LSL1337

What's the usual memory usage by the this docker image for you guys? for me, it is over 1GB, even when i'm not running anything. seed a few shows for a few days, I delete them, but the docker is still at well over 1 gig of usage (docker stats command) is this normal? shouldn't the usage go down after a while? thanks!

thank you, i will wait until the next stable.

This morning I found the following lines in the log: Deluge wasn't responding this morning (webui), so I restarted the docker, and it was fine. I'm not sure both are related, but deluge is mentioned in the above log. Any idea? I haven't seen this before, and I never had problem with the deluge docker not responding. I updated to 6.5.0 late last week. I was on 6.4.x before that. thank you!

what am I doing wrong. I have the official plex docker. i stopped it, i want to try to create a seperate one, with this image. i stop the main one, so it wouldn't conflict with this in host mode. when i start up this docker, i go to the local IP:32400, I log in, and there is no server set-up screen, only the normal plex web interface, which is looking for my offline server WTF? i started a new server, why cant it start to set it up? (i have plexpass, so it should be possible to have multiple servers under the same plex.tv account, but it doesn't matter, since i cant claim the server with a brand new plex account either...)

ok, i was the stupid one, but this plug-in really ****** me over tonight. listed few old dockers which i didn't use for a while binhex-sonarr etc but it was pointing to the linxserver-sonarr folder (i guess the old binhex docker used the same one I use currently) now my 'prod' sonarr is GONE. ofc it's my fault, but why did it list a folder, if it belonds to az in use application?............. :SIGH:

But before that, If I edited, and it failed to start, it would be still there, to fix the docker start up settings. I will remove the plug-in I hope it will fix it! thanks!

I have a similiar problem. After I edited dockers, (maybe i messed up something, and it didn't start) they went missing, and 'became' an orphan image. I've been using unraid for over a year now, it never happened in 6.3.x few days ago I updated to 6.4.1, and it started happening (could be unrelated) what could cause this? what is an orphan image anyway? why can't I just edit the xml, it wasn't from a template

yeah, it would be great if implemented. i don't want to run a VM, so i can host a drive via isci on a storage OS, lol it's a 2 year old topic, probably never will happen... yeah annoying, i can't install to SMB drives in windows in some cases...

Thank you very much!

I'm a linux noob, but is it possible to write a scipt, which logs my disk status (hdparm -C /dev/sdX) (spin up/spin down) every or 15 or 60 minutes, and writes the results to a file. (maybe extend it with a HDD temp as well, maybe even fan rpm?) preferably all the results (from every different run) in the same file, or seperate disks to seperate files maybe? ( i know i can see a spin down event in the extended log, but it doesn't tell me, when a disk was spun up or down for example) thank you very much!

so, something weired happened, I'm not sure they are connected. Yesterday I noticed there is a new release for radarr, And I went and checked my docker, System/updates, but it didn't show anything there, the page wouldn't load. I pressed restart on the system page, and still nothing showed up, but radarr was working. hmm, though nothing of it. Today I can see, that the mono process is 100% hammering my CPU (on 1 thread), since last night, around the time, I checked the update page and restarted radarr (not the docer) so now I restarted the docker itself, mono cpu load went back to normal. Still the updates page didn't load, after radarr restart, the cpu is 100%. So, Should I not use the restart button in radarr while I'm running it in docker? Or is this a bug, and related to the not loading updates page. Is the updates page loading for you guys? I get error in the logs :"Request Failed: Sequence contains more than one matching element" Even at 100% cpu, the app still works. thanks!

since everything is possible, I'm sure you can somehow put it into the plex docker, but it's not straigforward, like 2 commands in ssh. check the plex forums, I think some guys tried it, but if you are a linux beginner, It's probably over our heads

