Multiple repeated errors in the console

[MattDale]

Server Version#: 1.42.2.10156
Player Version#: N/A

I’m not sure how long these have been occurring as I don’t monitor the logs but when trying to uncover some downloading issues I have come across some troubling errors and while I’ve found threads with the same log entries, none of the solutions work for me.

ERROR - [Notify] Failed to add watch for “/mnt/media/video/films” (28: No space left on device)

A line like this for every media directory when there is plenty of space on all devices. They’re on a NAS.

Host is Ubuntu 24.05. Not using docker. Total number of directories are less than 4000 so definitely not an inotify problem.

ERROR - [Req#7d4a] Unknown metadata type: folder

Loads of these. I think maybe benign but it shouldn’t be spamming the logs

[Req#7ce6/Transcode/f905e98d-e9e4-4c89-91a7-e2cc89782500/4e56b718-84ca-4a16-82dc-f08cdcfe5bb7] [eac3_eae @ 0x7ac223e5b4c0] EAE timeout! EAE not running, or wrong folder? Could not read ‘/tmp/pms-43a06417-9cc9-4612-b108-140e90bf44ce/EasyAudioEncoder/Convert to WAV (to 8ch or less)/f905e98d-e9e4-4c89-91a7-e2cc89782500_2726350-0-145.wav’
ERROR - [Req#7df0/Transcode/f905e98d-e9e4-4c89-91a7-e2cc89782500/4e56b718-84ca-4a16-82dc-f08cdcfe5bb7] [eac3_eae @ 0x7ac223e5b4c0] error reading output: -5 (I/O error)
ERROR - [Req#7df1/Transcode/f905e98d-e9e4-4c89-91a7-e2cc89782500/4e56b718-84ca-4a16-82dc-f08cdcfe5bb7] Error while decoding stream #0:1: I/O error

Again being spammed by this. No transcoding is happening. I’ve run the DBRepair.sh and have deleted the EAE codec but neither made a difference.

I’m at a bit of a loose end to where to go. While I haven’t noticed any specific issues it can’t be good to have your log files spammed.

[ChuckPa]

See if this applies ?

Linux Tips

Increase the number of directories Linux can monitor (inotify) for PMS The default size of the Linux inotify table is 8192 directories (*1). When PMS is configured to update automatically when media is added, with properly organized media, especially music, this table rapidly fills and exceeds the default allocation requiring you to increase it. The solution to this is straight forward. Increasing the table size is easy. The first step is to determine how many directories to allocat…

I will also ask if the transcoder temp directory is on a network share ?

[MattDale]

It does not apply. I have less than 4000 directories to monitor and with the latest kernel on ubuntu it has 64K default.

The transcoder temp directory is /dev/shm so not on a share

[dbirch]

@MattDale

I have less than 4000 directories to monitor

But what is the complete size on the drive? That table is no just for plex media.

[MattDale]

On that share it is only media. On the whole system the count of directories is 272841 however I don’t expect they all need monitored. Is there a command to check this?

[dbirch]

@MattDale

I don’t expect they all need monitored.

the inotify table is a OS feature, not something in plex. Its an OS feature that provides notifications to apps that are watching specific directories. However the table is part of the OS kernel.

See Synology FAQ - Questions, Answers, and How-To's - #22 by ChuckPa

[ChuckPa]

Server DEBUG logs please which capture this happening?

Best logs are those which capture server startup and an attempted transcode which fails.

[MattDale]

Plex Media Server.log (1.6 MB)

This is my latest log with all the mentioned errors

[MattDale]

Would the inotify table need to be increased on the NAS or it only the host OS that matters for this?

[ChuckPa]

Thanks.

