[Support] HaveAGitGat - Tdarr: Audio/Video Library Analytics & Transcode Automation


Recommended Posts

1 hour ago, absolute_badger said:

UPDATE: tdarr 2.00.14 is confirmed broken. Do not upgrade to this version until a fix is available!

It's not broken. The new Server image contains a Node so you most likely have 2 Nodes running on the same port which are interfering with each other.

 

I've just released a small Server image update so you now have to set internalNode=true to enable the Node in the Server container. So pull the latest container and enable the internal Node as you see fit. Leaving it disabled should mean things run the same as before.

 

https://docs.tdarr.io/docs/installation/docker/run-compose/

  • Like 1
Link to comment
5 minutes ago, wgstarks said:

Does this mean that the tdarr_node docker is no longer required on unRAID? Or maybe the new node isn’t included in the unRAID build. I’m not seeing any evidence of it in the container configuration.

Yeah if you’re running the server and node on the same machine then you can just use Tdarr_Server and set internalNode=true. 
 

There’s a PR to update the unraid template:

 

https://github.com/selfhosters/unRAID-CA-templates/pull/317

  • Thanks 2
Link to comment
36 minutes ago, HaveAGitGat said:

It's not broken. The new Server image contains a Node so you most likely have 2 Nodes running on the same port which are interfering with each other.

 

I've just released a small Server image update so you now have to set internalNode=true to enable the Node in the Server container. So pull the latest container and enable the internal Node as you see fit. Leaving it disabled should mean things run the same as before.

 

https://docs.tdarr.io/docs/installation/docker/run-compose/

I don't know that this is correct exactly. I have not set interalNode=true after the update, and still have the node docker image running. I have all of the appropriate settings for my GPU setup in it (as they were before, while it was working under 2.00.13, verified the same), but while that particular node registers, updates plugins, etc, unlike my other 3 nodes (which work correctly, and transcode perfectly) I get the following error for every attempted transcode

 

[h264_cuvid @ 0x559e07dba1c0] Cannot load libnvcuvid.so.1

[h264_cuvid @ 0x559e07dba1c0] Failed loading nvcuvid.

Stream mapping:

Stream #0:0 -> #0:0 (h264 (h264_cuvid) -> hevc (hevc_nvenc))

Stream #0:1 -> #0:1 (copy)

Error while opening decoder for input stream #0:0 : Operation not permitted

Link to comment
4 hours ago, iammeuru said:

I don't know that this is correct exactly. I have not set interalNode=true after the update, and still have the node docker image running. I have all of the appropriate settings for my GPU setup in it (as they were before, while it was working under 2.00.13, verified the same), but while that particular node registers, updates plugins, etc, unlike my other 3 nodes (which work correctly, and transcode perfectly) I get the following error for every attempted transcode

 

[h264_cuvid @ 0x559e07dba1c0] Cannot load libnvcuvid.so.1

[h264_cuvid @ 0x559e07dba1c0] Failed loading nvcuvid.

Stream mapping:

Stream #0:0 -> #0:0 (h264 (h264_cuvid) -> hevc (hevc_nvenc))

Stream #0:1 -> #0:1 (copy)

Error while opening decoder for input stream #0:0 : Operation not permitted

 

The Docker deps and ffmpeg being used is the same as before, no changes have been made to that.

 

Make sure you pull the latest image, sounds like you might still be on the old one with the internal node causing interference.

 

https://github.com/HaveAGitGat/Tdarr/issues/562

 

Link to comment
4 hours ago, HaveAGitGat said:

 

The Docker deps and ffmpeg being used is the same as before, no changes have been made to that.

 

Make sure you pull the latest image, sounds like you might still be on the old one with the internal node causing interference.

 

https://github.com/HaveAGitGat/Tdarr/issues/562

 

I forced an update to the docker image for the server (2.00.14). I shutdown my node docker image. I added new variables for nodeID, nodeIP, Node Port (as a port), NVIDIA_VISIBLE_DEVICES, NVIDIA_DRIVER_CAPABILITIES, internalNode, and the extra parameter "--runtime=nvidia" to the tdarr image template, and it didn't work (I don't see a node registering through the UI, and no mention of it in the logs), so now I'm just going to remove all of that, and try setting internalNode=false, and try using my previous setup with the tdarr_node docker image (2.00.14 also) as it was previously configured and working before the update from 2.00.13 to 2.00.14.

