GJones

Members
  • Posts

    68
  • Joined

  • Last visited

Everything posted by GJones

  1. Though I know you might be aware, Plex is actually designed for this. It is derived from XBMC source and runs in a separate client/server model. I used XBMC for years but when I added more clients (now up to 3 Roku, 4 iPhone, 2 iPad, a MacBook Pro Retina and a few Chromebooks through PlexWeb), the Plex setup was much simpler to administer. Plex Media Server runs as a plugin on unRAID and works exceptionally well. I'm still a huge fan of XBMC, but it does not meet my needs as well. If you are running a single client or want separate HTPCs at each TV with separate libraries, XBMC may still be a better fit. Plex does have some disadvantages: no smart playlists, lack of PseudoTV functionality. But the advantages, for me, were too many to pass up: automated sync to iPhones/iPads centralized library management transcoding of files on the fly for different client types Excellent Roku support And I am in no way associated with Plex. I just hate to see someone get frustrated trying to get XBMC to do something which is more easily accomplished with something else.
  2. Actually, there are many uses for it. I use MakeMKV to rip the disks to unRAID. They sit there, being served from the MKV file under Plex and my automated script transcodes them into a smaller MP4 using HandBrake.
  3. Has anyone attempted to get HandBrake 0.9.9 working on unRAID 5.0? (The previous thread was somewhat dead)
  4. The syslog entry pertains to file permissions. No great concern.
  5. You will be fine without -C. I use it because I have four cores and only want to use three. I am looking at HD audio but my hardware does not support it. Updates coming soon for that.
  6. I am in the process of moving the search includes/excludes to tcOPTIONS. It is a balancing act to keep that somewhat maintainable and yet, efficient. That should be something up for testing next week. At that point, I will likely turn to get the options built into unmenu controls.
  7. I changed the name of the thread in the hope that it would make it more obvious what the package actually accomplishes.
  8. There is a log of every tcremkv and tcremkvblu job created by ts under /tmp. I had considered moving those files under the _transcode directory periodically, renamed according to the source file. The log you refer to would more than likely be based on the actions of tcrepack and tcrepackblu (still considering combining the two). This second log would be high-level, listing the actions of the two repack programs along with the end result of individual tcremkv and tcremkvblu tasks. It would also show files flagged as issues report on positive results as well. To do this, I would need to rework the current method for picking candidates in tcrepack and tcrepackblu. I would need to check each for a skip file in their directory prior to them being counted. This would be more user friendly in that a user that never logged in via telnet or ssh could fix their file and then remove the skip file from the directory.
  9. There are reasons to avoid renaming a failed file. One would be that the file will be completed at a later time. I may, however, write out a file to that folder pointing out that the file cannot be encoded at this time. Let me consider it. Renaming the file would be clean but I want to determine if there is another way. I do not want to rename every file in someone's collection just because they are using some strange encoding, etc. The "skip" file would allow a timestamp in there as well to keep tcre... from attempting to process it for a set period of time.
  10. Do you by chance have overlap between your HD and SD search directories in tcOPTIONS?
  11. Version 0961 Posted A new check avoids infinite loops by making sure that tcrepack and tcrepackblu aren't called rapidly. A new variable (MINBTWN) in tcOPTIONS controls the wait, which defaults to 300 seconds (5 minutes). If it runs more quickly than that, it exits with a message to avoid trying to transcode the same failing file forever.
  12. There are thousands of the files, but they contain one of two types of content: 1) ts-out.1NZ44J (as an example, but always the same) /usr/local/bin/tcremkv: line 34: [: !=: unary operator expecte unknown option (-C) **************************************** * Comparing File Sizes * **************************************** du: cannot access '/mnt/user/TV/Alias/Season_2/Alias_H264.S02E15.mkv': No such file or directory du: cannot access '/mnt/user/replaced/Alias.S02E15.mkv': No such file or directory **************************************** * ck * **************************************** BEFORE AFTER 2) ts-out.1c01Lj (other files contain two consecutive numbers) 70 71 Please let me know if you need something else. Thank you! I am testing the fix to avoid the race condition that reruns tcrepack or tcrepackblu. It stores the last run time for each and exits if it was less than 5 minutes ago. Please run tclanglist against the file it is failing to transcode. Then run tclang on it as well. Please post the results. This is an attempt to determine why the script dislikes your input file.
  13. Please post the contents of the /tmp/ts-out...... files where it finished immediately. It will help me determine the cause.
  14. Posting 096 now, corrected the last path issue in cron (cron jobs don't see /usr/local/bin in the path).
  15. In the meantime, run ts -K This will kill the queue.
