TrueHD doesn't decode?

Server Version#: 1.14.1.5488 (Synology NAS 918+)
Player Version#: Doesn’t matter

After uploading a MKV Blueray to my Plex media folder, it won’t play if I leave the default audio track set to TrueHD 7.1. It won’t play any player device in my house (Windows 10, Vizio, TCL, or Roku devices). I simply edit the movie and select ANY other audio format within Plex and then it streams just fine.

It’s a pain to keep doing this… Is there a “fix”? I’m lost in the hundreds of threads I find on this subject. Please keep it simple. :slight_smile:

//Brew

DEBUG log files please (the ZIP file)

can’t help fix without seeing what’s broken :wink:

Here you go!

Plex Media Server Logs_2019-01-22_18-25-40.zip (4.2 MB)

Thanks for the logs.

We have a few things to fix.

  1. You have more than 8192 diretories you’re trying to monitor. Linux’s default table size isn’t big enough. We need to increase that. Thanks to Synology getting “sneaky” we now must modify their stuff.

  2. As root (sudo sh command) change lines 16 & 17 of /etc/init/s2s_daemon.conf, save the changes, then restart DSM. You will need check this after each DSM update to make certain they didn’t undo you enhancement.

 16:  env MaxWatches=32768
 17:  env MaxQueueEvents=163840
  1. This might be bad codec download. It’s impossible to tell with all the iNotify errors in the way. It also could be obsolete due to age
Jan 05, 2019 12:31:49.843 [0x7f9893a7d700] WARN - Transcode runner appears to have died.
Jan 05, 2019 12:31:49.844 [0x7f989a511700] WARN - Transcode runner appears to have died.
Jan 05, 2019 12:31:57.818 [0x7f9893a7d700] ERROR - [Transcoder] [ac3 @ 0x217cb80] frame sync error
Jan 05, 2019 12:31:57.818 [0x7f989a511700] ERROR - [Transcoder] [ac3 @ 0x32959c0] frame sync error
Jan 05, 2019 12:31:57.819 [0x7f989937d700] ERROR - [Transcoder] Error while decoding stream #0:2: Invalid data found when processing input
Jan 05, 2019 12:31:57.819 [0x7f989a7ff700] ERROR - [Transcoder] Error while decoding stream #0:2: Invalid data found when processing input
Jan 05, 2019 12:31:57.833 [0x7f989378f700] ERROR - [Transcoder] Could not write header for output file #0 (incorrect codec parameters ?): No such file or directory
Jan 05, 2019 12:31:57.834 [0x7f9899c47700] ERROR - [Transcoder] Error initializing output stream 0:0 --
  1. This, however is a major problem. You are attempting to perform scanning/database activities while trying to transcode (perhaps subtitles?) ? the database is grossly fragmented and preventing PMS from communicating with itself properly.
Jan 22, 2019 17:28:32.796 [0x7fedd8cd7700] WARN - SLOW QUERY: It took 16090.000000 ms to retrieve 79 items.
Jan 22, 2019 17:28:32.818 [0x7fedcc447700] WARN - SLOW QUERY: It took 16110.000000 ms to retrieve 16 items.
Jan 22, 2019 17:28:32.842 [0x7fedd17ff700] WARN - SLOW QUERY: It took 16200.000000 ms to retrieve 79 items.

The remedy for this:

  1. Stop all Scanning
  2. Optimize database
  3. Make certain the iNotify changes above are implemented
  4. Rescan ONE library section at a time until all have caught back up
  5. Optimize database between each library section scan
  6. Retest video transcoding once complete.

OK… Let me try to understand…

  1. Line 16 (MaxWatches) was updated… But line 17, was already at 163840 so I didn’t touch it.

  2. No idea what you’re asking me to “change” when it comes to iNotify?

  3. Scanning… This is odd. Where is scanning configured? My media files haven’t changed in a bout 6 months since I loaded the NAS… I also have Weekly optimization enabled… All scheduled tasks are configured via Plex starting at 5am and ending at 7am… Is it possible they have NEVER finished properly??

That is the MaxWatches and MaxQueueEvents settings.

To manually scan a library, in Plex Web, click on the ellipses next to the library name, then choose “Scan Library Files.”

To manually optimize the database, click on the ellipses to the right of “Libraries” and choose “Optimize Database.”

  1. I am showing you how it should look. Some DSM versions are different.

  2. These “MaxWatches” is the name of the kernel’s data table used by the iNotify monitoring service. PMS uses that service to be notified when you add media to your media folders.

  3. From what I can see, either the scanning never finished or never finished matching everything. It’s not possible to tell yet.

