Search the Community
Showing results for tags 'HTTP-Response'.
_jackflack posted a topic in General/WindowsFirst off, I do not hold anyone accountable except myself for these errors until proven otherwise. I have tried everything I can think of and read all sorts of different "solutions" for problems similar, yet never resolved. So I am here to ask/beg for some additional guidance. When my Emby (3.4.0 - 3.5.2) server is trans-coding a video file, I will get these HTTP Response ms. spikes that I just cannot explain. When I stream without trans-coding I DO NOT experience this issues that I am aware of. This is only during trans-coding. Internal External Trans-coded Internal This is what I have running. Intel i7-4770s 3.1Ghz 4-Core (8-Core with HT) (stock no OC) (2) 8GB DDR3 1333 1.5v C:\ 120 GB SSD M.2 - OS WIN10 Pro x64 running Emby Server D:\ 120 GB SSD 6GB SATA - Trans-Code Directory (manual setup) (2) Realtek PCIe GbE Family Controller Driver 10.28.615.2018 (updated when started having this issue) FIREWALL is DISABLED in Windows 10 \\Drobo5n\Data - Direct Share in Emby (movies, songs, tv shows) Cisco RV325 Gateway/Switch Wireless Unifi AP-AC-LR ATT - Fiber 1Gbps Charter - Cable 150/25Mbps -------------------------- Ways I've tested this. Local Network: I copied a movie, can be any movie, but this one is libx265 at roughly 5k to the local D: drive on the server. Trans-coding from the local D: drive would allow me to bypass the drobo5n as the possible problem and most of the network for the initial tests since we are dealing with basically lag spikes and reads and writes. I set the max threshold for steaming at 3Mbps (for testing) forcing even on the local network that the file played had to be trans-coded. Threads is set to max (8 total), Path is set to D:\Emby a empty SSD drive, Volume 1, H264 Encoding Fast, CRF 23, Allow subtitle extraction on the fly (set on or off does not change anything). I get ms spikes and the movie will eventually time out on the Emby client on a Amazon Fire-Stick. Client will time out also on a iPad running IOS 11.4.1, iPhone running 10.3.3. Anything on the local network having the movie trans-coded will get ms spikes and eventually time out. Trans-coded to the WAN: Essentially the same setup, however I have substituted the Cisco RV325 for a HP 2530, no changes. I removed the internal network as a factor and placed my ATT fiber connection into pass-thru mode and ran it directly to the emby server Ethernet interface 1 using all of AT&T's settings including a static IP. Same result with ms spike over 30000ms I did the same thing with the Charter connection and achieved the same results. Yes, all network cables have been checked and replaced in some form or fashion. Other than reformat a very pristine install of Windows 10, that is completely patched thru ver. 1803 OS Build 17134.254 I am not sure what could be causing this. I have also tried Intel Quick Sync under Hardware Acceleration and it made no difference if I got a HTTP-Response ms spike. Anyone got a idea on what is causing this? I would like to figure out why the HTTP Response 200 or 204 or 500 just seems to die and then comes back 30000ms later. Hopefully its something dumb I have done and just can't see/find. -jack embyserver.01.txt embyserver.02.txt ffmpeg-transcode-3a894504-e98f-4318-a42c-c343840d0d23.txt ffmpeg-transcode-7e725ccc-cf04-4617-b922-019e3c9454ed.txt