[Support] binhex - Plex


Recommended Posts

17 minutes ago, jebusfreek666 said:

In fairness though, I followed the steps in Q3 of the FAQ you have posted and this was not mentioned. You might want to add it to avoid more of the same questions from popping up.

thanks, added in.

Link to comment
Posted (edited)

For some reason I can't get my my binhex-plex docker to update.

 

It says it's updated successfully, but then always immediately shows an update is available again. Also doesn't seem to be updating the version of plex, as plex itself is still reporting there's an update available.

 

Other containers (linuxserver, etc.) are updating appropriately, as are other containers from binhex (binhex-sonarr, binhex-radarr). Any troubleshooting tips?

 

 

Screen Shot 2021-05-07 at 2.24.29 PM.png

 

Screen Shot 2021-05-07 at 2.24.59 PM.png

Edited by dgallaher
Link to comment
  • 4 weeks later...
Posted (edited)

Hi all,

 

I'm having an issue where overnight sometimes plex will become unresponsive in the morning, i.e. TV, phone won't connect to server and webUI won't load (Error Connection time out). It fixes itself temporarily with a restart of the container.

 

In the container log, this repeats a few times, so it seems to be trying to repeatedly get thumbnails for 2 old show episodes I've had for a long time:

 

---

2021-05-29 02:26:35,175 DEBG 'plexmediaserver' stdout output:
Generated new chapter thumbnails for Wicks.

2021-05-29 02:26:35,586 DEBG 'plexmediaserver' stdout output:
Generated new chapter thumbnails for Behold a Pale Rider.

2021-05-29 02:41:23,123 DEBG fd 8 closed, stopped monitoring <POutputDispatcher at 22907605937744 for <Subprocess at 22907605937648 with name plexmediaserver in state RUNNING> (stdout)>
2021-05-29 02:41:23,177 DEBG fd 12 closed, stopped monitoring <POutputDispatcher at 22907605709968 for <Subprocess at 22907605937648 with name plexmediaserver in state RUNNING> (stderr)>
2021-05-29 02:41:23,177 INFO exited: plexmediaserver (terminated by SIGKILL; not expected)
2021-05-29 02:41:23,178 DEBG received SIGCHLD indicating a child quit
2021-05-29 02:41:23,426 INFO spawned: 'plexmediaserver' with pid 9091
2021-05-29 02:41:24,427 INFO success: plexmediaserver entered RUNNING state, process has stayed up for > than 1 seconds (startsecs)
2021-05-29 02:41:42,244 DEBG 'plexmediaserver' stdout output:
Critical: libusb_init failed

 

---

 

But this didn't seem critical to me so I looked at the Plex Media Server.log and the only Errors are below:

 

---

 

May 30, 2021 03:05:38.306 [0x152fe9715b38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 03:06:31.309 [0x152fe906cb38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 03:06:31.309 [0x152fe906cb38] ERROR - SSDP: Error parsing device schema for http://192.168.0.16:9080
May 30, 2021 04:05:38.672 [0x152fe9a0fb38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 04:40:11.641 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:11.741 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:11.841 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:11.941 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:12.041 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:12.141 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:12.242 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:13.625 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:13.725 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:13.826 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:13.926 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:14.026 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:14.126 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:14.227 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:18.811 [0x152fe9643b38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 05:05:38.833 [0x152fe9a0fb38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 06:05:38.979 [0x152fe9643b38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 07:05:39.190 [0x152fe9a0fb38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 08:05:39.488 [0x152fe9a0fb38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 08:48:39.895 [0x152fe906cb38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 08:48:39.895 [0x152fe906cb38] ERROR - SSDP: Error parsing device schema for http://192.168.0.3:2870/dmr.xml
 

---

 

Nothing jumping out at me as an obvious problem. Not getting database corruption notifications on my plex phone app as I have had in the past but fixed. 

 

Any ideas?

