Jump to content
linuxserver.io

[Plugin] Linuxserver.io - Unraid Nvidia

883 posts in this topic Last Reply

Recommended Posts

4 minutes ago, trurl said:

The plugin is actually a custom build of Unraid that includes nvidia drivers.

 

If you upgrade Unraid, then you have removed the custom build.

 

If you upgrade the plugin, you have installed a new custom build of Unraid.

Thank you!

Share this post


Link to post

OK, fixed.

 

Here's the breakdown.

 

Tried to compile on v6.7.0 and it failed complaining of missing libraries.  Now one of the dependencies we use is Mesa which includes these libraries  So I tried a known version of Unraid which I compiled 8 days ago, v6.7.0rc8, using the same packages that I used for v6.7.0 today and it also failed to build with the same error.  There's been no updates to the Slackbuild of Nvidia drivers since I compiled v6.7.0rc8 so that ruled out a problem with that.  So I started to look at the packages I was installing to spot any issues.

 

There has been a major version change between the GCC compiler version Limetech use and the version I was pulling from slackware-current (v8.3 to v9.1) which I thought may have been the issue, so my working theory was downgrade GCC and the other GCC bits we use and see if that worked.  I started to build GCC from source, which was taking a VERY long time, as there are no longer any Slackware packages available to download GCC v8.3, but in the meantime managed to find an out of date Slackware mirror which still had them, so switched everything back to v8.3 expecting it to work.

 

Unfortunately, although everything looked "cleaner" when compiling, it still failed with the same error.  So shot that theory down.  I chose at this point to pull the already published DVB builds as I'd compiled them with v9.1 and it seemed like it could be asking for trouble.

 

I then took a look at Mesa again and a bit more closely and decided to examine the slackware-current changelog for any changes to that which may have caused the failure.  It was then  I noticed just this morning there had been a change to slackware-current which meant mesa now had a new dependency libglvnd which was previously not required as it was previously included in the mesa package.

 

Essentially, if I'd compiled this at the time of release (I couldn't I was asleep) then I wouldn't have hit this error until the next version of Unraid

 

Quote

libglvnd-1.1.1-x86_64-1.txz: Added. This is the GL Vendor-Neutral Dispatch library, which allows multiple drivers from different vendors to coexist on the same machine. When libglvnd is present, the NVIDIA driver will not overwrite any system files. Note that this is known to work when installing the NVIDIA driver using the .run installer. Other methods may require adjustment. This library is now a dependency of mesa

 

Once I added that package it all worked and compiled fine, so I tried booting my VM with the Nvidia build and that was successful but I don't have a card in it to test, so then I upgraded my bare metal Unraid with my 1050ti in it and it worked with no issues.

 

I've typed this out so those of you that wonder what the delays are have your answers.  If you follow all that then feel free to swing by Discord and offer to help out with stuff.

 

Basically Slackware doesn't spoon feed you, if you installed Mesa on Ubuntu for instance it would pull all the dependencies in automatically with it's package manager, Slackware you have to figure them out for yourself.  There are currently 90 dependencies for the Nvidia build.

  • Like 1
  • Upvote 5

Share this post


Link to post

So I have got everything up and running. But now I have got this error in plex. Any ideas on what i am doing wrong. Thanks.

 

Screen Shot 2019-05-11 at 11.02.08 AM.png

Share this post


Link to post
Posted (edited)
4 minutes ago, acozad1 said:

So I have got everything up and running. But now I have got this error in plex. Any ideas on what i am doing wrong. Thanks.

No.  We need far more information.  Unraid version (screenshot of the plugin page would be good for that)  Plex docker run command, any logs from Plex that might be useful.  Screenshots of your Plex setup like I posted in the setup guide at the beginning of this thread.  Are you using any of the extra scripts people have been posting in here for Plex?  At least give us something to go on.

 

So far it's kind of like you've pushed a car into the garage, turned to a mechanic and said, "What's wrong with it?"  Nobody can tell at the moment.

Edited by CHBMB

Share this post


Link to post
7 minutes ago, CHBMB said:

No.  We need far more information.  Unraid version (screenshot of the plugin page would be good for that)  Plex docker run command, any logs from Plex that might be useful.  Screenshots of your Plex setup like I posted in the setup guide at the beginning of this thread.  Are you using any of the extra scripts people have been posting in here for Plex?  At least give us something to go on.

 

So far it's kind of like you've pushed a car into the garage, turned to a mechanic and said, "What's wrong with it?"  Nobody can tell at the moment.

Did you get a chance to fix the issue I had?

 

 

Quote from nvidia forums:

Looking at the logs though I see the driver continuously initializing/deinitializing in fast succession which points to that you're missing the nvidia-persistenced. If you get additional load on the bus then this might run into timing problems possibly leading to the issue you're observing. Please enable the nvidia-persistenced to start on boot and check if that resolves the issue.

 

Thank you again for all the hard work to make this possible.

Share this post


Link to post
1 minute ago, Dazog said:

Did you get a chance to fix the issue I had?

 

 

Quote from nvidia forums:

Looking at the logs though I see the driver continuously initializing/deinitializing in fast succession which points to that you're missing the nvidia-persistenced. If you get additional load on the bus then this might run into timing problems possibly leading to the issue you're observing. Please enable the nvidia-persistenced to start on boot and check if that resolves the issue.

 

Thank you again for all the hard work to make this possible.

nvidia-persistenced is installed as part of the compile process.

So unless you know of a way to enable it if it's not running or at least check if it is, I got nothing,

Share this post


Link to post
Posted (edited)

yes, I am using this script for transcoding which worked when I was on rc8. But since the downgrade, which i think is giveing me this problem.

