1 hour ago, IamSpartacus said:

 

Here you go.  https://gist.github.com/Xaero252/9f81593e4a5e6825c045686d685e2428.  You can put this script in User Scripts so that you can run it each time Plex gets updated.

 

And yes, I'm seeing that GPU memory usage is about what I'm seeing as well.

Thanks, I'll go research how to actually do that. LOL

 

So that's normal usage for GPU, thought it might be too low but cool if that sounds right.

On 3/26/2019 at 6:50 PM, CHBMB said:

I can't tell you why your copy of that file is getting deleted, but that's what I would bet is causing it.  Try running the above command and see what the output is, both from a fresh boot and after the problem occurs.

The issue was the recycle bin plugin. I didn't have it excluding the /tmp folder. The recycle bin plugin was prompting the cleaning of that folder and removing GPU data on the Nvidia plugin page. 
Thanks for all your work that you put into this as well all the containers you maintain. 
LSIO team rocks!!!

The issue was the recycle bin plugin. I didn't have it excluding the /tmp folder. The recycle bin plugin was prompting the cleaning of that folder and removing GPU data on the Nvidia plugin page. 
Thanks for all your work that you put into this as well all the containers you maintain. 
LSIO team rocks!!!

Ah, nice find, I'll take a look at the recycle bin plugin and see if I can get a work around for the issue. Thanks for reporting back.

Sent from my Mi A1 using Tapatalk

v6.7.0rc7 uploaded

On 4/2/2019 at 9:49 AM, IamSpartacus said:

What's the current easiest method to be able to constantly monitor watch nvidia-smi?  Netdata docker?

Yes.

 

I have submitted a request with telegraf to expand the api calls for use with Grafana.

 

https://github.com/influxdata/telegraf/issues/5564

Edited April 5, 20197 yr by Dazog

Hi All

 

I'm running UnRaid 6.6.7 as a VM on ESXi 6.7. I'm passing my GT 710 graphics card through (see pic) however the nvidia setting reports it can't see the card correctly:

 

"Unable to dertermine the device handle got GPU 0000:1B:00.0: Unknown error"

 

Checking the logs I see the following entry:

 

kernel: NVRM: RmInitAdapter failed! (0x23:0x56:483)
kernel: NVRM: rm_init_adapter failed for device bearing minor number 0
kernel: NVRM: RmInitAdapter failed! (0x23:0x56:483)
kernel: NVRM: rm_init_adapter failed for device bearing minor number 0

 

 

Has anyone got it to work with UnRaid as a VM? Anyone give me any pointers?

 

Edited April 5, 20197 yr by IrishBiker

8 minutes ago, IrishBiker said:

Hi All

 

I'm running UnRaid 6.6.7 as a VM on ESXi 6.7. I'm passing my GT 710 graphics card through (see pic) however the nvidia setting reports it can't see the card correctly:

 

"Unable to dertermine the device handle got GPU 0000:1B:00.0: Unknown error"

 

Checking the logs I see the following entry:

 

kernel: NVRM: RmInitAdapter failed! (0x23:0x56:483)
kernel: NVRM: rm_init_adapter failed for device bearing minor number 0
kernel: NVRM: RmInitAdapter failed! (0x23:0x56:483)
kernel: NVRM: rm_init_adapter failed for device bearing minor number 0

 

 

Has anyone got it to work with UnRaid as a VM? Anyone give me any pointers?

 

We only do testing when unraid is running on bare metal / host os. I suspect it could be to do with nvidia not liking their cards running in a virtual environment which is why you're getting this issue.

So I went to do an upgrade from the RC6 to RC7 and I couldn't get it to boot. Rolled back to RC6(nvidia) and it still wouldn't boot. I keep getting an error in the log right before it gets to the login prompt that says "Error: Check Nvidia Driver!" and then it reports a segfault.

 

So neither the RC6 or RC7 will boot. Before I was giving me driver errors with the dynamix system temp plugin so I removed that and attempted it again but can't get past the nvidia driver error check. Not sure what I did wrong but even starting over with  a fresh install and then installing the nvidia version would not work. 

 

My GPU is a GTX970

 

I won't be able to do much until I get home at 6pm EST 

 

Just thought I'd share my issue

GPU Assignment dosen't work

Hi, im running into the problem that no matter wich "uuid" i enter as value Plex is always going to start transcoding with the first gpu in line. If I start my vm first, wich uses my GTX 1060, then plex is going to do the transcoding on the Quadro P400. Any suggestions how to solve this? Thx in advance

3 hours ago, IrishBiker said:

I'm passing my GT 710 graphics card through (see pic) however the nvidia setting reports it can't see the card correctly:

 

"Unable to dertermine the device handle got GPU 0000:1B:00.0: Unknown error"

do your card working correctly, if you pass it to Windows VM for example?

1 hour ago, uldise said:

do your card working correctly, if you pass it to Windows VM for example?

Yes, well it was recognized in Device Manager. I'm getting hold of a HP NVIDIA QUADRO P620. Will try that when it arrives. 

4 hours ago, IrishBiker said:

Yes, well it was recognized in Device Manager. I'm getting hold of a HP NVIDIA QUADRO P620. Will try that when it arrives. 

You may have to manually supply the BIOS for the consumer cards. The Quadro P620 should play well in a VM, however.