 

Edit: Found this in the unRAID log:

 

May 30 03:05:27 BlackBox kernel: Call Trace: May 30 03:05:27 BlackBox kernel: out_of_memory+0x3dd/0x410 May 30 03:05:27 BlackBox kernel: Out of memory: Killed process 9564 (Plex Media Serv) total-vm:6431820kB, anon-rss:3845312kB, file-rss:0kB, shmem-rss:4kB, UID:99 pgtables:7788kB oom_score_adj:0

 

Is this the culprit? I recently turned DLNA on for no good reason, which apparently can cause memory leak. I've turned it off now. 

Edited by xMaverickx
Link to comment
Hi all,
 
I'm having an issue where overnight sometimes plex will become unresponsive in the morning, i.e. TV, phone won't connect to server and webUI won't load (Error Connection time out). It fixes itself temporarily with a restart of the container.
 
In the container log, this repeats a few times, so it seems to be trying to repeatedly get thumbnails for 2 old show episodes I've had for a long time:
 
---
2021-05-29 02:26:35,175 DEBG 'plexmediaserver' stdout output:
Generated new chapter thumbnails for Wicks.

2021-05-29 02:26:35,586 DEBG 'plexmediaserver' stdout output:
Generated new chapter thumbnails for Behold a Pale Rider.

2021-05-29 02:41:23,123 DEBG fd 8 closed, stopped monitoring (stdout)>
2021-05-29 02:41:23,177 DEBG fd 12 closed, stopped monitoring (stderr)>
2021-05-29 02:41:23,177 INFO exited: plexmediaserver (terminated by SIGKILL; not expected)
2021-05-29 02:41:23,178 DEBG received SIGCHLD indicating a child quit
2021-05-29 02:41:23,426 INFO spawned: 'plexmediaserver' with pid 9091
2021-05-29 02:41:24,427 INFO success: plexmediaserver entered RUNNING state, process has stayed up for > than 1 seconds (startsecs)
2021-05-29 02:41:42,244 DEBG 'plexmediaserver' stdout output:
Critical: libusb_init failed
 
---
 
But this didn't seem critical to me so I looked at the Plex Media Server.log and the only Errors are below:
 
---
 
May 30, 2021 03:05:38.306 [0x152fe9715b38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 03:06:31.309 [0x152fe906cb38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 03:06:31.309 [0x152fe906cb38] ERROR - SSDP: Error parsing device schema for http://192.168.0.16:9080
May 30, 2021 04:05:38.672 [0x152fe9a0fb38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 04:40:11.641 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:11.741 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:11.841 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:11.941 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:12.041 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:12.141 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:12.242 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:13.625 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:13.725 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:13.826 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:13.926 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:14.026 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:14.126 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:14.227 [0x152fe906cb38] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
May 30, 2021 04:40:18.811 [0x152fe9643b38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 05:05:38.833 [0x152fe9a0fb38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 06:05:38.979 [0x152fe9643b38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 07:05:39.190 [0x152fe9a0fb38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 08:05:39.488 [0x152fe9a0fb38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 08:48:39.895 [0x152fe906cb38] ERROR - Error issuing curl_easy_perform(handle): 28
May 30, 2021 08:48:39.895 [0x152fe906cb38] ERROR - SSDP: Error parsing device schema for http://192.168.0.3:2870/dmr.xml
 
---
 
Nothing jumping out at me as an obvious problem. Not getting database corruption notifications on my plex phone app as I have had in the past but fixed. 
 
Any ideas?
 
