witek

[witek]

Hi,

If pactl list sources | grep nx did not return any output I suggest to update nxnode configuration file. To do that run:

sudo /usr/NX/bin/nxnode --audiosetup <path>

where <path> is an optional argument to specify the absolute path to Pulseaudio configuration files (e.g. client.conf, daemon.conf, …) in a non standard location.

After that restart NoMachine server and verify if audio works when you connect to CentOS:

sudo /etc/NX/nxserver --restart

Did you change anything inside Pulseaudio configuration files?

[witek]

Hi,

Have you tried to choose explicitly the sound device used by nxplayer on CentOS 7? Run the NoMachine session, go to Menu Panel -> Audio settings and select the sound device from the list (instead of using “Autodetect”):

https://www.nomachine.com/DT10O00158#6.1

Are you able to get sound from any other application on this CentOS?

[witek]

Hi,

It seems that NoMachine audio service is not able to connect to Pulseaudio when nxplayer wants to get a list of available sound devices. In this case audio will not work inside NoMachine session. How do you open nxplayer? Through NoMachine tray? Please, can you try to open nxplayer via terminal (run /usr/NX/bin/nxplayer) and check if there is any sound device available in the sound settings when you connect to some server?

[witek]

Hi,

We could not reproduce this issue in your evironment. What output device are you using on laptop (built in speakers or something else)? Did you try to uncheck and check again “Mute audio on the server” option?

Please, can you provide also the output of these commands when the NoMachine session is running:

1. List files in /dev/snd:

ls -lrt /dev/snd

2. Check if PCM mixer control is available for specified sound card:

for dev_id in $(ls /dev/snd | grep controlC | sed -e 's/controlC//g'); do amixer -c $dev_id get PCM; done

3. List available sinks

pacmd list-sinks > list_sinks.log

and attach “list_sinks.log” in you reply.

[witek]

Hi,

Sound on the server will be not muted if NoMachine audio service does not work properly. In this case it seems that nxaudio kernel extension is not loaded. If you did not approve NoMachine kernel extensions, you can do this by following steps below:

1. Try to load nxaudio module (the path is different than on MacOS 10.15):

sudo kextutil -vvv /System/Library/Extensions/nxaudio.kext/

2. Access System Preferences -> Security&Privacy -> General panel. Click on the “Allow” button (see kext_approval.png).

3. Check if nxaudio is loaded properly:

kextstat | grep nxau

Output of the command above should be similar to

265 0 0xffffff7f83cd6000 0x5000 0x5000 com.nomachine.driver.nxau (4.1.b2) 78172556-B6A4-3112-BA41-E79D18108476

4. Restart NoMachine server using command:

sudo /etc/NX/nxserver --restart

If you want to transmit audio from the server to the client, Microphone permission has to be also properly set. When you connect to the iMac for the first time you should get a request that “nxnode would like to access the microphone” (see microphone_request.png). After accepting it, you should be able to see nxnode inside Microphone permission (see microphone_settings.png).

If audio still does not work, please send us the output of kextutil command (point 1) in your reply.

Attachments:

1. 

   kext_approval.png
   

2. 

   microphone_request.png
   

3. 

   microphone_settings.png
   

[witek]

So I guess that the problem is still there, isn’t it?

Yes, nxaudio kext cannot be loaded because it is rejected. Without it, audio inside NoMachine session will not work. I suspect that it is related to the article that I mentioned in my previous message (https://www.nomachine.com/AR01P00962).

To resolve this problem, please run the command sudo kextload /Library/Extensions/nxaudio.kext on your Mac and then access System Preferences -> Security&Privacy -> General panel. Click on the “Allow” button (see attachment) and restart NoMachine server using command:

sudo /etc/NX/nxserver --restart

Attachments:

1. 

   kext_approval.png
   

[witek]

Hi,

It seems that nxaudio kernel extension (kext) is not loaded on MacOS Catalina. It is responsible for creating NoMachine audio adapter. You can try to load it manually by following steps below:

1. Load nxaudio kext using command:

sudo kextutil -vvv /Library/Extensions/nxaudio.kext

2. Check if nxaudio is loaded properly:

kextstat | grep nxau

Output of the command above should be similar to

265 0 0xffffff7f83cd6000 0x5000 0x5000 com.nomachine.driver.nxau (4.1.b2) 78172556-B6A4-3112-BA41-E79D18108476

3. Restart NoMachine server:

sudo /etc/NX/nxserver --restart

4. Run NoMachine session and verify audio.

If no output is shown in second step, we will need result of kextutil command from point 1. Did you also approved loading NoMachine kernel extensions in the Security&Privacy panel? More information you can find here:

https://www.nomachine.com/AR01P00962

[witek]

Hi,

When you connect to the Windows machine, NoMachine will take default output device and read audio stream from this device to transmit it to the other side. If you want to change it, you have to close the session, switch to another sound device and then connect again via NoMachine. Based on this information, can you set speakers as the default output device and then connect to the server?

