New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Provider, do you permit Plex on your server ?
Hi ! I have several movies that I want to watch on my trip, in my local network (home) I'm using Plex hooked to Raspberry Pi and it has been working great.
Although I don't know much about transcoding and all that stuff, I think my movies is pre-transcoded, it's a YIFY releases and it run really well on my Pi— consume only 4% - 6% cpu when streaming to my Samsung Smart TV.
Provider usually only set a limit for CPU and don't have regulation regarding Plex usage, the only vps provider I know to have a clear stance on Plex is SSDNodes.
If I were to rent your service, would you allow me to run Plex with said configuration ?
Comments
yes
Private plex? You can run it anywhere.
I get that you mean for personal use only? Yeah, I won't be sharing Plex with anyone else.
No one should have an issue with private Plex unless you start hammering the cpu with transcoding
Yeah, I allow it
I thought Plex is a Nigh Sect thing.
When I watch movies, I go to
/files/
listing of my @PulsedMedia account, copy the link, and paste into VLC Player on the Android tablet.Most content works great.
Only two TV shows require transcoding because bitrate exceeds available bandwidth even in early morning.
So I read ffmpeg manpage and run the necessary command over SSH, and playback the output file instead.
I have media running remotely via plex as plex is the only player that compatible with dolby vision on my tv. Direct stream does not consume much cpu, it's the bandwidths that need to be monitored. If u streaming large media ~ 100gb, then probably should look at the FUP of the provider on bandwidth.
How have your speeds been? I think I remember you posting somewhere on LET that you had to connect through one of your servers for it to be usable. I haven't really posted about it here publicly, but it's been several weeks since I originally contacted support and I'm still waiting for it to get better.
When I first got it, it was great. But, within weeks, pulling files off it was pretty much impossible because even connections to the WebUI or SSH server would time out. As an actual seedbox, it generally seems to work fine for the price, unless I actually want to get any files from it.
I have since found that tunneling my connection through a SOCKS proxy in Europe yields usable speeds. But, that is dependent on other traffic through that proxy and it's a terrible and insecure workaround to have to do just to be able to access or pull files off my server. It also requires being able to run connections through a SOCKS proxy, so I can't simply log into my NAS and rsync the files/folders I want to pull like I previously did.
Unfortunately, I can also no longer get decent speeds when SSH tunneling through one of my higher-bandwidth VPS. When I previously tested tunneling through that server, it resulted in decent speeds.
The following tests are using iperf3 in TCP mode, matching what's used during HTTPS streaming.
Verizon FiOS is my local ISP.
I'm running iperf3 client on a computer connected to wired network.
PulsedMedia to Verizon FiOS: direct connection is unusable for streaming
PulsedMedia to EvolutionHost Roubaix: fast enough within Europe
EvolutionHost Roubaix to Verizon FiOS: okay for streaming
SSH tunnel is terrible.
It's running TCP over TCP, which messes up congestion control algorithm.
SOCKS and HTTP proxies are fine too.
There would be two separate TCP flows, one between origin and proxy, the other between proxy and client.
It's not TCP over TCP, so that congestion control can still work properly.
I'm using WireGuard tunnel.
There's only one TCP running, so that congestion control is working as it should.
PulsedMedia to Android tablet direct: bufferbloat is strong here
PulsedMedia to tablet over WireGuard tunnel at Roubaix: lower but more stable bandwidth
@PulsedMedia is still using TCP CUBIC.
Maybe they should switch to BBR for better performance.
Maybe they should switch to BBR for better performance.
On Deb 10 servers BBR should be used.
All of that is because of Twelve99, and our transit provider keeps switching back to them. Contract papers has been sent in for a new provider over which we plan to reroute ourselves everything that goes to Twelve99/Telia. This has been a major pain point.
We are very far from maxing out our own links, it's the next hop which is having this issue and refusing to fix it.
@yoursunny just sent email to request update on delivery schedule ... after the new links have been hooked up it might take a few weeks to automate stuff, but just send us a ticket for rerouting.
How's the situation now? I think Level3 did some changes which resulted in several networks among which was FiOS i believe, thus routing going back to Twelve99, ought to be fixed now. If not, we can still try RETN and by april GlobalConnect ought to be connected as well.
As for BBR, if you are on Deb10 server please open a ticket and we'll check why it's not using BBR.
I use GreenKVM. I don't see any issues with Plex if you disable transcoding. Below is my 2GB RAM / 1 core VPS while streaming from Plex.
What dashboard is that?
swizzin