Edit: Found this in the unRAID log:
 
May 30 03:05:27 BlackBox kernel: Call Trace: May 30 03:05:27 BlackBox kernel: out_of_memory+0x3dd/0x410 May 30 03:05:27 BlackBox kernel: Out of memory: Killed process 9564 (Plex Media Serv) total-vm:6431820kB, anon-rss:3845312kB, file-rss:0kB, shmem-rss:4kB, UID:99 pgtables:7788kB oom_score_adj:0
 
Is this the culprit? I recently turned DLNA on for no good reason, which apparently can cause memory leak. I've turned it off now. 
I would say there is a very good chance that out of memory killer (OOM Killer) is your culprit, stop any memory hungry containers and see how you go

Sent from my CLT-L09 using Tapatalk

Link to comment
Posted (edited)
On 5/30/2021 at 6:19 PM, binhex said:

I would say there is a very good chance that out of memory killer (OOM Killer) is your culprit, stop any memory hungry containers and see how you go

Sent from my CLT-L09 using Tapatalk
 

 

Thanks for replying. I think we are on the right track but it happened again overnight and I got a fix common problems warning this time (OOM), so DLNA not the issue. Always happens at 3am so I think it must be when Plex runs it's scheduled tasks? unRAID Mover runs at 2200. As a bit more background I only have 8Gb RAM (just ordered 8 more), but my system is pretty slim. No VMs, and 12 containers (the usual stuff; Plex, the arrs, duckdns, krusader, delugeVPN, etc). Nothing CURRENTLY using more than 0.03% of RAM. I do have plex hardware transcoding set up to /dev/dri, but I don't believe that would have been running at 3am. I've copied the warnings and errors from overnight below and attached my diagnostics file if that helps. Again, the plex log says it was trying to generate the same thumbnails for 2 old episodes. I wonder if there is an error with this. I've moved the episodes out of the library for now.

 

Do I just need more RAM? Would have thought 8Gb would be adequate, though not beefy.

Do I limit the amount of RAM plex can access or does this make it worse?

 

Thank you!

 

Jun 1 02:20:44 BlackBox kernel: CPU: 6 PID: 29792 Comm: unraid-api Not tainted 5.10.28-Unraid #1

Jun 1 02:20:44 BlackBox kernel: Call Trace: Jun 1 02:20:44 BlackBox kernel: out_of_memory+0x3dd/0x410

Jun 1 02:20:44 BlackBox kernel: oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0,global_oom,task_memcg=/docker/7e8e5c060ae6100a95183b3624127cca31f52022f073c4fcd95231f9cb60bcd9,task=Plex Media Serv,pid=6112,uid=99

Jun 1 02:20:44 BlackBox kernel: Out of memory: Killed process 6112 (Plex Media Serv) total-vm:6483760kB, anon-rss:3703864kB, file-rss:0kB, shmem-rss:4kB, UID:99 pgtables:7528kB oom_score_adj:0

Jun 1 02:41:49 BlackBox kernel: CPU: 5 PID: 29792 Comm: unraid-api Not tainted 5.10.28-Unraid #1

Jun 1 02:41:49 BlackBox kernel: Call Trace: Jun 1 02:41:49 BlackBox kernel: out_of_memory+0x3dd/0x410

Jun 1 02:41:49 BlackBox kernel: oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0,global_oom,task_memcg=/docker/7e8e5c060ae6100a95183b3624127cca31f52022f073c4fcd95231f9cb60bcd9,task=Plex Media Serv,pid=7547,uid=99

Jun 1 02:41:49 BlackBox kernel: Out of memory: Killed process 7547 (Plex Media Serv) total-vm:6427644kB, anon-rss:3788660kB, file-rss:0kB, shmem-rss:4kB, UID:99 pgtables:7644kB oom_score_adj:0

Jun 1 03:02:51 BlackBox kernel: CPU: 3 PID: 20694 Comm: Plex Media Serv Not tainted 5.10.28-Unraid #1

Jun 1 03:02:51 BlackBox kernel: Call Trace: Jun 1 03:02:51 BlackBox kernel: out_of_memory+0x3dd/0x410

Jun 1 03:02:51 BlackBox kernel: oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=7e8e5c060ae6100a95183b3624127cca31f52022f073c4fcd95231f9cb60bcd9,mems_allowed=0,global_oom,task_memcg=/docker/7e8e5c060ae6100a95183b3624127cca31f52022f073c4fcd95231f9cb60bcd9,task=Plex Media Serv,pid=22324,uid=99

