Context :
Server Version#: 1.13.9.5439 on Windows Server 2016
Player Version#: 2.5 and 2.20 on Mac OS X 10.12.6
Player Version#: 2.20 on Windows 10 1803/1809
Network : Gigabit LAN Direct Play to a PMP Client
Network : Gigabit WAN Direct Play to a 100Mbit fiber PMP client
Network : Gigabit WAN Transcode to a 10Mbit ADSL PMP client
I need to troubleshoot a buffering issue that randomly cropped up in the past three weeks on all the PMP clients I’ve tested, the only common point being the server.
Randomly, there is some strange 300/400KB/s hard cap to streams once the initial network buffer is consumed by the client (default still at 75MB and it’s done instantly at linerate). If I exit PMP and restart the stream, the issue is cleared in 95% of the cases.
Are any of the plex logs relevant in this case?
I have troubleshooted nearly every possible venue :
- OS network performance is at linerate, impact on local resources is inexistant beyond the expected disk I/O
- NIC Firmware and Drivers are up to date (intel i219-LM)
- the Switches and router operate at linerate without any packet loss or delay
- happens on any kind of video media (H264, H265, with bitrates ranging from 1800kbps up to 20Mbps)
- I’ve wiped clean the PMP installs on the clients
Untested yet :
- Transcode to iOS Plex players
- Completely wiping the PMS install