Screen Shot 2019-05-11 at 11.09.00 AM.png

Edited by acozad1

Share this post


Link to post
1 minute ago, acozad1 said:

yes, I am using this script for transcoding which worked when I was on rc8. But since the downgrade, which i think is giveing me this problem.

Screen Shot 2019-05-11 at 11.09.00 AM.png

I don't support the patch, try without and if it still isn't working then post what I asked.  Otherwise you'd need to figure it out for yourself or speak to whoever is maintaining that patch

Share this post


Link to post

Ok, Thank you for taking a look at it for me. I will do what you recommended and run it without and see what happens. By the way, you guys do amazing work. Thank you.

Share this post


Link to post
Posted (edited)

@CHBMBWhen you compiled version 6.7 final, did you have to substitute "kernel-modules-4.19.41-x86_64-1.txz" for "kernel-modules-4.19.42-x86_64-1.txz" and "kernel-headers-4.19.41-x86-1.txz" for "kernel-headers-4.19.42-x86-1.txz"?

Trying to compile the HP RMRR patched version but I can't find any traces to the .41 files.

Edited by AnnabellaRenee87

Share this post


Link to post
1 minute ago, AnnabellaRenee87 said:

@CHBMBWhen you compiled version 6.7 final, did you have to substitute "kernel-modules-4.19.41-x86_64-1.txz" for "kernel-modules-4.19.42-x86_64-1.txz" and "kernel-headers-4.19.41-x86-1.txz" for "kernel-headers-4.19.42-x86-1.txz"?

Trying to compile the HP RMRR patched version but I can't find any traces to the .41 files.

I didn't, but Slackware-current sometimes throws some issues I find where the packages have been updated but the manifest hasn't and I'm going to guess that's what you're being hit with.

 

If I were you, wait a day or two anyway as you'll probably want to modify your script some on my DVB changes I'm working on.  I think it's a recipe for disaster to use GCC 9.1 from slackware-current when Unraid was compiled with v8.3

 

Where's your github repo for your script and I'll try and make a PR later as I'm currently changing around a lot of stuff and it'll probably break your stuff.

Share this post


Link to post
3 minutes ago, CHBMB said:

I didn't, but Slackware-current sometimes throws some issues I find where the packages have been updated but the manifest hasn't and I'm going to guess that's what you're being hit with.

 

If I were you, wait a day or two anyway as you'll probably want to modify your script some on my DVB changes I'm working on.  I think it's a recipe for disaster to use GCC 9.1 from slackware-current when Unraid was compiled with v8.3

 

Where's your github repo for your script and I'll try and make a PR later as I'm currently changing around a lot of stuff and it'll probably break your stuff.

Thanks for the update, I was just running the same scripts you made before, here they are if ya wanna see them.

 

I figured they pulled from your DVB editions, I will wait a few days.

rmrr-patch-kernel-compile-module-1.sh rmrr-patch-kernel-compile-module-2.sh

Share this post


Link to post
Posted (edited)

I installed the driver, and rebooted, but when I try to open my VMs page (or even VM settings) it won’t load. My VMs don’t even start. Is there something I’m missing? Something else I noticed is that on my Dashboard page, none of my VMs or Dockers show in the spots. When I install regular unRAID 6.7 everything works fine. NVIDIA unRAID 6.7. 

Edited by urhellishntemre

Share this post


Link to post
37 minutes ago, urhellishntemre said:

I installed the driver, and rebooted, but when I try to open my VMs page (or even VM settings) it won’t load. My VMs don’t even start. Is there something I’m missing? Something else I noticed is that on my Dashboard page, none of my VMs or Dockers show in the spots. When I install regular unRAID 6.7 everything works fine. NVIDIA unRAID 6.7. 

Are any of your VMs set to autostart?

Share this post


Link to post
17 minutes ago, CHBMB said:

Are any of your VMs set to autostart?

Yes one is set to auto start. Should I go back to base 6.7, disable auto start and go back to NVIDIA?

Share this post


Link to post
27 minutes ago, urhellishntemre said:

Yes one is set to auto start. Should I go back to base 6.7, disable auto start and go back to NVIDIA?

Might not be a bad idea, is that VM using a passed through Nvidia card?

Share this post


Link to post
36 minutes ago, CHBMB said:

Might not be a bad idea, is that VM using a passed through Nvidia card?

Hahaha, yes it is. I can see the issue now. I will update later.  

Share this post


Link to post
Posted (edited)
9 hours ago, CHBMB said:

nvidia-persistenced is installed as part of the compile process.

So unless you know of a way to enable it if it's not running or at least check if it is, I got nothing,

Tracked down it's a amd/linux  kernel encryption thing.

https://devtalk.nvidia.com/default/topic/1043952/linux/linux-driver-410-73-gtx-980-nvrm-rminitadapter-failed-/2

 

 

 

 

 

Edited by Dazog

Share this post


Link to post

 

Im getting this over and over and over in my syslog since patching and implementing a 1070ti into my setup with plex. Is this an error? is it normal? does it even matter? I have the card in a 16x slot if this matters or not.

cap.PNG

Share this post


Link to post
On 5/11/2019 at 7:00 AM, CHBMB said:

We have to sleep.  Turnaround time is when I get to it, and if it builds successfully, which at the moment, it is not.....

 

ETA:  God only knows.....

 

Please remember that this is not official and relies on someone (me) manually running the builds, waiting for it to compile, (Typically around 30 mins using 32GB RAM and 24 cores) and uploading it to the server.  Delays can come from my other commitments, family, job, other life things, and as it states in the original post, we rely on a lot of upstream work, so there are no guarantees that we will release any given version.  Very dependent on things just working as we have no control over the upstream.

You're THE man! Great Team!

Share this post


Link to post

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.