Hello,
I updated to Plex 1.10.1.4602 recently, running on a Synology DS1817+ / DSM 6.1.4-15217 Update 5.
Since the Plex update I have seen the following issue:
For some movies - (mkv with x264 codec, HD resolution) - the playback hangs after some minutes. I use a TV app on Samsung TV, and the playback stops with “network issue” error. After this error happens I cant start another movie on the TV nor on the Android App nor Browser playback. It seems like some part of the server (transcoder?) crashed permantly. It happened with different settings (AutoPlay, Direct Play, Direct Streaming). Not for all movies, just for some. Note: before the update these movies played flawlessly.
If I want to resolve the issue I need to restart(!) my synology or to kill the process id of plex. It is not sufficient trying to stop/restart the process, the process is not stopping. Something still runs and prevents proper restart. Not from the DSM or console. Also it is not healing by itself over time.
Please note: I still can start all the Plex apps, connect and browse the movie library. Just starting any movie is not working anymore.
How can I provide the proper information to fix this bug, what part to investigate?
Thank you, I changed the setting and now I am waiting. 
On the network:
I am running the network using a AVM FritzBox 7490. The Syno is in a separate connection, as you have seen with link aggregation. 2 network ports from the Syno are connected to a Netgear Router GS108Tv2. Some other NAS connected to the same Router. The uplink of the Netgear is connected to the FritzBox. The TV is connected via network cable to the FritzBox. So there is no wifi connection for Plex server and Plex client, all network cable based.
[Fritz]
|… |
TV…[Netgear]
…||… …||
…Syno…Other, usually switched off
Syno running static IP, TV running DHCP.
Hi,
maybe first: no, I cant observe any unusual heating of my Netgear.
For the network, I need to investigate when I have a little bit more time. It is running for more than 1 year without change, maybe some issue appeared in the meantime.
The incident just happened, I sent the logs to you.
I doubt somehow that it comes from the network, as most movies play without problems. It is just for 2 or 3 movies where this appears.
I can’t resolve the issue from restarting the server on the DSM user interface, I really have to reboot (or ssh and kill).
So the logs contain:
- incident around 12:47
- trying to stop process (DSM acknowledged it works)
- trying to start the process, but web interface not responding
- reboot
This all should be in the logs.
Thanks for your reply, but in my opinion you are now too fast with your conclusion. Syno says:
“2018/01/31 12:49:07 admin System started counting down to reboot.”
So the network event in the log at 12:49:11.833 is coming from the shutdown of the Syno. It happened after the incident occured. It was 12:47 when playback stopped, I immediatelly logged on the server, tried to stop the Plex server (12:47:58.326), restarted, accessed the web server several times, it is not coming back, so reboot Syno.
Any cause must be at times <=12:47:xx
I can try in the next days to remove the LAG but please consider:
- I can play 2 hours movies without any interruption in general
- so far I identified 3 movies which crash the server most of the time after a while, especially when I restart at a certain time (“want to start where you stopped last time”) but only in conjunction with the Samsung TV app, not when playing on Android or in the browser (in later case both accessing via wifi but the Syno is still on LAG)
From a software point of view I wonder if there isnt a watchdog built into the transcoder, and if the playback is halted the transcoder should be hard stopped and re-initialized (like in embedded systems). It is pretty clear that playback stops in case of major problems on the movie or network, still the transcoder should not remain in such status for hours.
So for sure these movies have an internal corruption in some form and I wonder if it is possible to track the issue and to harden the transcoder against this flaw to increase overall stability.
Hi, the configuration changed in the meantime:
- I put the Syno to a standard single wired LAN, no LAY anmore
- server is now running 1.12.1.4885
It took a while till it happened again, just now. I also think it relates to the Bitrate of the movie, this one was running at 27Mbit/s, I used streaming mode in the app. Seams it never happens below 10-14Mbit (Somewhere in that range, not 100% sure).
Same behavior as before:
- video paused, connection problem display
- I cant stop PLEX, CPU load was at 95%, caused by Plex Transcoder
- I need to reboot the Syno to make PLEX run again
Actually I have no issue with the fact that the connection can break the playback. What I really dont understand why there is no software mechanism to reset the transcoding module if it locks up. It seems there is no watchdog counter implemented which is used to reset the transcoder when it stops.
Hi, I sent the log file as private msg, thanks for your reply.