Jun 1 03:02:51 BlackBox kernel: Out of memory: Killed process 22324 (Plex Media Serv) total-vm:6420672kB, anon-rss:3782076kB, file-rss:0kB, shmem-rss:4kB, UID:99 pgtables:7700kB oom_score_adj:0

Jun 1 04:21:03 BlackBox root: Fix Common Problems: Error: Out Of Memory errors detected on your server

blackbox-diagnostics-20210601-0902.zip

Edited by xMaverickx
Link to comment
8 hours ago, xMaverickx said:

Jun 1 02:20:44 BlackBox kernel: Out of memory: Killed process 6112 (Plex Media Serv) total-vm:6483760kB,

what this is telling you is that process 'Plex Media Serv' was consuming 6.4 GB of RAM, now if you only have 8GB then that is going to be a problem, thus OOM killer kicks in and kills Plex, question is, why is it consuming that amount of memory, is this normal or this is unusual?, is it performing a memory intensive background process at 3am, perhaps database cleanup or something similar, you would need to trawl your Plex log to find out what it does at around 3am that could cause this high memory usage.

Link to comment
  • 3 weeks later...

I have been thinking about setting my disks to spin down after a certain amount of time. I realize that I will have to turn off periodic checks in plex as this will keep the disks spinning. My question is, I had read that plex will kick out errors if the disk that the media is on has to spin up. I don't know if this is an old issue or something that has been resolved some way in recent builds. Is this still the case? I have like 10 data disks and my media is spread across all of them as this is primarily a plex server. I will probably have to skip doing spin downs if my plex users are going to get errors all the time. 

Link to comment
1 hour ago, jebusfreek666 said:

I have been thinking about setting my disks to spin down after a certain amount of time. I realize that I will have to turn off periodic checks in plex as this will keep the disks spinning. My question is, I had read that plex will kick out errors if the disk that the media is on has to spin up. I don't know if this is an old issue or something that has been resolved some way in recent builds. Is this still the case? I have like 10 data disks and my media is spread across all of them as this is primarily a plex server. I will probably have to skip doing spin downs if my plex users are going to get errors all the time. 

I can only speak to my recent experience with plex/unraid on a new build, but I have not had any issues for users with disks having to spin up other than maybe a few extra seconds load time. Basically nothing the user has noticed. My disclaimer is I've only been running this setup for a bit over a week so issues may yet arise, but I tested the loading as a specific curiosity of mine before I moved my content over to this server en-masse.

 

As for scanning new files, I disabled the periodic checks and either manually start a scan if I have added new content to a library, or I use Sonarr/Radarr to update the library itself when they add new content as they will be my main source moving forward. They seem much more efficient at just telling Plex "here's this new thing I just added for you" rather than Plex having to rescan everything. Depending on how long your scans take, and your spindown time, I imagine this should still be achievable, however I would be wary of the "monitor for changes" setting as I can only imagine how this would behave in an unraid world (I haven't tested, purely a thought exercise says there is more room for error with that one without understanding how the fuse filesystem works in more detail).

Link to comment
3 hours ago, hoppers99 said:

I use Sonarr/Radarr to update the library itself when they add new content as they will be my main source moving forward. They seem much more efficient at just telling Plex "here's this new thing I just added for you" rather than Plex having to rescan everything. 

 

Where is the setting in Sonarr/Radarr to have it make plex update its library?

Link to comment
5 minutes ago, jebusfreek666 said:

Where is the setting in Sonarr/Radarr to have it make plex update its library?

 

Have a look for "Connection Settings" under settings. Similar to connecting your Indexer and Download Client, this area will give you options to connect Plex in a similar way. My setup currently has Radarr/Sonarr successfully communicate the requirement for a file to the Download Client, and then moves it into the correct location for Plex to see it when the download is complete, then updates Plex to make it available. If you get stuck I can confirm more specific settings/setup when I'm back home this evening but there were a number of good guides around google that I used to get mine setup.

