DietPi-Software | Plex Media Server not responding

Created on 14 Mar 2019  路  8Comments  路  Source: MichaIng/DietPi

Plex Server not responding

I had a working installation of Plex Media Server and it stopped working without any explanation.The service is listed as running but the server cannot be connected to . I tried uninstalling and installing again but the issue persists. I include the plex log which has errors I do not know how to interpret.

Required Information

  • DietPi version | cat /DietPi/dietpi/.version
#!/bin/bash
G_DIETPI_VERSION_CORE=6G_DIETPI_VERSION_SUB=21
G_DIETPI_VERSION_RC=1G_GITBRANCH=master
G_GITOWNER=Fourdee
  • Distro version | echo $G_DISTRO_NAME or cat /etc/debian_version
9.8
  • Kernel version | uname -a
Linux Nal 4.14.98-v7+ #1200 SMP Tue Feb 12 20:27:48 GMT 2019 armv7l GNU/Linux
  • SBC device | echo $G_HW_MODEL_DESCRIPTION or (EG: RPi3)
RPi 3 Model B+ (armv7l)
  • Power supply used | (EG: 5V 1A RAVpower)
    5V 2.4A Official Raspberry Pi
  • SDcard used | (EG: SanDisk ultra)
    SanDisk 32GB

Additional Information (if applicable)

  • Software title | Plex Media Server (42)
  • Was the software title installed freshly or updated/migrated?
    It was working, the issue appeared, uninstalled and reinstalled but the issue persists
  • Can this issue be replicated on a fresh installation of DietPi? I do not know

Steps to reproduce

  1. Install Plex Media Server on latest version

Expected behaviour

  1. Working Plex installation

Actual behaviour

  1. Plex Media Server seems to be running and I can connect to localhost:32400 and see the plex web interface but the Server itself is listed as unavailable and cannot be connected to.

Output of Logs/Plex Media Server.log is included here

Most helpful comment

I believe I figured it out

The issue was that after the reinstall, Plex needed a connection from the local network to be configured. I was doing this through a OpenVPN connection since the RPi is offsite. The VPN allowed me to access the plex web client through 192.168.1.X:32400 but did not allow me to do the initial server setup. It did not give any error/warning either, I guess they could improve this.

I was able to get it working with SSH port forwarding using -L 8888:localhost:32400.

My best guess is that my plex had the bug @WolfganP was referencing and with the reinstall the user configuration was wiped causing this issue.

Apparently all the errors in the logs were benign (had to google each one of them) and were related to missing folders and DNS with IPv6 configuration.

All 8 comments

@JJGO
Thanks for your report. I will try to replicate.

Errors from your log:

Mar 14, 2019 20:25:42.665 [0x70dff400] ERROR - Error issuing curl_easy_perform(handle): 7
Mar 14, 2019 20:25:42.665 [0x70dff400] WARN - HTTP error requesting POST http://127.0.0.1:32600/devices/discover (0, No error) (Failed connect to 127.0.0.1:32600; Connection refused)
...
Mar 14, 2019 20:25:46.367 [0x745435d0] ERROR - Transcoder: Error cleaning old transcode sessions: boost::filesystem::directory_iterator::construct: No such file or directory: "/var/lib/plexmediaserver/Library/Application Support/Plex Media Server/Cache/Transcode/Sessions"
...
Mar 14, 2019 20:25:46.816 [0x717fd400] ERROR - SQLITE3:(nil), 539, recovered 3 pages from /var/lib/plexmediaserver/Library/Application Support/Plex Media Server/Cache/gn_lists.gdb-journal
...
Mar 14, 2019 20:25:48.371 [0x6dfff400] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
... repeatedly

With v6.22 we will switch to the new Plex v1.15 packages, although I am currently not yet sure if we should automatically update existing installs or leave them to not mess with running systems. The new installs on ARM do not use the 3rd party repo anymore, since with v1.15 official ARM binaries are available: https://github.com/MichaIng/DietPi/pull/2554

@JJGO
On x86_64 it works fine, although tested with the latest v1.15 install.


Found an old thread: https://forums.plex.tv/t/error-error-issuing-curl_easy_perform-handle-28-on-an-sony-bravia-tv/93260/36

  • Just what can be tried additionally is disabling all DNLA features from within the web UI (if possible).

