Plex Media Server - Constant CPU usage

Plex Media Server Desktops & Laptops
1

I’ve just noticed my plex server PC is running slower. Plex Media Server.exe is running at a constant 25% CPU with no activity (no shows playing or clients connected).

I looked in the main log, I see a lot of this type of activity:

Summary

Feb 10, 2019 12:48:13.027 [11844] DEBUG - PubSubManager: Time to connect to 66.228.52.149 was 151 ms.
Feb 10, 2019 12:48:13.027 [11844] DEBUG - PubSubManager: Time to connect to 172.104.209.60 was 158 ms.
Feb 10, 2019 12:48:13.027 [11844] DEBUG - PubSubManager: Time to connect to 45.79.211.188 was 195 ms.
Feb 10, 2019 12:48:13.027 [11844] DEBUG - PubSubManager: Time to connect to 45.79.10.123 was 195 ms.
Feb 10, 2019 12:48:13.027 [11844] DEBUG - PubSubManager: Time to connect to 184.105.148.101 was 206 ms.
Feb 10, 2019 12:48:13.027 [11844] DEBUG - PubSubManager: Time to connect to 82.94.168.54 was 309 ms.
Feb 10, 2019 12:48:13.027 [11844] DEBUG - PubSubManager: Time to connect to 172.104.133.220 was 382 ms.
Feb 10, 2019 12:48:13.027 [11844] DEBUG - PubSubManager: Time to connect to 185.3.95.210 was 407 ms.
Feb 10, 2019 12:48:13.027 [11844] DEBUG - PubSubManager: Time to connect to 139.162.115.125 was 599 ms.
Feb 10, 2019 12:48:13.028 [11844] DEBUG - PubSubManager: Time to connect to 139.162.21.199 was 781 ms.
Feb 10, 2019 12:48:13.028 [11844] DEBUG - PubSubManager: Updating best ping time for 45.79.10.123 to 195 ms.
Feb 10, 2019 12:48:15.317 [5188] DEBUG - HTTP 200 response from PUT https://plex.tv/api/servers/a5fa0e6cbf155de3b9997ca8299f0c6389781ec7/connectivity?X-Plex-Token=xxxxxxxxxxxxxxxxxxxx&asyncIdentifier=82ccefe9-5e6a-4b63-ac7e-fcf1e07dec0a
Feb 10, 2019 12:48:15.317 [6048] WARN - PubSub: Received notifyConnectivity event with incorrect async identifier (9f977dd2-5047-4544-96b9-4b1ba7e185cb, expected 82ccefe9-5e6a-4b63-ac7e-fcf1e07dec0a)
Feb 10, 2019 12:48:15.317 [7988] DEBUG - EventSource: Got event [data] ‘’
Feb 10, 2019 12:48:15.317 [7988] WARN - PubSub: Received notifyConnectivity event with incorrect async identifier (025e5e84-af81-4a58-a51c-6c92e79f2d6e, expected 82ccefe9-5e6a-4b63-ac7e-fcf1e07dec0a)
Feb 10, 2019 12:48:15.317 [7988] DEBUG - EventSource: Got event [data] ‘’
Feb 10, 2019 12:48:15.317 [7988] WARN - PubSub: Received notifyConnectivity event with incorrect async identifier (b33a06a9-2973-4985-8ed8-34563571fa4b, expected 82ccefe9-5e6a-4b63-ac7e-fcf1e07dec0a)
Feb 10, 2019 12:48:15.317 [6048] DEBUG - Completed: [127.0.0.1:55691] 200 PUT /myplex/refreshReachability (10 live) GZIP 7893ms 268 bytes (pipelined: 1)
Feb 10, 2019 12:48:15.317 [7988] DEBUG - EventSource: Got event [data] ‘’
Feb 10, 2019 12:48:15.318 [7988] DEBUG - PubSub: Got notified of reachability: 0 for :0
Feb 10, 2019 12:48:15.321 [7988] DEBUG - Auth: authenticated user 1 as dk4dk4
Feb 10, 2019 12:48:15.322 [4912] DEBUG - Request: [127.0.0.1:55691 (Loopback)] GET /myplex/account (10 live) GZIP Signed-in Token (dk4dk4)