This confirms the table is full

Jan 06, 2026 00:56:19.226 [127145329314616] ERROR - [Notify] Failed to add watch for "/mnt/media/video/films" (28: No space left on device)
Jan 06, 2026 00:56:19.230 [127145376467768] WARN - JobManager: Could not find job for handle 2720335
Jan 06, 2026 00:56:20.054 [127145329314616] ERROR - [Notify] Failed to add watch for "/mnt/media/video/films/1917 (2019)" (28: No space left on device)
Jan 06, 2026 00:56:20.081 [127145329314616] ERROR - [Notify] Failed to add watch for "/mnt/media/video/films/A Scanner Darkly (2006)" (28: No space left on device)
Jan 06, 2026 00:56:20.082 [127145329314616] ERROR - [Notify] Failed to add watch for "/mnt/media/video/films/Airplane! (1980)" (28: No space left on device)
Jan 06, 2026 00:56:20.083 [127145329314616] ERROR - [Notify] Failed to add watch for "/mnt/media/video/films/Alita - Battle Angel (2019)" (28: No space left on device)
Jan 06, 2026 00:56:20.089 [127145329314616] ERROR - [Notify] Failed to add watch for "/mnt/media/video/films/Almost Human (2019)" (28: No space left on device)
Jan 06, 2026 00:56:20.104 [127145329314616] ERROR - [Notify] Failed to add watch for "/mnt/media/video/films/Almost Human (2019)/Subs" (28: No space left on device)
Jan 06, 2026 00:56:20.105 [127145329314616] ERROR - [Notify] Failed to add watch for "/mnt/media/video/films/An Inspector Calls (2015)" (28: No space left on device)
Jan 06, 2026 00:56:20.106 [127145329314616] ERROR - [Notify] Failed to add watch for "/mnt/media/video/films/Animal Farm (1999)" (28: No space left on device)
Jan 06, 2026 00:56:20.107 [127145329314616] ERROR - [Notify] Failed to add watch for "/mnt/media/video/films/Apex (2021)" (28: No space left on device)
Jan 06, 2026 00:56:20.108 [127145329314616] ERROR - [Notify] Failed to add watch for "/mnt/media/video/films/Automata (2014)" (28: No space left on device)
Jan 06, 2026 00:56:20.109 [127145329314616] ERROR - [Notify] Failed to add watch for "/mnt/media/video/films/BIOS (2021)" (28: No space left on device)
Jan 06, 2026 00:56:20.110 [127145329314616] ERROR - [Notify] Failed to add watch for "/mnt/media/video/films/Battlefield 2025 (2020)" (28: No space left on device)

You say 4000 directories, and Ubuntu has a 64K default.

The errors indicate one of those values is incorrect.

Confirm the inotify table size please

sudo sysctl -a | grep max_user_watches

Might you have an error in the list of directories you’ve added to your libraries ?

[ChuckPa]

@MattDale

Would the inotify table need to be increased on the NAS or it only the host OS that matters for this?

The inotify table is populated on the host which runs PMS.

PMS uses the kernel’s filesystem notification mechanism to detect changes
This is how “Automatically Update my library when changes are detected” works.

For details on inotify, type: man inotify
The built in electronic help pages will tell you all about it.

[MattDale]

fs.epoll.max_user_watches = 7237431
fs.inotify.max_user_watches = 65536

I ran

find /mnt/media/ -type d | wc -l

Which results in 3851

[MattDale]

It looks like this was in fact the solution. I increased that setting to 128K and then restarted PMS and it appears to have stopped that error. There are others now but the server is now functional.

This error still remains

Unknown metadata type: folder

[dbirch]

@MattDale

Unknown metadata type: folder

you can safely ignore that, its a bit of debug info that the devs have left in the logs, doesnt really mean there is an error.

[MattDale]

For anyone else coming to this, I found the following very useful

GitHub - mikesart/inotify-info: Linux inotify info reporting app

It revealed that the maximum watches had in fact been breached and the culprit in this case was Jellyfin rather than Plex although of course it could affect any other processes too

[ChuckPa]

To add,

This will happen when multiple apps (servers) set inotify watches.
The table is in the kernel.
Apps do mot share slots. Each app has their own entry.