 

Edit:
I see you are running esxi, my bad;

try doing this:

 right click your VM > Edit Settings > VM Options > Advanced > Edit Configuration > Add Parameter:

hypervisor.cpuid.v0 = "FALSE"

Someone on Reddit reported that to work getting the drivers installed on Windows, hopefully it does for Linux as well. Should only apply to *consumer* cards. The Quadro, again, should play well in a VM.

Edited April 5, 20197 yr by Xaero

I figured out my failure... If you don't wait for the "Done" button to appear and you exit the window any sooner, you will corrupt the install...

 

Is there any way to show write process in the window after the checksum pass? That way we know it's still in operation?

20 hours ago, MowMdown said:

I figured out my failure... If you don't wait for the "Done" button to appear and you exit the window any sooner, you will corrupt the install...

 

Is there any way to show write process in the window after the checksum pass? That way we know it's still in operation?

I've made the routine a little more verbose and added a warning.  Just update the plugin....

 

11 minutes ago, CHBMB said:

I've made the routine a little more verbose and added a warning.  Just update the plugin....

 

Go one step further and remove the "x" altogether.  Add this in before the call to openBox()

$("#sb-nav-close").css("visibility","hidden");

 

3 minutes ago, Squid said:

Go one step further and remove the "x" altogether.  Add this in before the call to openBox()

$("#sb-nav-close").css("visibility","hidden");

 

Cool, gives me an excuse to check the CA API for updating....

 

4 hours ago, CHBMB said:

Cool, gives me an excuse to check the CA API for updating....

 

What's to check?  It's 3 lines of code, only 1 of which does anything.

What's to check?  It's 3 lines of code, only 1 of which does anything.

Until I see it working with my own two eyes.....

TL:DR It did work, it was uneventful.....

Sent from my Mi A1 using Tapatalk

55 minutes ago, CHBMB said:

Until I see it working with my own two eyes.....

Ye of little faith

well, i have also this:

 

and Plex ist stuttering well. HP NHL40, unRaid nvidia Version  6.7.0-rc7.

 

Edited April 8, 20197 yr by dibux

42 minutes ago, dibux said:

well, i have also this:

You should edit your screenshot to get rid of the plex claim token.

1 hour ago, dibux said:

well, i have also this:

 

and Plex ist stuttering well. HP NHL40, unRaid nvidia Version  6.7.0-rc7.

 

The nvidia 710 doesn't seem to have NVENC so it's not going to work / plus not supported by nvidia-smi.

I am getting an unknown flag error. I can't figure this out, it worked a week ago then suddenly I checked on plex and the app was entirely gone, I was able to restore it from the Apps tab. Is this common? I can't tell if it is a problem with the specific grab of PLEX I got from Apps. Is there a different one that people use? This is the error I receive.

 

/usr/local/emhttp/plugins/dynamix.docker.manager/scripts/docker run -d --name='Plex' --net='eth1' --ip='192.168.0.3' --privileged=true -e TZ="UTC" -e HOST_OS="Unraid" -e 'PLEX_CLAIM'='claim-***********' -e 'PLEX_UID'='99' -e 'PLEX_GID'='100' -e 'VERSION'='latest' -v '/mnt/user/Yggdrasil/Transcodes/':'/transcode':'rw' -v '/mnt/user/Yggdrasil/Movies/':'/Movies':'rw' -v '/mnt/user/Yggdrasil/TV Shows/':'/TV Shows':'rw' -v '/mnt/user/Yggdrasil/':'/Data':'rw' -v '/mnt/user/Yggdrasil/Plex Media Server/':'/config':'rw' --runtime=nvidia 'plexinc/pms-docker' 
unknown flag: --runtime
See 'docker run --help'.

4 minutes ago, djlcurly said:

I am getting an unknown flag error. I can't figure this out, it worked a week ago then suddenly I checked on plex and the app was entirely gone, I was able to restore it from the Apps tab. Is this common? I can't tell if it is a problem with the specific grab of PLEX I got from Apps. Is there a different one that people use? This is the error I receive.

 

/usr/local/emhttp/plugins/dynamix.docker.manager/scripts/docker run -d --name='Plex' --net='eth1' --ip='192.168.0.3' --privileged=true -e TZ="UTC" -e HOST_OS="Unraid" -e 'PLEX_CLAIM'='claim-***********' -e 'PLEX_UID'='99' -e 'PLEX_GID'='100' -e 'VERSION'='latest' -v '/mnt/user/Yggdrasil/Transcodes/':'/transcode':'rw' -v '/mnt/user/Yggdrasil/Movies/':'/Movies':'rw' -v '/mnt/user/Yggdrasil/TV Shows/':'/TV Shows':'rw' -v '/mnt/user/Yggdrasil/':'/Data':'rw' -v '/mnt/user/Yggdrasil/Plex Media Server/':'/config':'rw' --runtime=nvidia 'plexinc/pms-docker' 
unknown flag: --runtime
See 'docker run --help'.

Delete --runtime=nvidia and try typing it in manually NOT copying/pasting from the forum.  For some reason it adds extra (non-visible) characters.

1 hour ago, CHBMB said:

Delete --runtime=nvidia and try typing it in manually NOT copying/pasting from the forum.  For some reason it adds extra (non-visible) characters.

I feel very dumb... but oh well, it worked