Server Version#: 1.18.9.2571
Player Version#: 7.30.1.16475
Hello guys,
PMS is installed on my QNAP 653a and I’m using as client a Nvidia Shield Pro 2019, therefore the playback is managed completely by the last one.
Recently I noticed that the playback gets sometimes stuck for a second or 2 while watching 4K with high bitrate. I started to investigate and it seems not related to the file itself, if I re-play the same scene not necessarily it get stuck again, files aren’t transcoded as well by the NAS, so I had a look to the CPU of my NAS it looks quite under stress while playing.
The NAS is now managing 6 HDDs and in normal conditions CPU is always around 67%, but it reaches around 90-95% while playing 4Ks, so I’m starting to think that the already not so excellent CPU is starting to suffer the whole workload.
In light of this I was thinking if it might be beneficial to migrate the PMS (just that) somewhere else (maybe a NUC?), or if there are any other solutions to apply. I tried also running a PMS on the Shield itself but also here I’m not entirely satisfied by smoothness of the playback.
Thanks in advance for your support.
67% just for having 6HDDs seems awful high unless you are running some kind of VMware or other application.
Also a jump of 25% just to transfer data seems high (your model does not support 4k transcoding using Plex according to the Plex compatibility chart so it should only be transferring the data your Shield should do the rest via direct play)
Might want to share more details of what your sever is actually doing so posters here could provide better insights?
As far as the pause, if I were to guess I would suggest maybe your network itself is bottle-necking your data rates?
Hi, yes I’m currently running a VM (W7).
About the bottle neck, it seems a bit strange to me. Both NAS and Shield are hard wired to the modem (Fritz!Box 7590), therefore everything should be 1 Gbps. Is there anything else that can interfere with data rate?
Would you mind ,
- Verify DEBUG logging on, VERBOSE logging off
- Recreate the playback session - just long enough to capture what’s happening
- Stop Playback
- wait 30 seconds for logs to be written
- Settings - Server - Troubleshooting - Download Logs.
HDDs and CPU utilization aren’t related
Virtualization Station showing 49% of the 63% total, given what’s changed in QTS recently, seems to be closer to root cause (the VM). The cost of I/O in QTS has increased – sadly.
Hi Chuck,
I’m not sure what happened lately but the last few updates have really thrown a stick in the spokes. I’m on a TS-1677x with a Ryzen 1700, 64gb ram and a Quadro P4000. I’m having all sorts of issues. Mainly the library disappears and even starting and stopping the service doesn’t fix it. Please advise.
disappears? Can you show me with screenshots?
If you see it dropping the files, can I have the logs of them going in/out , whatever you can capture as it happens?
Hi Chuck,
Once the issue will reappear I’ll follow your procedure, thanks!
Actually, now I cannot even get Plex to work. This last update has really borked things. How can I find an old version for QNAP x86?
There are no 32 bit QNAP binaries anymore (at QNAP’s request)
Discontinuing support for QTS 4.2 is what ended the support.
Which QNAP NAS do you own and which QTS version is it?
TS-1677x QTS 4.4.1.1216
Plex server keeps going unreachable all the time now since last update.
The Ryzen 3 CPU has 6300 Passmarks (about that of an i5)
May I see some DEBUG logs please which capture what you’re trying to share?
I will look at what PMS is doing and then tell you what I see. Hopefully my reply can point in a direction or offer explanation.
I cannot even access it anymore. It’s a Ryzen 7 1700 with 64gb of ram and a Quadro P4000. Everything was working just fine and the last update just broke everything. Nothing even works locally. I don’t know how to get any access.
You still have access to the QNAP, correct?
Install this package. It’s written by Dane22 (a ninja here).
It will find your Plex installation folders and create a Shared Folder in File Station you can access it with.
Once you install it, please open FileStation, go into the share, navigate down until you see Logs.
Right-click Logs and create a ZIP file.
Download and attach that ZIP file here please.
I’ll look at what’s happening.
I have a 1277 with same CPU and a lesser GPU.