 

I assume that won't work, and if so, I'll just downgrade everything back to 2.00.13... I'm not sure why all of the needed (for internal node use) variables aren't in the template by default, because I feel like maybe I'm not needing all of those variables I added, or I've configured it wrong somehow, but I can't really see where I did it incorrectly. This has to be basically affecting everyone who was using their tdarr setup on their unraid server for local transcode work, so I think some change needs to happen to the image template to make this less confusing.

 

I've basically got a 2080super here I can't use whatsoever for tdarr transcode work, so I'm about 75% of my previous capacity in terms of workload handling, while right in the middle of transcoding the entirety of my media libraries.

 

I'm not really buying that the internal node was "interfering", because I saw the node registering with its appropriate name when I updated everything initially, but maybe someone here can disambiguate how to verify that, and I can check.

Link to comment
14 hours ago, HaveAGitGat said:

It's not broken. The new Server image contains a Node so you most likely have 2 Nodes running on the same port which are interfering with each other.

 

I've just released a small Server image update so you now have to set internalNode=true to enable the Node in the Server container. So pull the latest container and enable the internal Node as you see fit. Leaving it disabled should mean things run the same as before.

 

https://docs.tdarr.io/docs/installation/docker/run-compose/

 

I'm on the latest but I don't see those options.

Link to comment
2 hours ago, wgstarks said:

Should be there now. I see it in mine. You might need to check for updates maybe.

 

 

Still nothing, even after checking for updates.

Web UI is still failing semi-frequently (going unresponsive). 4 times so far today...

Edited by Econaut
Link to comment
10 hours ago, Econaut said:

 

I'm on the latest but I don't see those options.

I didn't have the options in the template, I just added those variables to it, though I'm now running perfectly after deleting all but the internalNode one on the tdarr(server) container, and just setting it to false... I'm just using the external node as its own separate docker container. Apparently there was some sort of conflict, but it's working now...

 

Basically if you're coming from 2.00.13 to 2.00.14 and you have(had) a functional setup, ensure they're as up to date as possible, and just create a new variable in the template called interalNode, and set its value to false, then you should be able to connect as previously done, and it will work.

 

Maybe they'll update the template to have the correct variables to setup for internal node usage, or someone will make a guide on how to get it up with an internal node, and then maybe I'll switch, but probably not at this point, because this works and I don't see a reason to switch from a functional setup unless someone can describe some tangible benefits to it aside from losing a small amount of overhead running the second container. I do applaud the team for trying to get it all setup into one package for that reason, and it surely would've been the way I went if I'd started this a little later than I did.

  • Thanks 1
Link to comment
33 minutes ago, wgstarks said:

That has already been done.

Really? In unraid I show up-to-date, but I don't see any variables in the template for internal node stuff, other than the one I added to turn it off explicitly. I'm mid-transcode, and am not going to force update it to find out rn, but maybe I'm misunderstanding how I need to update it to get the latest and greatest?

Link to comment
8 minutes ago, iammeuru said:

Really? In unraid I show up-to-date, but I don't see any variables in the template for internal node stuff, other than the one I added to turn it off explicitly. I'm mid-transcode, and am not going to force update it to find out rn, but maybe I'm misunderstanding how I need to update it to get the latest and greatest?

My template updated just a few minutes after the pull request was merged today. I don't know enough about how the template updating works to offer an explanation about yours not updating. Perhaps @Squid might.

Link to comment

Hey, my install failed, is this variable correct? it says there is an invalid entry in docker pull down. What is the proper variable or is it port, should It be port or variable here, this is on the template so mabye there is a mistake, it fails unless i put it as a variable and when it does install it wont load web ui from the node nor node show up in tdarr. help me please, why is this installation failing?

NODEPORT.png

INVALID.png

Link to comment
3 hours ago, RYZEN5000 said:

Hey, my install failed, is this variable correct?

The only way Ive found to get a successful install is to delete the path for node port. This will allow the TDARR docker to run but you won't be able to use the internal node. You'll need to also run the TDARR_NODE docker. Or you can wait until the bugs are worked out in the template.

