Issues with plex playback (Synology)

[turb3k]

Server Version#:1.20.4.3517-ab5e1197c
Player Version#:1.23.0.1435-e10e14cf

Hello, I’m having issues with my Plex Media Server. I keep getting “EAE timeout! EAE not running, or wrong folder? Could not read” Errors when people are attempting to playback media. If i restart the server it seems to fix the issue for a while (a few hours to days). However, it eventually comes back and starts having issues again. I have tried:-

• creating a new Tmp folder and directing plex there
• Making sure Plex has full control and ownership of all folders and files in the plex directory
• Rolling back the plex version
• Making sure plex is the latest version
• deleting the codecs folder
• uninstalling plex and reinstalling

I don’t think it’s an issue I have read about with “the number of folders limit” if that were the case why would it work when I reboot the server for a while?
I have had the server for around 4/5 years and never had this issue until the last month, if anyone could help me i would be really happy as its beyond my ability to fix now.

Thanks in advance.

Russ

[turb3k]

I have attached the logs, however, i have rebooted the server in the last hour to correct the issue so don’t know how far these logs will go back?

[turb3k]

@trumpy81

Something is being played and transcoded successfully.

Yes, after I restarted the server I became able to play the media file, however, before the restart I was unable to.

@trumpy81

This normally indicates that you have too many folders being watched by Plex. The iNotify table is not big enough to hold them all.

See this article to correct that:

Ok, so i need to figure out the amount of media folders on the entire system or just within plex itself?

@trumpy81

Plex is unable to locate some files.

Yes, this is just a disconnected USB drive from the Synology unit.

@trumpy81

To hopefully, correct the EAE issue, do the following:

1. Stop Plex from running.
2. Open Control Panel > Shared Folder and edit the Plex folder. On the permission tab, user plex should have custom or R/W permissions, then click on OK.
3. Open File Station and right click on /Plex/tmp_transcoding and select Properties. Check that the owner of the folder is user plex. On the General tab, enable Apply to this folder, sub-folders and files then click on OK.
4. Right click on Plex/tmp_transcoding again and select Properties. On the Permission tab, enable Apply to this folder, sub-folders, and files then click on OK.

I did manage to find this information in another threat and did action these instructions.
However, it has not seemed to fix the issue with the correct EAE

[turb3k]

@trumpy81

This normally indicates that you have too many folders being watched by Plex. The iNotify table is not big enough to hold them all.

See this article to correct that:

Plex is saved to volume 1 on my system and my media is saved to volume 2
All the file and folder info i can find is below, with as much detail as i can give.

The total number of folders plex is watching :-

Video folder (Volume 2)

• Folders : 10414
• Files including metadata & jpeg etc : 174746
• Size : 47.68TB

Plex Folder (Volume 1)

• Folders : 756141
• Files including plex’s metadata & images : 799306

Music Folder (Volume 2)

• Folders :1341
• Files : 12475

There are other folders on the system, however plex is not pointed at them currently. Do i need the file and folder information from them also?

Thanks so much for you’re help with this.

[turb3k]

Well, this is about the maxium level of my ability now.
From reading the article i understand it as this:

1. First, get the total number of directories in use for your media (movies, series, music, and photos). For each share containing media:
   a. File Station.
   b. Left Pane
   c. Right-click the share
   d. Properties
   e. Write down the number of folders it finds.

Video folders = 10414, Music Folders = 1341 Total = 11,755
Ignore the plex folder

a. Add 32768 to the value = 44,523
b. Divide by 32768 = 1.358734130859375 (1.35)
c. Add 1 to the the Integer result (No fractions here) = 2.35
e. Multiply by 32768 = 77004.8
f. Use this final result. (integer part only) = 77004.8 or 77004?
g. If the number of folders you are using is less than 8192, stop here unless you are about to add a lot more media. = The number is higher.

Create a Scheduled task you will run ONE TIME -OR- when DSM updates and you need to increase the table again. = I need to do this as my number is higher?

Remove any previous s2s_watches_max entries (avoid clutter)

sed -i ‘s/^s2s_watches_max=.*//' /etc/synoinfo.conf sed -i 's/^s2s_watches_max=.*//’ /etc.defaults/synoinfo.conf

This is an example only. Use your actual result below to add the new value

echo ‘s2s_watches_max=77004’ >> /etc/synoinfo.conf
echo ‘s2s_watches_max=77004’ >> /etc.defaults/synoinfo.conf

or

Remove any previous s2s_watches_max entries (avoid clutter)

sed -i ‘s/^s2s_watches_max=.*//' /etc/synoinfo.conf sed -i 's/^s2s_watches_max=.*//’ /etc.defaults/synoinfo.conf

This is an example only. Use your actual result below to add the new value

echo ‘s2s_watches_max=770048’ >> /etc/synoinfo.conf
echo ‘s2s_watches_max=770048’ >> /etc.defaults/synoinfo.conf

or

Remove any previous s2s_watches_max entries (avoid clutter)

sed -i ‘s/^s2s_watches_max=.*//' /etc/synoinfo.conf sed -i 's/^s2s_watches_max=.*//’ /etc.defaults/synoinfo.conf

This is an example only. Use your actual result below to add the new value

echo ‘s2s_watches_max=77004.8’ >> /etc/synoinfo.conf
echo ‘s2s_watches_max=77004.8’ >> /etc.defaults/synoinfo.conf

Could you assist with this number please so i can be sure to enter the correct one?

