Server Version#: Version 1.21.1.3795
Player Version#: XBOX One current version (no updates available)
I’m running into an issue where, I am suddenly on dead stalls with playback. This doesn’t just occur on the XBOX One, it occurs on the Playstation4 and Roku. Every device. They will play for a few minutes, then die. It occurs with 1080P or 4K.
This is interesting because I connect to the NAS that stores the data via a private 10Gb network connection (straight cable, on a 10.0.10.x network) which is only available to the server, and to the rest of the network through a common internet linked connection (192.168.5.x) This has worked for more than a year with no issue, including 4K BD ISOs. Now, I can’t even stream 720P since some update a few weeks ago. What the heck happened here? I’ve tested file transfers from my QNAP TS963X which is 40TB in 8TBx6 with 4 1TB SSD for Cache. I’m still getting 600Mbs data transfer straight from that unit to my PC in a file copy.
(File removed)
So, I have ZERO idea why Plex is suddenly choking so hard when it was so reliable a few weeks ago.
Host server:
Windows 10 x64
64GB Ram
Ryzen 5950X
GTX1080 TI
Asus 10Gbe Network Card
Onboard 1Gb for outbound to network (where things like XBOX etc. are)
Just in case, I tested with another 10Gbe adapter, same results. I did the update today that was available… Version 1.21.1.3830 and get the same result still
Is the 10 GB NIC with a chip from Broadcom, by chance?
My colleague is just in the process of swapping all NICs for Intel, because the other adapters suddenly stop after a while.
Except that this may lead to a drastic increase in CPU load. And the full speed of 10 gbps may also not be achievable any longer.
With a 10 gbps adapter, those hardware offload features are practically mandatory.
Oh! To clarify, I suggest it only as a troubleshooting step in response to traffic stalls.
TSO/LSO and other “stateful” offloads have caused a pretty remarkable number of problems. Thank goodness it’s gotten better in the last few years. Offload also replaces the TCP congestion control built into the OS, and is often less sophisticated.
Because file transfers on the 10GbE interface are working well, perhaps there’s a problem with the 192.168.5.x interface. Even with all offload disabled, there should be no danger of slowdown on a 1GbE interface.
Do you have a way to test file transfers on the 192.168.5.x interface? It would be nice to confirm whether it’s Plex-specific or the interface/network.
Can you test a few movie files stored locally on the server, and take the 10GbE and NAS out of the equation? It would be nice to narrow down the problem.
Over the 192.168.5.x interface, I still get full 1Gb transfers between NAS. If I route Plex to get media from there (creating a new library, different folders, same device different network route) I get same result.
One thing about this is that all of this worked FLAWLESSLY for months in this configuration until suddenly about 3 weeks ago it stopped. I’ve checked the health of the NAS, no issues. Checked routes, no changes. I am stumped. I’ve tried another service test (I still have a license to a competitor product in emby) and no issues there like this one, or KODI. So, I have no idea at all what is going on with Plex to generate this.
Plex shows full data transfer for about 15 minutes then it drops to zero data transfer.
Well, I’m still working on this but it appears as though subtitles being on may be a huge part of this. If I go into plex server and I completely disable it by putting it to “manual” on the server, then it doesn’t happen nearly as often. I’m still looking for a full solution as nothing else results in this issue (not Kodi, Emby, etc.)