Link to comment
6 hours ago, wgstarks said:

The only way Ive found to get a successful install is to delete the path for node port. This will allow the TDARR docker to run but you won't be able to use the internal node. You'll need to also run the TDARR_NODE docker. Or you can wait until the bugs are worked out in the template.

I installed it by deleting the entry for node port completely and by creating a variable sort of like the one for server port but its not a port its a variable and it still doesn't work after the install. I wonder how long this will take before it gets addressed. months?...... I think its worth searching for an older version of the app where i can install it with a docker command line and mabye get it funtional that way but knowing that the version needs to match for all components I don't know if I can find the correct versions for all my windows nodes too.

Edited by RYZEN5000
Link to comment
5 minutes ago, RYZEN5000 said:

I installed it by deleting the entry for node port completely and by creating a variable sort of like the one for server port but its not a port its a variable and it still doesn't work after the install. I wonder how long this will take before it gets addressed. months?...... I think its worth searching for an older version of the app where i can install it with a docker command line and mabye get it funtional that way but knowing that the version needs to match for all components I don't know if I can find the correct versions for all my windows nodes too.

The new templaplate has already been merged and should be installed in the next few hours but if you want you can delete the node port and add a new port-

 

678069216_ScreenShot2022-02-24at3_11_25PM.thumb.png.f0d2170bb976297b0466584da9f59cb3.png

Link to comment
6 hours ago, RYZEN5000 said:

I installed it by deleting the entry for node port completely and by creating a variable sort of like the one for server port but its not a port its a variable and it still doesn't work after the install. I wonder how long this will take before it gets addressed. months?...... I think its worth searching for an older version of the app where i can install it with a docker command line and mabye get it funtional that way but knowing that the version needs to match for all components I don't know if I can find the correct versions for all my windows nodes too.

Tdarr is working now! Docker template is fixed now, I was playing around with docker compose scripts for a few hours trying to install it that way but I went back and tried it again and its been updated to beta version with node installation included. It installed the node automatically.

But what a pain in my arse it was soo difficult to install it by docker compose, it installed alright but the port mappings were not working at all, I'm glad somone fixed the docker template and included the node this time, thats a better way to do it, and here I was thinking it wouldn't get fixed for a very long time but it was faster than expected and I'm just getting ready to map for my other windows PCs to make them nodes now that all my programing from before is probably not going to work but we will see about that soon! Im Transcoding now, hope it doesn't take long to catch up I'm mostly converted to HEVC but each week I do another run and shrink my new downloads and keep on downloading, keep torrenting, I'll never stop, not till I cant fit anymore 16TB drives in my case, I've been torrenting on this new server since October 8th it was AGKSERVERS birthday and hes a real beast but I haven't fully added all the drives yet just 48TB plus 5.5TB cache. 4 more bays left them I'm adding more drive racks inside because the case is huge its packed but theres room if I move some fans and wires I think I can fit 20 drives and I want to buy more 16TB or 20TB drives. Tdarr is working! Health checks are intensive on my ryzen 7 5800X but when they finish my cpu will run around 5% transcoding with GPU. I hope my build inspires others to build a nice new Unraid server. It is my favorite thing to play with.AGKSERVER1.thumb.png.cbbd203594b028bf31ec0e052ab08257.png

TDARR12.png

20220112_133918.jpg

20220112_200843.jpg

20220112_200828.jpg

20220112_134002.jpg

20220112_133859.jpg

20220112_175025.jpg

20220112_175900.jpg

jellyfin1.png

 

 

 

 

 

 

jellyfin2.png

 

 

 

jellyfin3.png

 

AGKSERVER2.png

20220224_195304.jpg

 

Edited by RYZEN5000
  • Like 2
Link to comment

i found something strange.

I have only one node. Server and node are dockers on my unraid server with default parameters except for ip of course.

If i change the node name which is defaulted at "node name", the queue is stuck.

As soon as i set it to "node name" back the workers start.

Link to comment
2 hours ago, caplam said:

i found something strange.

I have only one node. Server and node are dockers on my unraid server with default parameters except for ip of course.

If i change the node name which is defaulted at "node name", the queue is stuck.

As soon as i set it to "node name" back the workers start.

What are the transcode options set at for the new node?

 

 

4B5F7B18-54B7-424F-8845-38FE09CF65A3.thumb.jpeg.41e4089cd417f355b6e434fa1dd47a0f.jpeg

Link to comment

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.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.