  16. Cron does not see /usr/local/bin on the path. I will be posting a fix today. The failure is in it not seeing HandBrakeCLI in the path.
  17. When testing is complete, this is the objective. ISO files and VIDEO_TS files present a challenge because they include multiple titles. Single title formats are easily accomplished, though.
  18. The transcoding server listed in the User Customizations also uses HandBrake to transcode all identified files. It contains options to find files automatically (cron) or to transcode them one at a time.
  19. Asterisk can be somewhat resource-intensive depending on codecs used. The dependencies and maintenance of an Asterisk box are pretty heavy for your average unRaid user and hardware. I would advise against it.
  20. I am using shares, but this folder doesn't appear in the list of network folders in OS X (not sure if you have any Mac experience). It does appear in the terminal when I browse to /mnt/user/replaced though... Should I delete the directory /mnt/user/replaced, and then create a share in the UnRaid interface? Do not delete it. Just add the share if you want it to be a share. Then you can see it on the Mac as well.
  21. I assume you mean you are setting TEMPDISK to /mnt/disk4/Downloads and not TEMPDIR to /mnt/disk4/Downloads. The only thing going to /tmp/ts-out..... is the log. The log is using RAM but the output video file is being written to /mnt/disk4/Downloads/.transcode/thegreatest_VIDEO.mkv, using the directory you requested. You can read the log by typing ts -t (just like running tail -f /mnt/ts-out.logfilename, and will monitor the log until the command finishes or you hit CTRL-C). I will look into removing the log going to /tmp to save RAM, but it hadn't been an issue on my 4GB machine. Keep in mind that ts is part of the solution but not my program.
  22. Ah... but when I back up my movies to ISO, I only grab the primary video track, and everything is remuxed into one .m2ts file. I know it's a little lame, but Dune players don't support subtitles over MKV, so I'm working with hardware constraints. Do subtitles and lossless audio represent any stumbling blocks? Lossless audio and PGS subtitles are a problem only because HandBrake does not deal well with them. I use makemkv to pull the core audio and only forced subs. This lets HandBrakeCLI handle the audio sync correctly with the core DTS track and then I remux in the PGS subtitles if they are truly forced. I use mkvmerge (not part of this tool) to remux when necessary. It is rarely necessary for me. DVD subs work perfectly as-is.
  23. Yes, there is a very complex technical reason for this. Just kidding. I like the file moves to take as little time as possible. The quicker moves happen, the less can go wrong. An on-disk move is instantaneous. Moving across disks takes far far longer. My goal is to get those files out of the "main" directory as fast as possible. I check my "replaced" directory about once a week to make sure that the transcodes had no issues and empty the directory out. If you are using user shares (and I am assuming you are), /mnt/user/replaced is one central location, even if it is across several disks. I am trying to accomplish both fast and convenient.
  24. NEW VERSION 095 Changed final directory processing to create the directory if needed on the cache drive and send the file there Fixed one typo that was a # instead of a " Fixed a missing tcLIBUPD in tcreblu Fixed a PATH problem in the cron
  25. Thank you GJones - I was trying to be more specific to prevent the script from re-transcoding all my files without having to rename everything _H264. But I've take the advice given earlier and renamed all spaces to "_"... doesn't look as pretty, but it works! May I ask about the logic for putting the replaced files on the same disk as the original? Seems to me it might be easier to have them all kept centrally (on the cache drive for instance), rather than having to go through each disk to clear out? Maybe I'm missing something here though. The goal was to put the final file "back" on the same disk as the original, while really putting it on the cache disk. To make a really complex topic short, unraid does not handle this the way I thought it would. I now specifically write them out to the cache drive and let unraids mover put them back later. This means an immediate move for the old file (final dir to replaced dir on the same disk) and the new file (from the transcode directory to the final directory on the cache drive). As to the overall goal, it is pretty straightforward. I pull movies in using MakeMKV (not running on the unraid box, but wouldn't that be nice). I store them immediately because I want to be able to watch the movie on my media center immediately. The last step in tcremkv and tcreblu is to update the library on the XBMC server: Store the "raw" mkv Index it in XBMC tcrepack or tcrepackblu finds the file and transcodes it Put the transcoded file back in the original directory Move the original file to the replaced directory (in case of a problem) Re-index to find the new file and remove the old (now missing) file Yes, now that I think of it, I could have put those in separate directories but I think I avoided that simply from maintenance. I don't really like looking in separate places (or configuring XBMC to do so) for movies. I separate kids' movies from other movies so I can serve them up to the kids in one chunk. HD movies are separated only because of different storage needs (balancing larger files).