Buffering after a few minutes mostly on PC

[jnantg]

Server Version#: 1.18.2.2058
Player Version#: 4.8.4
OS Server: Ubuntu 18.04.3 LTS x64 (Kernel 5.0.0-37) running in VMWare
Client OS: Ubuntu 18.04.3 LTS x64
Logs are in the buttom
**System
**
IP server: IP: 10.100.104.15
IP NAS: 10.100.104.13

VM specs:
Intel Xeon E5-2620 2Ghz (8 cores assigned to this VM. Tried 4 cores 4 threads also but not much difference)
16GB ECC Ram
Disk: 100GB HDD 67GB/95GB Used
Internet connection: Dedicated 1000/1000Mbit business fiber line through a Zyxel USG110 Firewall.

Storage:
Synology DS1819+
The NAS has 4 LAN ports of which 2 are in use. LAN 1 is connected to a switch for out local network here it can be accessed from 10.100.100.0 and LAN 4 is connected to the firewall so the VM can reach it on network 10.100.104.0. All ports are being forwarded and the firewall is currently not enabled.

the server is using.
/mnt/
/mnt/downloads
/mnt/series
/mnt/movies

Issue:

The plex server is running on something called Openflixr and is more or less running flawlessly. Even with SSL however there are some issues with the server.

IT BUFFERS! But it is super hard to reproduce consistently. I haven’t found out as to why it happens. Yesterday i was at a friends house and we watched “Good Boys” on his Xbox using the Plex app. After 5 minutes it started buffering and we had to stop the playback and start it again for it to work. This happened every 5-20 minutes.

At some point it was buffering for a solid 30 minutes and nothing worked. (restarting the server most likely would have)
Later we watched “Shazam!” and it had no issues for the entire movie. No buffering, all smooth.

The issue seems to be worse when watching on PC, but on other devices it usually works perfectly.

I have a suspicion the issue might be related to the encryption. But i haven’t been able to confirm it. However the issue seems to be the same when accessing both http://domain.com and https://domain.com.

I restarted the plex server, reproduced the error and exported logs.
The logs are attached but i replaced IP addresses and domain name. In the logs my domain will show as domain.com and my client IP address will show as 95.174.78.192
I also attached console log from Chrome throwing errors as well as some pictures of settings and other stuff i thought could be relevant. All is in the buttom of this post.

Troubleshooting and errors:

I am 98% sure it is not a resource issue as it happens both when transcoding and when direct streaming. CPU, Ram and network is not hitting any limits.

The only error the console really shows is

error issuing curl_easy_perform(handle):51

and super rarely

error issuing curl_easy_perform(handle):28

But those don’t even appear when the movie stops and starts buffering.

After buffering for a while the error might happen. But it doesn’t seem related.

Another warning that is rarely shown is

NAT: PMP, timed out waiting for response.

But again doesn’t seem related.

DLNA is OFF.

I really hope someone is able to help. I have been dealing with this issue for weeks and i have no idea where to look for the issue anymore. Feel like i have tried it all.

Edit: Tried converting my cert to PFX and moving it to /var/lib/plexmediaserver/ and chown plex:plex. I have updated the attached screencap of the settings. But now my info shows a cert thing sometimes. Not an error, just info. Don’t know if it matters, going back to the previous config does not fix this.

CERT: incomplete TLS handshake: sslv3 alert certificate unknown

Plex logs:
Plex Media Server Logs_2019-12-05_10-16-50.zip (4.4 MB)

Chrome console log:
domain.com-1575539358754.log (255.9 KB)

Relevant screencaps:
screencaps.zip (2.7 MB)

[ChuckPa]

No idea what that even is and clearly never been tested / blessed as supported.
It looks like a proxmox style configuration - which isn’t supported.

You seeing the cert errors is the show stopper.
Your logs show

Dec 05, 2019 10:05:45.578 [0x7fcfaeffd700] DEBUG - Request: [95.174.78.192:44430 (WAN)] GET /playQueues/243?repeat=0&own=1&includeChapters=1&includeGeolocation=1&includeExternalMedia=1 (11 live) TLS GZIP Signed-in Token (jnantg)
Dec 05, 2019 10:05:45.580 [0x7fcfaeffd700] DEBUG - PlayQueue: total generated: 1, before cursor: 0, after cursor: 0
Dec 05, 2019 10:05:45.581 [0x7fcfaeffd700] DEBUG - PlayQueue: No need to refresh, active window is operational.
Dec 05, 2019 10:05:45.595 [0x7fcfeffff700] ERROR - Error issuing curl_easy_perform(handle): 51
Dec 05, 2019 10:05:45.604 [0x7fcfaeffd700] DEBUG - PlayQueue: Client dozfz9qs4u4z8vf0w1jyj2ef requested ownership of play queue 243, but already had it.
Dec 05, 2019 10:05:45.607 [0x7fcfaeffd700] DEBUG - We're going to try to auto-select an audio stream for account 1.
Dec 05, 2019 10:05:45.608 [0x7fcfaeffd700] DEBUG - Selecting best audio stream for part ID 2602 (autoselect: 1 language: en)

Curl error 51 is the certificate error blocking it from working.

If you’ve got Ubuntu running in VMware, why not run the native Plex DEB file (from Plex.tv/downloads ) ?

[jnantg]

It is not a Proxmox setup. It is the native Plex DEB file running on this ubuntu server and being reverse proxied by Nginx. Which is why it is so strange. I even tried reinstalling Plex but without any luck.

I don’t really know what my next step in terms of troubleshooting would be.

[ChuckPa]

Allow me to be clear:

Openflixr isn’t known. It’s never been tested. It can’t be supported until such time as it passes validation test.

I’m sorry.

[jnantg]

So i tried to re-install Plex Media server from the .DEB file. Now it is just a Plex Media server running on Ubuntu. I am still getting the same errors. Even though i tried converting my SSL cert to PKSC#12 and add that to the Plex server.

Disabled everything on the server except for Nginx, Pihole and Plex Media server. Plex media server is not being blocked by the Pihole.