If the above does not work, please uninstall Plex (as this was a fresh Plex install, right?) and apply the current DietPi Beta: https://github.com/MichaIng/DietPi/issues/2632
Then install Plex again. This will install the new official binaries instead of those from the 3rd party repo, which was required for ARM before.

  • Due to lack of time and as of new releases I am not going to spend time into investigating/fixing the old 3rd party ARM installs.

@JJGO stop Plex service and try running:

LD_PRELOAD=/usr/lib/arm-linux-gnueabihf/libarmmem-v6l.so; sudo rm /etc/ld.so.preload

Ref: https://forums.plex.tv/t/plex-server-suddenly-not-starting-on-raspberry-pi-after-crashed-update/387681/5

Just some hours ago v1.15 has been pushed to the dev2day repo as well: https://forums.plex.tv/t/official-pms-for-armv7-what-will-happen-with-this-community-one/372290/21

So I suggest to update the package as well, before applying the above fix:

G_AGI plexmediaserver-installer

The repo will be dropped ~1 month later. An official migration guide (to the official binaries/deb package) is still missing, so still not sure if it is safe in every case to simply overwrite the dev2day package with the official one and remove the repo.

Thanks for all the suggestions.

I updated the plexmediainstaller with G_AGI, reinstalled the plexmediaserver, and ran the coimmand @WolfganP mentioned. The behaviour is still the same. The Logs contain the following errors

Mar 17, 2019 18:27:29.078 [0x6fcff450] ERROR - Error issuing curl_easy_perform(handle): 7
Mar 17, 2019 18:27:34.199 [0x73c4e540] ERROR - Transcoder: Error cleaning old transcode sessions: boost::filesystem::directory_iterator::construct: No such file or directory: "/var/lib/plexmediaserver/Library/Application Support/Plex Media Server/Cache/Transcode/Sessions"
Mar 17, 2019 18:27:35.841 [0x6dcff450] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1

With the last error repeating every 100ms continuously.

I will try disabling the DLNA part.

How can I install the new beta you mention?

@JJGO it seems your issue is related to the file structure as some files/folders are missing.

If you want to install Plex manually, download the image from https://www.plex.tv/media-server-downloads/ and then execute:
sudo apt remove plexmediaserver-installer && sudo dpkg -i yourdownloadedimagefile.deb

@JJGO
馃 Lets see if disabling DLNA works.

Did you reboot after updating Plex and applying WolfganP suggesting?


Yeah about the missing file/dir it is not the first error, so might be related to the above ones, but you could test to create the dir manually:

mkdir -p /var/lib/plexmediaserver/Library/Application Support/Plex\ Media\ Server/Cache/Transcode/Sessions

How to apply the Beta: https://github.com/MichaIng/DietPi/blob/beta/BRANCH_SYSTEM.md#steps-to-use-the-dietpi-beta-branch

  • Just note that this does not update your Plex or something, especially not since the repo already provides v1.15. So this along will not solve your issue most likely.
  • You could try to purge Plex completely and install it freshly after applying the Beta. The installed package/binaries are different, although the same version, so there is a small chance that this will solve your issue.

Otherwise I suggest to report this to the Plex forum, respectively search for similar issues. Important to know that the issue occurred on v1.14 as well.

I believe I figured it out

The issue was that after the reinstall, Plex needed a connection from the local network to be configured. I was doing this through a OpenVPN connection since the RPi is offsite. The VPN allowed me to access the plex web client through 192.168.1.X:32400 but did not allow me to do the initial server setup. It did not give any error/warning either, I guess they could improve this.

I was able to get it working with SSH port forwarding using -L 8888:localhost:32400.

My best guess is that my plex had the bug @WolfganP was referencing and with the reinstall the user configuration was wiped causing this issue.

Apparently all the errors in the logs were benign (had to google each one of them) and were related to missing folders and DNS with IPv6 configuration.

Was this page helpful?
0 / 5 - 0 ratings

Related issues

k-plan picture k-plan  路  3Comments

Invictaz picture Invictaz  路  3Comments

Invictaz picture Invictaz  路  3Comments

Fourdee picture Fourdee  路  3Comments

Kapot picture Kapot  路  3Comments