All sorts of issues the last few days and 3 updates.
Wont load and had to reboot NAS !
Worst is the remote access dr0ps every few minutes - some times for hours (and generally comes back 2-3 minutes later )
Then runs for a few hours with no dropouts !!
I am running 1.19.1.2589
Facts, Plex latest update for Qnap is 100% a ■■■■ show.
I’m going to disagree. How about we find out what the issues are and resolve them ?
To resolve them, I’m going to need quantifiable , DEBUG logs which capturing the issues. I will also need the XML of the particular media being played so I can look at what PMS is seeing.
Hello Chuck,
I got another event, on the 7th, more or lesse between 21:30 - 22:00. Attached you’ll find the logs. Don’t know if you need anything else.
Plex Media Server Logs_2020-04-08_00-15-27.zip (4.4 MB)
In the logs, I see a playback to the Shield.
DirectPlay=1
Apr 07, 2020 21:30:50.920 [0x7f9df3735700] DEBUG - Streaming Resource: Reached Decision id=86394 codes=(MDE=1000,Direct play OK.) media=(id=209890 part=(id=211122 decision=direct play protocol=* streams=(Video=(id=367354 decision= width=3840 height=2160) Audio=(id=367355 decision= channels=0 rate=0))))
Apr 07, 2020 21:30:50.921 [0x7f9e2b16e700] DEBUG - Completed: [192.168.178.31:41602] 200 GET /video/:/transcode/universal/decision?audioBoost=100&autoAdjustQuality=0&directPlay=1&directStream=1&directStreamAudio=1&fastSeek=1&hasMDE=1&location=lan&maxVideoBitrate=200000&mediaBufferSize=209664&mediaIndex=0&partIndex=0&path=%2Flibrary%2Fmetadata%2F86394&protocol=*&session=2171f07960115647-com-plexapp-android&subtitleSize=100&videoBitrate=200000&videoQuality=100&videoResolution=3840x2160 (11 live) TLS GZIP 19ms 1375 bytes (pipelined: 23)
Apr 07, 2020 21:30:51.136 [0x7f9e2b45c700] DEBUG - Auth: authenticated user 1 as Sgt Hartman
I do see where the database needs optimizing. (Settings - Server - Scheduled Tasks does regular maintenance)
Signed-in Token (Sgt Hartman)
Apr 07, 2020 21:34:00.311 [0x7f9e2a8a4700] DEBUG - Setting container serialization range to [0, 50] (total=-1)
Apr 07, 2020 21:34:00.382 [0x7f9daeb7d700] WARN - SLOW QUERY: It took 870.000000 ms to retrieve 51 items.
Apr 07, 2020 21:34:00.390 [0x7f9daeb7d700] DEBUG - Setting container serialization range to [0, 50] (total=552)
Apr 07, 2020 21:34:00.423 [0x7f9da4829700] DEBUG - It took 60.000000 ms to retrieve 1 items.
Apr 07, 2020 21:34:00.466 [0x7f9e2b45c700] DEBUG - Completed: [192.168.178.23:55084] 200 GET /hubs/home/recentlyAdded?type=1&includeExternalMedia=1 (14 live) GZIP Page 0-50 544ms 20465 bytes (pipelined: 1)
Apr 07, 2020 21:34:00.482 [0x7f9df3735700] DEBUG -
There were no visiblity tests during this time.
What did you experience? (as detailed as possible please)
Hi Chuck,
I quote you “I do see where the database needs optimizing. (Settings - Server - Scheduled Tasks does regular maintenance)”. Shall I do anything about it?
Then, “What did you experience? (as detailed as possible please)”. While playing suddenly it stops for a second or two then restarting, then stopping again, something like 5-6 times. I closed and restarted the Plex, replaying the same scene without having the same issue, sorry but this is all I can explain.
That’s enough of a description.
I looked at the timestamps and now need ask.
Since this is DirectPlay, is any of this on WiFi and, if WiFi, is the player near the edge of the WiFi coverage?