Thank you

[ChuckPa]

@turb3k

Video folder (Volume 2)

• Folders : 10414
• Files including metadata & jpeg etc : 174746
• Size : 47.68TB

@turb3k

Music Folder (Volume 2)

• Folders :1341
• Files : 12475

If I may be of assistance?

1. When you opened Filestation and obtained the folder counts for your media folders (don’t worry about Plex’s folders – they don’t count)

2. You got ?
   a. 10414 (for video)
   b. 1341 (for music)

11755 folders ?

3. To apply that:
   a. 11755 / 32768 = 0.3587, now discard the fractional part → 0
   b. 0 + 1 = 1
   c. 1 * 32768 = 32768

32768 is the value we will use

4. now create the task as I show in the How-To using your values. Your script’s contents will be this:

# Remove any previous s2s_watches_max entries (avoid clutter)
sed -i 's/^s2s_watches_max=.*$//' /etc/synoinfo.conf
sed -i 's/^s2s_watches_max=.*$//' /etc.defaults/synoinfo.conf

# Custom calculated for up to 32768 folders
echo 's2s_watches_max=32768' >> /etc/synoinfo.conf
echo 's2s_watches_max=32768' >> /etc.defaults/synoinfo.conf

After running this script (manual “RUN” button), Restart the NAS

[turb3k]

@turb3k

77004

Hello, I have now done this however i didn’t apply the
now discard the fractional part → 0
0 + 1 = 1

So used the number 77004, will this have a negative effect? i have restarted the system after running the command and the red text in console about not enough inotify space has stopped on boot up and now reports that notify is now watching the folders that had been showing up as red due to not enough space.

Thank you

[ChuckPa]

Screenshot please? Nothing should be in red.

May I ask where 77004 came from?
Did I miss something?

[turb3k]

I can’t take a screen shot as it seems to be fixed and working now but i was seeing red messages in the console on start up that refered to notify. Something along the lines of “notify unable to watch folder due to insufficient space. (then the name of a folder)” But after applying the :

echo ‘s2s_watches_max=77004’ >> /etc/synoinfo.conf
echo ‘s2s_watches_max=77004’ >> /etc.defaults/synoinfo.conf

It reports that “notify is watching XYZ folder” which had previously reported as unable to be watched.

The 77004 number came from my bad maths, will it cause any problems being set higher than 32768?

Thank you.

[ChuckPa]

how much memory do you have? 1GB / 2 GB ?

(Control Panel - Info)

[turb3k]

Its a DS3617xs, I think that comes with 16gb of RAM standard, i have not upgraded it yet to 32gb

[ChuckPa]

then 77004 would not be an issue.
However, the fraction 77004.8 you showed above would probably drive it crazy and it sounds like it did! lol

[ChuckPa]

this time, about 2 minutes after PMS starts, download the logs ZIP please

Attach that here.

I can quick look and see what’s going on.

[turb3k]

Its ok i didnt use the fraction, i only used 77004 as the number. haha

Ok will create a new log file now. 1min

[turb3k]

(File removed)

New log attached

[ChuckPa]

I asked about the fraction because you had posted this:

@turb3k

echo ‘s2s_watches_max=77004.8’ >> /etc/synoinfo.conf
echo ‘s2s_watches_max=77004.8’ >> /etc.defaults/synoinfo.conf

As for a screenshot tool. This works very well on Windows.

Greenshot

Did you restart the Synology after running the Scheduled Task?

[turb3k]

Yes, I did restart the entire system after running the task.
Sorry, I didn’t mean to post the 77004.8 in there.

Do those logs still show an issue?

[ChuckPa]

The logs you provided do not show PMS having restarted.

PMS will restart when the Syno restarts.

1. Synology will increase the table sizes at system startup only.
2. When PMS starts , it will be able to benefit from the increased table size.
3. Fresh PMS logs, downloaded 2 minutes after system restart, will confirm everything is as it should be

[turb3k]

@ChuckPa

The logs you provided do not show PMS having restarted.

I restarted the entire Synology unit and then waited 2mins then downloaded logs.

@ChuckPa

When PMS starts , it will be able to benefit from the increased table size.

Do I need to run the script via the Task Scheduler on every system restart?

@ChuckPa

Fresh PMS logs, downloaded 2 minutes after system restart, will confirm everything is as it should be

Can i just restart the Package and wait 2mins or do i need to restart the whole Nas drive again?
It’s strange as i restarted the Nas completely then let the PMS package restart then waited then downloaded the logs

[ChuckPa]

You must restart the whole NAS.

The table values you changed are the Operating System (DSM) itself.

PMS relies on the additional table size provided by the OS. This is how Linux works.

PS: My Syno restarts in 4 minutes