Feb 10, 2019 12:48:15.323 [6048] DEBUG - Completed: [127.0.0.1:55691] 200 GET /myplex/account (10 live) GZIP 1ms 1152 bytes (pipelined: 2)
Feb 10, 2019 12:48:15.325 [6048] DEBUG - Auth: authenticated user 1 as dk4dk4
Feb 10, 2019 12:48:15.325 [0644] DEBUG - Request: [127.0.0.1:55691 (Loopback)] GET /myplex/account (10 live) GZIP Signed-in Token (dk4dk4)
Feb 10, 2019 12:48:15.326 [7988] DEBUG - Completed: [127.0.0.1:55691] 200 GET /myplex/account (10 live) GZIP 1ms 1152 bytes (pipelined: 3)
Feb 10, 2019 12:48:20.191 [6048] DEBUG - handleStreamRead code 2: End of file
Feb 10, 2019 12:48:20.198 [7988] DEBUG - Auth: authenticated user 1 as dk4dk4
Feb 10, 2019 12:48:20.198 [11908] DEBUG - Request: [127.0.0.1:55691 (Loopback)] GET /player/proxy/poll?deviceClass=pc&protocolVersion=1&protocolCapabilities=timeline%2Cplayback%2Cnavigation%2Cmirror%2Cplayqueues&timeout=1 (9 live) GZIP Signed-in Token (dk4dk4)
Feb 10, 2019 12:48:20.199 [11908] DEBUG - Beginning read from two-way stream.
Feb 10, 2019 12:48:25.097 [7988] DEBUG - Auth: authenticated user 1 as dk4dk4
Feb 10, 2019 12:48:25.099 [5188] DEBUG - Request: [127.0.0.1:55689 (Loopback)] GET /:/prefs (9 live) GZIP Signed-in Token (dk4dk4)
Feb 10, 2019 12:48:25.099 [7988] DEBUG - Auth: authenticated user 1 as dk4dk4
Feb 10, 2019 12:48:25.100 [4912] DEBUG - Request: [127.0.0.1:55686 (Loopback)] GET /accounts/1 (9 live) GZIP Signed-in Token (dk4dk4)
Feb 10, 2019 12:48:25.101 [7988] DEBUG - Completed: [127.0.0.1:55686] 200 GET /accounts/1 (9 live) GZIP 1ms 518 bytes (pipelined: 2)
Feb 10, 2019 12:48:25.113 [6048] DEBUG - Auth: authenticated user 1 as dk4dk4
Feb 10, 2019 12:48:25.113 [7988] DEBUG - Auth: authenticated user 1 as dk4dk4
Feb 10, 2019 12:48:25.115 [0644] DEBUG - Request: [127.0.0.1:55686 (Loopback)] GET /myplex/account (9 live) GZIP Signed-in Token (dk4dk4)
Feb 10, 2019 12:48:25.116 [6048] DEBUG - Completed: [127.0.0.1:55686] 200 GET /myplex/account (9 live) GZIP 3ms 1152 bytes (pipelined: 3)
Feb 10, 2019 12:48:25.116 [11908] DEBUG - Request: [127.0.0.1:55566 (Loopback)] GET /system/:/prefs (9 live) GZIP Signed-in Token (dk4dk4)
Feb 10, 2019 12:48:25.117 [11908] DEBUG - [com.plexapp.system] Sending command over HTTP (GET): /system/:/prefs
Feb 10, 2019 12:48:25.117 [11908] DEBUG - HTTP requesting GET http://127.0.0.1:55553/system/:/prefs
Feb 10, 2019 12:48:25.126 [11908] DEBUG - HTTP 304 response from GET http://127.0.0.1:55553/system/:/prefs
Feb 10, 2019 12:48:25.126 [11908] DEBUG - [com.plexapp.system] HTTP reply status 304, with 0 bytes of content.
Feb 10, 2019 12:48:25.127 [6048] DEBUG - Completed: [127.0.0.1:55566] 304 GET /system/:/prefs (9 live) GZIP 13ms 324 bytes (pipelined: 5)
Feb 10, 2019 12:48:25.143 [6048] DEBUG - Completed: [127.0.0.1:55689] 200 GET /:/prefs (9 live) GZIP 45ms 6510 bytes (pipelined: 2)
Feb 10, 2019 12:48:26.762 [7988] DEBUG - Auth: authenticated user 1 as dk4dk4
Feb 10, 2019 12:48:26.762 [4912] DEBUG - Request: [127.0.0.1:55689 (Loopback)] GET /diagnostics/logs (9 live) GZIP Signed-in Token (dk4dk4)
Feb 10, 2019 12:48:26.762 [4912] DEBUG - Diagnostics: Building logfile zip