  • Thanks 1
Link to comment

--Plex quicksync hw encoding for library--

 

So I just built a new server with a 11600k and trying to optimize my library to 1080p.  I can hw transcode 4k to 1080p in a stream but I wanted to optimize the whole library anyway for others.   I can't seem to force plex to transcode using quicksync. Its sits on the cpu doing 1.5-1.7x speed.  I just asked someone on reddit to run a 4k file to 1080p and he's saying on his 8600k he's doing 20x+.  What am I doing wrong here? What do I need to setup.

 

Or does anyone know of another program that will batch encode my library using quicksync?

Link to comment

Hoping you guys could help me with this issue I am having. Recently whenever I watch a video that comes from a large file my entire server locks up after a little bit and I have to restart the Plex container. What I have come to notice is that basically Plex is filling my ram with whatever the file is I am watching as I am watching. For example if I am watching a 40GB BluRay movie that is 90 minutes, after 90 minutes my Plex container will be using 40GB, issue is I only have 32GB of ram in my server and I cant even give it all of that amount. When I stop watching the movie the ram is given back.

 

Some notes that might help: DLNA is off Everything is generally direct play (exception of audio transcoding) I have updated to the newest verion of PMS When I stop watching the video all of the memory is given back.

 

This is after watching Top Gun for 30 minutes:

image.png.61eed6efda827447ca3078eab0489c16.png

 

Here is the config if that helps:

image.thumb.png.1fcf03d615da06ade9190beaff7610c6.png

Let me know if I can pull any other logs that would help with this

 

Any help is appreciated.

Link to comment

I have been experiencing a lot of Plex just crapping out on me.  Works totally fine for a few hours and then it just disconnects my users and no one can get back in.  No errors in unraid, no errors in the plex logs.  I have to restart plex and it works again for a few hours.  I noticed it starting this a week or two ago.  Happens locally and remote.  I have a really beefy server too so I dont think its that.  I also have high end networking equipment and a fiber connection.  So I think its just an issue with Plex.  Anyone else having issues with it?

Link to comment
I have been experiencing a lot of Plex just crapping out on me.  Works totally fine for a few hours and then it just disconnects my users and no one can get back in.  No errors in unraid, no errors in the plex logs.  I have to restart plex and it works again for a few hours.  I noticed it starting this a week or two ago.  Happens locally and remote.  I have a really beefy server too so I dont think its that.  I also have high end networking equipment and a fiber connection.  So I think its just an issue with Plex.  Anyone else having issues with it?
I use this image myself and other then 1 crash several months ago It had been solid for literally years. If you are doing GPU transcoding then I would encourage you to try different versions of the driver, as it could well be related to that

Sent from my CLT-L09 using Tapatalk

Link to comment
6 hours ago, binhex said:

I use this image myself and other then 1 crash several months ago It had been solid for literally years. If you are doing GPU transcoding then I would encourage you to try different versions of the driver, as it could well be related to that

Sent from my CLT-L09 using Tapatalk
 

I’ll play around with that.  I did just update my gpu driver last night.  Maybe that will help.  

Link to comment
On 6/26/2021 at 4:37 AM, binhex said:

I use this image myself and other then 1 crash several months ago It had been solid for literally years. If you are doing GPU transcoding then I would encourage you to try different versions of the driver, as it could well be related to that

Sent from my CLT-L09 using Tapatalk
 

Tried installing a Nvidia driver update, didn’t help.  Also rebooted a few times and tried without vms running.  Didn’t help.  Anymore ideas?

Link to comment
6 hours ago, sittingmongoose said:

Tried installing a Nvidia driver update, didn’t help.  Also rebooted a few times and tried without vms running.  Didn’t help.  Anymore ideas?

try ruling out gpu transcoding, turn that off and see if you get a stable experience, if it is stable then you need to post on the gpu driver plugin thread, if it isn't stable then you need to post on the plex forum and see if they have any suggestions for you.

