[email protected] Posted September 29, 2021 Share Posted September 29, 2021 13 hours ago, danimal86 said: I ran through the spaceinvader one tutorial and am running into a file path issue (at least i think it is) My docker image is filling up. I'm getting critical errors: I'm guessing that its re-encoding to somewhere where it shouldn't, but i cant figure out why. I did make a new share called "temp" and set it to Prefer for the cache settings. here's the tdarr template: tdarr node: in tdarr under libraries (for my movies library) Edit: So i've got the dockers running right now, and not getting any big errors, but docker memory is still high: you have to increase the image docker size or delete, if you choose to delete the image docker size just follow this guide Quote Link to comment
mike2246 Posted September 29, 2021 Share Posted September 29, 2021 On 9/28/2021 at 3:40 AM, 5hurb said: double check you have --runtime=nvidia in advanced extra param on the node. Now the issue I have, I have 2x GPU's how the F do I get this working? Will takes months to get through my database at the rate its working at with just 1 I've got it working as well with 1 Docker Node, I tried to make a second Docker node keeping all settings the same excpet Name/Node Port/GPU but the server doesn't detect it unless I use node port 8267 which I can't do on both at the same time. Is there a way around this? or can 2 GPU's be added to 1 node? Quote Link to comment
ShangHangin Posted September 29, 2021 Share Posted September 29, 2021 (edited) Found my problem...I had not set one of the library transcode caches to /temp (hand left as default .) Delete the docker, which cleaned up the docker image (dropping about 20% usage), reinstalled and restarted....all is good! TDARR is filling up my docker container. I had installed TDARR the other day, and everything was running find - transcoding for a couple days. I then had a power outage and after a restart, during parity check, I got a docker image space usage error. I stopped TDARR and the image stopped filling. After the parity check was complete, I restarted TDARR - and more filling. I have attached the TDARR log files. Advice appreciated. TDARR Edited September 30, 2021 by ShangHangin Fixed my problem Quote Link to comment
5hurb Posted September 30, 2021 Share Posted September 30, 2021 11 hours ago, mike2246 said: I've got it working as well with 1 Docker Node, I tried to make a second Docker node keeping all settings the same excpet Name/Node Port/GPU but the server doesn't detect it unless I use node port 8267 which I can't do on both at the same time. Is there a way around this? or can 2 GPU's be added to 1 node? I've tried to add the 2nd GPU to the node it only uses one unfortunately. Anyone else got ideas? Quote Link to comment
purplechris Posted September 30, 2021 Share Posted September 30, 2021 Issue with remote nodes if someone could please help? Main Unraid server working great with local node and local GPU I have another Unraid server I am using as a remote node, everything connects, the transcode takes place, but it cannot move the file. All the paths are the same for media and transcode cache and I can move files manually via the server, but the remove tdarr node cannot. Quote Link to comment
danimal86 Posted September 30, 2021 Share Posted September 30, 2021 On 9/28/2021 at 8:32 PM, [email protected] said: you have to increase the image docker size or delete, if you choose to delete the image docker size just follow this guide This doesn't seem like it would fix my problem with the docker image getting too big. I'm not sure whats going on with it, it ran for a solid 8hrs and my docker image was around 63%, then all of a sudden when we started watching plex, it skyrocketed to 99% Quote Link to comment
danimal86 Posted September 30, 2021 Share Posted September 30, 2021 15 hours ago, ShangHangin said: Found my problem...I had not set one of the library transcode caches to /temp (hand left as default .) Delete the docker, which cleaned up the docker image (dropping about 20% usage), reinstalled and restarted....all is good! TDARR is filling up my docker container. I had installed TDARR the other day, and everything was running find - transcoding for a couple days. I then had a power outage and after a restart, during parity check, I got a docker image space usage error. I stopped TDARR and the image stopped filling. After the parity check was complete, I restarted TDARR - and more filling. I have attached the TDARR log files. Advice appreciated. TDARR I'm having the same issue, i've tried moving my transcode folder to a spot within /media. It was working fine for about 8hrs (docker image wasn't rising in size), but when we started watching plex and it shot up to 99% Quote Link to comment
nicksphone Posted September 30, 2021 Share Posted September 30, 2021 1 hour ago, danimal86 said: I'm having the same issue, i've tried moving my transcode folder to a spot within /media. It was working fine for about 8hrs (docker image wasn't rising in size), but when we started watching plex and it shot up to 99% You are probably using tmp as the transcode location on both programs change the location to either a cache drive or a share location should fix it. Quote Link to comment
