker exec.

     

    Cheers

    Option 2 seems easiest, created the following script below and set it to run every 5 min using the "Users Scripts" plugin

    #!/bin/bash
    docker exec -u www-data Nextcloud php -f /var/www/html/cron.php
    exit 0

    Thanks!

    • Like 4
    • Thanks 3
×
×
  • Create New...