OK… So I update the iNotify values and rebooted… I Optimized the DB which took about 15 seconds… Then I scanned my 3 libraries (Music, Videos, Photos)… Each scan took a few seconds. Then I optimized the DB again which took maybe 10 seconds.

Videos configured for TrueHD still won’t decode and play. The movie I tested was “A Quiet Place” and the player was my Windows 10 machine using Chrome…

Logs attached for the next review please.

Plex Media Server Logs_2019-01-23_18-43-40.zip (3.3 MB)

Right now, it is going crazy trying to tag all your photos.

The database is overwhelmed and I bet your CPU usage is VERY high

Jan 23, 2019 00:08:54.360 [0x7fd661111700] WARN - SLOW QUERY: It took 54700.000000 ms to retrieve 1 items.

That’s a full minute to retrieve one database item.

How much memory is in the unit?

16GB… and 99% of the time the CPU is practically sleeping…

2019-01-23%2019_23_15-Microsoft%20Edge

Baffling no?? lol. This is so strange!

Actually, you just answered it.

The J3455 CPU can only address 8GB.

You saw one of those “16GB in your Synology NAS” things, didn’t you?

Remove one of the 8GB sticks.

Yes, you’re correct… but then what I will “see”? My NAS is extremely responsive… I get massive throughput on my GB network (115Mbps on file copies)… I can stream multiple movies simultaneously to multiple devices in my home. Everything works like a champ and I see no CPU hits at all… It even runs two Virtual Windows 2012 Servers, and an ActiveSysnc backup server and none of it has a problem. No performance issues at all.

My only issue is transcoding TrueHD via Plex so I’m not sure removing RAM will do anything for me…

What am I missing?

My Synology DS1815+ has 4x 1GB and with 8GB of RAM and will deliver a composite 468 MB/sec.

Using 4 NIC’s that’s right… so using only one is 115-120 which is what I’m getting without skipping a beat.

I Really don’t think the RAM is my issue at this point. Are you sure there isn’t something else in the logs?

As of now, I have 67,006 files and 25,038 directories in the Plex-Media share.

Right. all 4 nics, bonded into LACP adapter. It will saturate the wire when sending. Synology has always been good at that .

OK ok the MaxWatches value of 32768. That should hold you a while.

Were you able to resolve the DNS lookup error (Gateway / DNS) ?

It also was in your logs. It couldn’t find Plex.tv to get codecs.

OK that’s REALLY strange… I’m logged in my plex pass so the NAS has to resolve DNS… and it auto downloads DSM updates so I know it gets out to the web…

How else can I test?

Quick screenshot of nslookup from a console session…

2019-01-23%2020_17_02-Downloads

For now, Would you consider turning off Autotagging of your photos?

I’m wondering if we’re running into a resource contention problem.
(tagging a ton of photos at once will drive it nuts and that’s what it appears to be attempting.

This, at midnight, bothers me

Jan 23, 2019 00:03:51.355 [0x7fd65fe23700] WARN - Held transaction for too long (../Library/MetadataCollection.cpp:236): 0.150000 seconds
Jan 23, 2019 00:04:03.093 [0x7fd661111700] WARN - AutoTagging: /volume1/Plex-Media/Our Pictures/Vacations/Charelston/2016/SouthCarolina2016-172-Edit-Edit.tif is over maximum size of 104857600 bytes, skipping
Jan 23, 2019 00:05:47.272 [0x7fd661111700] WARN - SLOW QUERY: It took 114490.000000 ms to retrieve 1 items.
Jan 23, 2019 00:05:55.938 [0x7fd6599a1700] WARN - AutoTagging: /volume1/Plex-Media/Our Pictures/Household/2435 Black Forest Drive/July 30, 1999/04_01.JPG is below the minimum pixel count (307200), skipping
Jan 23, 2019 00:05:55.948 [0x7fd6599a1700] WARN - AutoTagging: /volume1/Plex-Media/Our Pictures/Household/2435 Black Forest Drive/August 18, 1999/04_01.JPG is below the minimum pixel count (307200), skipping
Jan 23, 2019 00:05:55.952 [0x7fd6599a1700] WARN - AutoTagging: /volume1/Plex-Media/Our Pictures/Household/2435 Black Forest Drive/July 30, 1999/05_02.JPG is below the minimum pixel count (307200), skipping
Jan 23, 2019 00:05:55.956 [0x7fd6599a1700] WARN - AutoTagging: /volume1/Plex-Media/Our Pictures/Household/2435 Black Forest Drive/August 18, 1999/05_02.JPG is below the minimum pixel count (307200), skipping

The DB was busy for almost 2 minutes?

Also, may I have an updated set of logs?