Link to comment
Posted (edited)

I had problems with my server and made a clean install of plex. Deleted docker + image and the appdata folder. Now whenever I try to sort my library the webUI has an error and on my fireTV the app just crashes. The Library has 500 shows in it but before i made the clean install it worked without any problems. The dashboard and recommended tabs dont load as well. Is there anything i can do about that?

Edited by Kinji
Link to comment
Posted (edited)

Plex Intro Detection processing causing very high memory usage and OOM

This post is more a headsup for others who might be running into similar issues. The actual problem is inside Plex (Intro Detection processing) and should be fixed there.

 

About a week ago my server started having memory related issues. Plex was being killed by the OOM reaper due to excessive memory usage. I haven't set any memory limits for Plex so the OOM was protecting the whole server from running out of memory. I have 32GB of mem and normal utilisation is ~60%.

 

When analysing the issue through Grafana statistics (TIG stack) I realised the issue had started already on 10.8.2020 and had been slowly creeping up until a week ago it passed the final limit causing the OOM. It also became evident that the problem occurred daily at the time when Plex scheduled maintenance tasks were running. I quite quickly found this thread on Plex forums:

https://forums.plex.tv/t/server-maintenance-leads-to-excessive-resource-utilization/611012/18

 

The root cause in that case was the intro detection processing stumbling on some badly encoded media. So I yesterday temporarily disabled intro detection (and only that) and the problem went completely away as can be seen from the docker memory utilisation graph below (last day missing the mem spikes). And yes, I have a weekly backup/update routine running which also restarts all dockers hence the drop on 3.7.

image.thumb.png.5e1f6f363ff47c928ff80771b3f3526e.png

 

So most likely I have added some bad media on 10.8.2020 since the intro detection feature was released way before that. The referenced thread also included ffmpeg scripts for detecting bad media. I think I will have to go that route too. Unfortunately the Plex logs do not tell anything about problematic files.

 

I will in case up the server memory 32 -> 64GB just to see if it makes a difference. But that is just a band aid. Plex should definitely not be using >15GB of memory.

Edited by henris
  • Like 1
Link to comment
6 hours ago, henris said:

Plex Intro Detection processing causing very high memory usage and OOM

This post is more a headsup for others who might be running into similar issues. The actual problem is inside Plex (Intro Detection processing) and should be fixed there.

 

About a week ago my server started having memory related issues. Plex was being killed by the OOM reaper due to excessive memory usage. I haven't set any memory limits for Plex so the OOM was protecting the whole server from running out of memory. I have 32GB of mem and normal utilisation is ~60%.

 

When analysing the issue through Grafana statistics (TIG stack) I realised the issue had started already on 10.8.2020 and had been slowly creeping up until a week ago it passed the final limit causing the OOM. It also became evident that the problem occurred daily at the time when Plex scheduled maintenance tasks were running. I quite quickly found this thread on Plex forums:

https://forums.plex.tv/t/server-maintenance-leads-to-excessive-resource-utilization/611012/18

 

The root cause in that case was the intro detection processing stumbling on some badly encoded media. So I yesterday temporarily disabled intro detection (and only that) and the problem went completely away as can be seen from the docker memory utilisation graph below (last day missing the mem spikes). And yes, I have a weekly backup/update routine running which also restarts all dockers hence the drop on 3.7.

image.thumb.png.5e1f6f363ff47c928ff80771b3f3526e.png

 

So most likely I have added some bad media on 10.8.2020 since the intro detection feature was released way before that. The referenced thread also included ffmpeg scripts for detecting bad media. I think I will have to go that route too. Unfortunately the Plex logs do not tell anything about problematic files.

 

I will in case up the server memory 32 -> 64GB just to see if it makes a difference. But that is just a band aid. Plex should definitely not be using >15GB of memory.