You can also check if the “exclusive mode” is not enabled for this device:

1. Open Control Panel -> Hardware and Sound -> Sound and select the Playback tab.
2. Select correct device from the list and click Properties.
3. Go to Advanced tab and ensure that options in “Exclusive Mode” are disabled.

[witek]

Hi,

What is your default input and output audio device on Windows? You probably have “NoMachine Microphone Adapter” set as a default output device. In this case NoMachine records audio stream from this device and transmits it to the other side, so that is why you can hear your voice on the client side.

If “NoMachine Microphone Adapter” is your only available option for output device you can try to add “NoMachine Audio Adapter”. Please run the following command as administrator using correct path of NoMachine executables, for example:

"C:\Program Files (x86)\NoMachine\bin\nxservice64.exe" --deviceadd --inffile "C:\Program Files (x86)\NoMachine\bin\drivers\nxaudio\amd64\nxaudio.inf" --hardwareid *NXVAC

After that, set “NoMachine Audio Adapter” as the default output and “NoMachine Microphone Adapter” as the default input in Windows sound settings. If you are doing it in a NoMachine session you can just use the command below to respawn nxnode process which will start audio service with new selected default device:

taskkill /f /fi "IMAGENAME eq nxnode.bin" /fi "USERNAME eq SYSTEM"

[witek]

In this case you can try to use PulseAudio loopback module which will redirect audio stream from your input device to the default output device. Please run the following command on the server and then verify audio on the client side:

pactl load-module module-loopback source=alsa_input.usb-Burr-Brown_from_TI_USB_Audio_CODEC-00.analog-stereo

This loaded loopback will be also visible in pavucontrol (Playback and Recording tab).

[witek]

Hi,

Do you want to transmit audio stream from the input device “PCM2901 Audio Codec Analog Stereo” to the client side (Ubuntu and MacOS)?

NoMachine uses default sink (output device) on the server side (Linux), then reads audio stream using monitor of this sink and transmits data to the other side. You cannot change this monitor when NoMachine start to record audio. That’s why you are not able to select source other than “Monitor of PCM2901 Audio Codec Analog Stereo” in the Recording tab of pavucontrol.

What is your default sink (output device) when the NoMachine session is running? You can check this in the pavucontrol or by using command in terminal:

pactl info | grep Default

[witek]

Hi,

You are using Windows OS on the server machine, right?

In this case, when you connect to the server, NoMachine will check what the default audio output device is and will transmit audio stream from this device to the other side.

If you want to change this default device used by NoMachine, you have to do this in Windows sound settings, then close the NoMachine session and connect again.

There is no option for now to choose it in NoMachine application when the session is running.

[witek]

Hi,

We have checked your environment but we did not notice any problem like that using Ableton Live.

NoMachine audio devices should be only used on the server side so on the client side your default audio output device should be set to headphones, speakers etc.

On the server side NoMachine application will take default output device and read audio stream from this
device (to transmit it to the other side). If you want to change default output device, you have to close the session
and connect again via NoMachine.

If you are able to hear audio from other application than NoMachine on the client
side, I suggest to focus on the remote machine.

1. What version of Ableton Live do you use?

2. Does audio work always correctly when you are not running Ableton Live?

3. What is your default sound output device? Can you please attach screenshot
of Playback tab in Sound panel? Go to Control Panel -> Hardware and Sound -> Sound.

4. Please check also if your default sound output device is running in the exclusive mode
and try to disable it.

https://answers.microsoft.com/en-us/windows/forum/all/disable-exclusion-mode-windows-audio-driver/92c92324-f189-4ae4-ba14-7c5dda7e6183

[witek]

Hi,

NoMachine on the client side on the Linux system is not creating any source or sink so because of that
you will get no result from these commands. NoMachine application should use default sound output (headphones, speakers etc.). If you are able to hear some audio from other applications I will suggest to leave it for now and focus on MacOS.

1. Do you use NoMachine 6.10.12 on this mac 10.15?

2. When you connect to the mac 10.15 for the first time via NoMachine
you should get request which reports that “NoMachine” would like to
access the microphone. Did you get it?

3. Please can you open Settings -> Security & Privacy -> Privacy -> Microphone panel and check if
“NoMachine” is available there (like on the attached screenshot)?

Attachments:

1. 

   microphone_permission.png
   

[witek]

Hi,

The audio quality slider is related to the Opus sample rate, frame size, channels etc. All these things affect how fast the audio stream can be processed and transmitted to the other side. We are using variable bitrate encoding
which means that bitrate can be changed depending on the audio stream to achieve more consistent quality.

1. The lowest point (the best performance):
Opus sample rate: 12000 Hz
Number of channels: 1

2. Middle position (compromise between performance and quality)
Opus sample rate: 24000 Hz
Number of channels: 2

3. The highest point (the best quality)
Opus sample rate: 48000 Hz
Number of channels: 2

[Author]

Posts