nicksphone Posted September 30, 2021 Share Posted September 30, 2021 1 hour ago, danimal86 said: This doesn't seem like it would fix my problem with the docker image getting too big. I'm not sure whats going on with it, it ran for a solid 8hrs and my docker image was around 63%, then all of a sudden when we started watching plex, it skyrocketed to 99% You are probably using tmp as the transcode location on both programs change the location to either a cache drive or a share location should fix it. Quote Link to comment
nicksphone Posted September 30, 2021 Share Posted September 30, 2021 3 hours ago, purplechris said: Issue with remote nodes if someone could please help? Main Unraid server working great with local node and local GPU I have another Unraid server I am using as a remote node, everything connects, the transcode takes place, but it cannot move the file. All the paths are the same for media and transcode cache and I can move files manually via the server, but the remove tdarr node cannot. The remote node server have 777 rights to the share location main server? Quote Link to comment
nicksphone Posted September 30, 2021 Share Posted September 30, 2021 5 hours ago, 5hurb said: I've tried to add the 2nd GPU to the node it only uses one unfortunately. Anyone else got ideas? Have you added another node for it so you would have a mode for each gpu Quote Link to comment
nicksphone Posted September 30, 2021 Share Posted September 30, 2021 On 3/27/2021 at 3:36 AM, Meller said: So is this anything like the docker unmanic? Where it will scan your libraries, and automatically convert any file that isn't a specific format? Like make all my movies x265 .mkv files? If so, has anyone used both dockers? Which one gives the best output in terms on video quality post transcoding? How does it handle transcoding HDR content? think of unmanic as easy mode. this is advanced mode where you can basicly tell it to do anything to the trascode if you know how to use ffmeg on the command line. Quote Link to comment
mike2246 Posted September 30, 2021 Share Posted September 30, 2021 23 minutes ago, nicksphone said: Have you added another node for it so you would have a mode for each gpu I've tried to add another Docker Node but it won't show up in the server if I use any other port than the default node port. struggling trying to get a Win 10 Node working. It shows up on the server gui but doesn't transcode with my video card Quote Link to comment
danimal86 Posted September 30, 2021 Share Posted September 30, 2021 37 minutes ago, nicksphone said: You are probably using tmp as the transcode location on both programs change the location to either a cache drive or a share location should fix it. I may have figured it out. On one of my libraries (within tdarr) i had the transcode cache set to /tmp not /temp. I bet that was it. I'll update as it goes. How do i get my docker image size back down to normal Quote Link to comment
nicksphone Posted September 30, 2021 Share Posted September 30, 2021 49 minutes ago, danimal86 said: I may have figured it out. On one of my libraries (within tdarr) i had the transcode cache set to /tmp not /temp. I bet that was it. I'll update as it goes. How do i get my docker image size back down to normal restart docker or the server will flush the tmp folder Quote Link to comment
nicksphone Posted September 30, 2021 Share Posted September 30, 2021 1 hour ago, mike2246 said: I've tried to add another Docker Node but it won't show up in the server if I use any other port than the default node port. struggling trying to get a Win 10 Node working. It shows up on the server gui but doesn't transcode with my video card you need to change the name of the docker for the node as well as the port Quote Link to comment
mike2246 Posted September 30, 2021 Share Posted September 30, 2021 10 minutes ago, nicksphone said: you need to change the name of the docker for the node as well as the port i did Quote Link to comment
danimal86 Posted September 30, 2021 Share Posted September 30, 2021 2 hours ago, nicksphone said: restart docker or the server will flush the tmp folder I restarted both tdarr and the node, didnt flush it, restarted the docker engine, didn't flush it. Left the docker engine off, restarted the server and then restarted the docker engine, didn't work. Still sitting at 96% Any other ideas? Looks like with the docker engine off, i can just resize the image allocation from 20gb to something else, but i'd rather fix it than band aid it. Quote Link to comment
nicksphone Posted October 1, 2021 Share Posted October 1, 2021 13 hours ago, danimal86 said: I restarted both tdarr and the node, didnt flush it, restarted the docker engine, didn't flush it. Left the docker engine off, restarted the server and then restarted the docker engine, didn't work. Still sitting at 96% Any other ideas? Looks like with the docker engine off, i can just resize the image allocation from 20gb to something else, but i'd rather fix it than band aid it. ok so you have 2 options can follow this guide https://medium.com/@samhavens/how-to-make-a-docker-container-smaller-by-deleting-files-7354b5c6c8f1 which is no simple or delete the docker image and reinstall the apps the settings should still be there. Quote Link to comment