Sadly this is not the case for me. Iam running only 8GB of ram (since my server is basicly only for plex) and plex never takes much of that. Especially the ram usage doesnt climb up.  But sometimes when i try to load recommended tab or try to sort my library is see 100% cpu usage on all cores. Sometimes it doesnt and crashes as well. I had 16GB of ram for a short time but somehow the stick let my dockers and unraid crash/freeze. After trying only the new stick my hole sever went yolo and my dockers got destroyed. So iam a little on the careful side when adding ram a second time...

Link to comment
7 hours ago, henris said:

So most likely I have added some bad media on 10.8.2020 since the intro detection feature was released way before that. The referenced thread also included ffmpeg scripts for detecting bad media. I think I will have to go that route too. Unfortunately the Plex logs do not tell anything about problematic files.

Did u try that? I had a similar issue where my activity stuck on one show. But i have no idea how to run a custom script.

Link to comment
Posted (edited)
13 minutes ago, Kinji said:

Did u try that? I had a similar issue where my activity stuck on one show. But i have no idea how to run a custom script.

Not yet. Unfortunately I don't have too much time to tinker with the server so I'm just running it with intro detection disabled. Did you follow the link to the original Windows based scripts? I would assume you should be able to use them almost as is and they might be easier than the Linux alternatives.

 

There have been several cases in past 10+ years with Plex suddenly acting up related to resource utilisation. HDR transcoding tone mapping and chapter thumbnail generation come to mind. But this (Unraid + Plex + NVidia Shields) is just like a dream compared to the days of HTPCs.

Edited by henris
Link to comment
On 7/5/2021 at 5:38 PM, Kinji said:

I had problems with my server and made a clean install of plex. Deleted docker + image and the appdata folder. Now whenever I try to sort my library the webUI has an error and on my fireTV the app just crashes. The Library has 500 shows in it but before i made the clean install it worked without any problems. The dashboard and recommended tabs dont load as well. Is there anything i can do about that?

I think i found the reason. I looked at the logs myself (holy you can see my email and username there?!!!) and found this "Audio Stream: 23211, Subtitle Stream: 23212". Seems like whenever i try to load anything plex chooses audio and subtitles for ALL my media which obviously takes long and is useless. I disabled "Automatically select audio and subtitle tracks" under settings > languages and now all loads instantly.

 

Sadly i found another issue. Plex doesnt do a partial scan on change. I tried to google it but did not found anything helpful for me. Do u guys know anything i can try?

Link to comment

I was wondering why my appdata backups were so huge. I looked at my appdata on QDirStat and noticed these two folders were taking up 99% of the space:

591343893_ScreenShot2021-07-06at4_21_49PM.png.831e6956663e57e1ac74487e26f40484.png

 

Is it safe to exclude /Media and /Cache from CA Backup and Restore? From what I understand Cache is just, well, cache. And /Media stores thumbnails which I do actually really like and prefer to keep on. My only concern about what I don't want to lose in a failure and restore situation is: collections, play history, user history, playlists, and custom posters/backgrounds that I've made and uploaded for various items.

Link to comment

Hi All,

 

I know that this has been answered, but I'm still a bit confused.

 

It looks like I'm still having transcoding problems when I'm trying to transcode to RAM.  I did follow @binhex FAQ on transcoding to RAM:

 

image.png.374ac863573f666e2cb1d424a7d75ec1.png

 

image.png.e30e6a46215780b8044c9849ed22d7ee.png

 

I'm just a bit loss in Plex...

 

Do I need to set the transcoder temporary directory to a path?  I currently have /transcode as the path.

 

I'm also using a P2000 and have hardware transcode enabled and that seems to be working fine when the transcode start for the beginning.  If I were to change the quality manually it just spins and never changes the transcode.

 

Also I'm getting these warnings in my logs:

image.thumb.png.ee31cf6ca91eb023cfd4121fd23d0b5e.png

 

This is the messages when I change quality on a stream from chrome.

image.png.f31c2366959e8aa15a00b5c11674caba.png

 

I think I might be over thinking this too.

 

Thanks in advance...

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.