I'm no expert, but as far as i know you wouldn't be better off eventually, well, even worse. on geforce (consumer) cards there is a transcode limit of 2, so you can't transcode more than that at the same time. on these intul GPUs you can do 5-6-7 1080p transcodes with minimal cpu usage, BUT at the moment, on linux they look like sh1t. intel is working on the new drivers (i think), and hopefully they will improove the quality, to be on par with windows, which is decent. there is no eta, it could be weeks, months, or years. 2 things i'm semi sure about: - intel QS is ok on windows (anyone can try it I guess, and decide for themselfs) and there is no stream limit (nvidia will only do 2 for sure, even on a 600 dollar 1080ti) - intel is working on a fix, there is a github for it with plenty of recent commits. (but will it fix this issue? idk) so there is hope. I think this feature was in beta since christmas. You used plex for months/years without it. Basically it works right now, it's just suboptimal. Hopefully a little more wait will improve it even further. On the bright side, you don't have to do anything, one update will just fix it (hopefully). keep checking the changelogs

yes, it works, you just need a few extra commands on unraid, and in the docker, check back 2-3 pages. I'll write up a noob 3 line guide in a few days.

Don't give up just yet, it works much better on windows. We just need updated(meaning better) linux drivers for windows, but it could take months, or even years.... so eventually...

when I tried it, (putty portable), it detected the (spacebar) before and after 'nobody:users' as an extra character 'ú' so it was 'únobody:usersú', and the command did not run, so I only managed the 777 command. today i'll try it in the GO script, maybe it works there. do I still need the 777 line as well? what about i915. Isn't i965 better/newer, or is that a whole other thing? thanks!

Ofc it's there, but it doesn't mean it will work (without the commands).

from 1080p x264 (15Mbit) to plex 8mbit (I guess h264) it starts at 240p for me as well, but after 2 sec, it's ok. but every 10-20 seconds, when there is proper movement, it goes back to youtube 240p levels of macroblocking. i'm gonna try it out on windows with nvidia gpu, whether or not it's the same. is it possible, that I have old drivers for the gpu or something? I'm on latest unraid (no beta) I'm on 4370T ( HD 4600) EDIT: Ok, IT's seems like it's a linux thing. (https://forums.plex.tv/discussion/290243/hardware-transcoding-linux-artefacts#latest) The drivers are not very good on linux for intel it seems. Maybe they'll update it. There is a 2.0 pre release (few days ago). Hopefully it will be better, and plex will include it in the docker. (right now 1.8.1 is in the container, and it overrides the OS version (as per my basic understanding after a few hours of digging around).

it happpens around once a month. what should I look for, when it happens again (for troubleshooting)? basic system log is enough?

so HW transcode is OK, but when the encoder runs out of 'bandwith', everything goes to shit? only solution is brute force (with more bitrate?), but for remote media streaming it's not viable, cos the only reason for the transcode, is to save bandwith. I can't believe how people find this usable... the picture above is not even the worst I've seen in the past 5 minutes. and when there is a scene change... No B-frames or anything? whatever, I'm getting off topic again, HUGE tanks for the linux help!

It works for me in priviliged mode only, but the video quality is HORRIBLE, I mean at 8mbit 1080p transcode from bluray remux source, it usually good, but without much macro blocking BUT, every 5 second the quality changes to youtube 360p, or lower.... for a few seconds, than it back to passable (still little lower then CPU 8Mbit, but okish) than back to youtube 240p... https://www.dropbox.com/s/qqio89866qs5vf0/8mbit.png?dl=0 can this be a driver issue? I wouldn't recommend this feature to anybody like this... I've read about GPU transcode being slightly inferior to CPU transcode, and i'm bit of a video quality nerd, but this is beyond usable imho. btw, any downsides to running the docker priviliged? anyone managed to get hw transcode to work without priviliged mode? thanks

well, that's not ideal so back to my original question. is it possble to stop the cache_dir plugin and start it again via ssh? than i can do a daily script. anyone? thanks

looks like hardware transcoding is now part of the main release (plexpass), no need to use old beta PMS to enable the feature. anyone got it working on unraid (docker)?

so any other plug-in for the functionality? Do most people don't use cache_dir? i like to keep my disks spun down, i VERY rarely use most of them. cheers

anyone else has issues with cache_dir? sometimes (once a month) it gets stuck, and I have 25% cpu usage, until i stop the plug-in? any idea? thanks