nicksphone Posted October 1, 2021 Share Posted October 1, 2021 (edited) 16 hours ago, mike2246 said: i did hop on the discord then will get faster support Edited October 1, 2021 by nicksphone spelling Quote Link to comment
djblu Posted October 1, 2021 Share Posted October 1, 2021 (edited) On 9/30/2021 at 10:03 AM, 5hurb said: I've tried to add the 2nd GPU to the node it only uses one unfortunately. Anyone else got ideas? Maybe use a VM with the 2nd GPU passed through, then set up a Tadrr node on the VM. Make sure you pass through the shares(tadrr transcode/Media) too. Edited October 1, 2021 by djblu Quote Link to comment
5hurb Posted October 1, 2021 Share Posted October 1, 2021 12 hours ago, djblu said: Maybe use a VM with the 2nd GPU passed through, then set up a Tadrr node on the VM. Make sure you pass through the shares(tadrr transcode/Media) too. That's definitely an option. I have had another play around today and checking the node logs if you change the port of the 2nd docker to 8268 the container still thinks its 8267. Looks to be a bug in the tdarr_node. I will get the blinking node on the web interface every time the node registers [32m[2021-10-02T10:33:14.191] [INFO] Tdarr_Node - [39mStarting Tdarr_Node Preparing environment [32m[2021-10-02T10:33:14.208] [INFO] Tdarr_Node - [39mUpdating plugins [32m[2021-10-02T10:33:14.213] [INFO] Tdarr_Node - [39mTdarr_Node listening at http://localhost:8267 [32m[2021-10-02T10:33:14.300] [INFO] Tdarr_Node - [39m---------------Binary tests start---------------- [32m[2021-10-02T10:33:14.308] [INFO] Tdarr_Node - [39mhandbrakePath:HandBrakeCLI [32m[2021-10-02T10:33:14.317] [INFO] Tdarr_Node - [39mffmpegPath:ffmpeg [32m[2021-10-02T10:33:14.326] [INFO] Tdarr_Node - [39mmkvpropedit:mkvpropedit [32m[2021-10-02T10:33:14.326] [INFO] Tdarr_Node - [39mBinary test 1: handbrakePath working [32m[2021-10-02T10:33:14.327] [INFO] Tdarr_Node - [39mBinary test 2: ffmpegPath working [32m[2021-10-02T10:33:14.327] [INFO] Tdarr_Node - [39mBinary test 3: mkvpropeditPath working [32m[2021-10-02T10:33:14.327] [INFO] Tdarr_Node - [39m---------------Binary tests end------------------- [32m[2021-10-02T10:33:14.626] [INFO] Tdarr_Node - [39mCloning plugins [32m[2021-10-02T10:33:15.327] [INFO] Tdarr_Node - [39mFinished downloading plugins! [32m[2021-10-02T10:33:16.221] [INFO] Tdarr_Node - [39mNode registered [32m[2021-10-02T10:33:17.516] [INFO] Tdarr_Node - [39m[2.891s]Plugin update finished [32m[2021-10-02T10:33:20.490] [INFO] Tdarr_Node - [39mNode registered [32m[2021-10-02T10:33:24.539] [INFO] Tdarr_Node - [39mNode registered [32m[2021-10-02T10:33:28.541] [INFO] Tdarr_Node - [39mNode registered I also found another wee problem. If you name your node ID: with just numbers it will do the same behavior and just blink in the web interface. It requires letter text. So you cant just call it 1060 Quote Link to comment
Select25 Posted October 2, 2021 Share Posted October 2, 2021 Heyho Just started to play arround with Tdarr the last few days and i wonder if there is a way to slove my Problem. I wanna encode my Library but only if the Input File is above a specified Size, does someone know about a Plugin or a way to let Tdarr check the Input size first and then encode it to x265? Quote Link to comment
questionbot Posted October 2, 2021 Share Posted October 2, 2021 (edited) Like everyone I just watched SpaceInvaders new video and thought I would give it a go... I was just wondering about how the cache works? Like won't all the new encoded files go into the cache drive? Meaning that only a few encodes and the drive will fill up and then... what? SpaceInvader says to use "prefer cache" for the transcoding tmp dir.. but how dose the other stuff work.. like the place the files are stored.. should be not use the "shares" to find the media folder.. or should be be browsing through the "disk" area and physically setting tdarr to the disks themselves? Is that making sense? I'm a plug and play kind of user.. so not super techy. Also, would anyone be willing to make a noob tutorial on how to connect to a unraid server a node on windows 10? Edited October 2, 2021 by questionbot Quote Link to comment
surferjsmc Posted October 2, 2021 Share Posted October 2, 2021 After a few days running I've gain almost 900GB of free space and still a few files left to transcode so I can say Tdarr works nice but, there is a small issue, inside the temp folder there are 21 files that where not moved yet to their original location after been transcoded and the OG files still in its place, it's like Tdarr moves almost but not all the files or it takes a long time to do it. Temp folder is configured as Prefer Cache. Anyone having same issue? Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.