Cdc ncm

Hello,

I am having issue trying to connect in headless device mode and it appears to me that its a driver issue. While in device manager I see that I have CDC NCM and its missing a driver. That being said I did read in another post that it was the version I was using(SDK) so i was looking to confirm that and if there was a work around?

Hello,

This is the wrong forum, there are no technical resources monitoring this section. What Nvidia product or service is this referring to? I can move this to the proper forum for you.

Tom

Jetson Nano

Moving this to the Jetson Nano forum.

thanks

Hi,
Please share more information about CDC NCM. We don’t have experience on this and it probably is not enabled in tegra_defconfig by default. With more information, other users can see this and share experience.

Are you referring to “CDC ACM” and not “CDC NCM”? The CDC ACM serial UART driver is quite common in the embedded world, so I am wondering if this is just a typo.

When I open up device manager on my Win10 box I have an entry under Other Devices CDC NCM. This is the Jetson Nano and it appears that I fixed the issue when I changed to the image to the NVIDIA DLI AI Jetson Nano SD Card Image v1.1.1 and not the SDK that I had in there. That being said is there a way to get the jupyter server working on the SDK version of the image?

Aha! Windows seems to use a different name for the driver of the same UART. It is also Windows which needs the driver, not the Jetson. I’m not sure where Windows would get the driver at, but what I know as the “CDC ACM” serial UART driver on Linux is a very common piece of hardware. Some naming convention may be making it so Windows does not understand this is the same driver. I have not used this with Windows, so I am unsure where you would get the driver or how to associate it with the device. Changing the Jetson itself would not have any effect on it if this is the case.

So I noticed that even with the other image I still see that issue but is does work so my guess would be that I would just need to install jupyter server on to the SDK.

How is networking provided between the Nano and the Windows machine? Are both connected to a common switch/router? Either way, can you ping the Jetson’s IP address? If you can ping, then what error message do you see from the web browser on the Windows side when trying to reach the jupyter web URL? You would of course need this installed on the Jetson and running to reach it, but it is also possible Windows is firewalling.

I am in headless mode via a USB cable and connected to http://192.168.55.1:8888/lab. It all appears to work now. The only issues I am having are with the learning curve now, trying to figure it all out. Currently working on the Hello Camera lab and getting some errors running the code but i do get an image so the camera is working