Something has changed in Plex this is not normal behaviour (constant CPU 25%) .
Please assist in finding the root cause.

2

I traced the process and this is the module that is sucking up 25% CPU.
I did full reinstall after I renamed the folder:
C:\Users<user>\AppData\Local\Plex Media Server

So all things are fresh still get 25% CPU in use…? Any thoughts anyone?

Image:
image

3

I found this thread and I disabled UPNP on my router, this fixed Plex from running at 25%CPU?
Why does Plex runaway with CPU if UPNP is enabled on my router?

1 Like
4

I would like to enable upnp again …

5

I was experiencing the same issue. Disabling UPNP fix it for me too but I’d like to turn that back on.

6

Does Plex not support this forum?

7

Hi same problem with upnp on, i hope PLEX resolve problem quickly, cant acces from outside now

8

Same issue with PMS on Linux too. Uses a whole CPU when uPnP is enabled on my router, even though I’m using manual port forwards for external access.

It seems to only matter that uPnP is enabled when PMS starts up or if it is enabled after startup. That is, if it’s enabled on start, CPU goes to 100% (of one core) and remains that way. If it’s disabled at startup, then it does not. If it is enabled after PMS startup, then CPU goes to 100%. If i disabled it after startup, CPU remains at 100%. So it’s like something detects uPnP is enabled, then goes into an infinite loop trying to process something and never exits.

v1.15.2.793 although it has been happened for at least the past few versions (have noticed this CPU pattern for the past few months).

I turned on debug logging, stopped plex, disabled uPnP, cleared the logs, started plex and waited for it’s initial processing/updating to be finished, observed that CPU was <1%. Then turned on uPnP, and nothing was written to the logs despite CPU immediately rising to 100%.

9

Can confirm. Also occurring on Ubuntu 18.04 server after upgrade to plexmediaserver_1.15.2.793-782228f99_amd64.deb. My CPU usage was ~60% though. Disabled UPnP, and server did NOT immediately reduce CPU usage. Rebooted router just to be sure, then restarted (stop/start didn’t seem to help but restart did) plexmediaserver.service, and CPU usage returned to normal.

Wasn’t happening on previous firmware, plexmediaserver_1.14.1.5488-cc260c476_amd64.deb.

Hope this helps.

10

I am seeing the same issue on Windows Server 2016 with Plex Media Server 1.15.4.919. Disabling UPnP on my OpenWRT router and restarting Plex gets the CPU usage back down to normal.

11

I can aknowledge this issu on my synology nas with Plex Media Server 1.15.6.1079.
Aftre disabling the UPNP on my Router the CPU goes back to normal on the NAS.
I also don’t use UPNP for Plex, I use a static Port forward then UPNP with Plex is actually not working. (My router is also a OpenWrt 18.06).

12

Hy all, at least I could fix the problem for me now, and this without disabling the upnp on my router.
I have more than one interface defined on my router which seems to cause a problem in upnp at all. It can not determ the external interface, wich ends in endless retrys on plex.

You need to specify the external interface in /etc/config/upnpd.

Under the
config upnpd ‘config’

you need to add the option:
option external_iface ‘wan’

On my router it is the standad wan.

Then restart the miniUpnpd /etc/init.d/miniupnpd restart

Aftre this restart your plex.

Hoppe this helps you aswell.

13

I tried this and restarted both. It had no effect on Plex CPU usage. Problem still exists.

14

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.