AGX Xavier is not responding

Autonomous Vehicles DRIVE AGX Xavier DRIVE AGX Xavier General
,
1

Please provide the following info:
Hardware Platform: DRIVE AGX Xavier™ Developer Kit
Software Version: DRIVE Software 10
Host Machine Version: native Ubuntu 18.04
SDK Manager Version: 1.2.0.6733

Hello everyone,

we suddenly having problems interacting with our Nvidia Drive AGX Xavier. We are using it now for about half a year and never really had a problem like that before:

When starting up the device, the fan starts working and everything seems normal. However, we don’t see a picture on the connected display, that is for both Xavier A and B.

Also, we cannot connect via ssh or minicom.

We just flashed it again with Drive OS 10, which seemed working, but that didn’t changed anything.

Is this a hardware error or is there anything else we can try?

Your help is much appreciated.

Best regards,
Thilo

2

Hi @t.mueller,

Do you have any suspicious messages on your host or target system? Could you try on another host system?

3

Hi @VickNV,

Using another host system doesn’t make a difference.

We don’t get any suspicious messages on the host system. However, it seems that the host system is unable to detect the target in any way.

Also to mention that we used to get the following error during flashing:

Error in local apt settings
SDK Manager received errors while using apt commands on your system.

We resolved this problem however by reinstalling Ubuntu. Flashing worked then fine.

Thanks,
Thilo

4

If you can flash DRIVE Software 10.0 successfully, that means serial communication works well.
But you said you cannot access Xavier A/B with minicom. Do you try ttyUSB2/ttyUSB6 mentioned in https://docs.nvidia.com/drive/drive_os_5.1.6.1L/nvvib_docs/index.html#page/DRIVE_OS_Linux_SDK_Development_Guide/MCU%20Setup%20and%20Configuration/mcu_uart.html#?

Are you able to list out all the devices?
$ lsusb -d 0x0403:
Bus 001 Device 015: ID 0403:6011 Future Technology Devices International, Ltd FT4232H Quad HS USB-UART/FIFO IC
Bus 001 Device 014: ID 0403:6011 Future Technology Devices International, Ltd FT4232H Quad HS USB-UART/FIFO IC
Bus 001 Device 013: ID 0403:6011 Future Technology Devices International, Ltd FT4232H Quad HS USB-UART/FIFO IC
Bus 001 Device 012: ID 0403:6011 Future Technology Devices International, Ltd FT4232H Quad HS USB-UART/FIFO IC

Could you also share your flash log (~/.nvsdkm/logs/DRIVE_Software_10.0_Linux_OS/NV_FLASH_XAVIER_PDKFLASH_PARALLEL_COMP.log)? Thanks!

5

Hi @VickNV,

Yes, we tried connecting via minicom ttyUSB2/ttyUSB6 as described but with no success.

We can list out all devices:
$ lsusb -d 0x0403:
Bus 001 Device 010: ID 0403:6011 Future Technology Devices International, Ltd FT4232H Quad HS USB-UART/FIFO IC
Bus 001 Device 009: ID 0403:6011 Future Technology Devices International, Ltd FT4232H Quad HS USB-UART/FIFO IC
Bus 001 Device 008: ID 0403:6011 Future Technology Devices International, Ltd FT4232H Quad HS USB-UART/FIFO IC
Bus 001 Device 007: ID 0403:6011 Future Technology Devices International, Ltd FT4232H Quad HS USB-UART/FIFO IC

NV_FLASH_XAVIER_PDKFLASH_PARALLEL_COMP.log (132.0 KB)

Please find attached the requested log file.

Thanks a lot for your help,
Thilo

7

I didn’t find anything abnormal in your NV_FLASH_XAVIER_PDKFLASH_PARALLEL_COMP.log.
Please provide the dmesg log around ttyUSB of your host system. It’s worth checking if there are any suspicious messages.

9

Hi @VickNV,

the dmesg log seems pretty normal I think:

$ dmesg | grep ttyUSB
[   53.154908] usb 1-1.4: FTDI USB Serial Device converter now attached to ttyUSB0
[   53.157374] usb 1-1.4: FTDI USB Serial Device converter now attached to ttyUSB1
[   53.158134] usb 1-1.4: FTDI USB Serial Device converter now attached to ttyUSB2
[   53.158786] usb 1-1.4: FTDI USB Serial Device converter now attached to ttyUSB3
[   53.159088] usb 1-1.5: FTDI USB Serial Device converter now attached to ttyUSB4
[   53.159399] usb 1-1.5: FTDI USB Serial Device converter now attached to ttyUSB5
[   53.164029] usb 1-1.5: FTDI USB Serial Device converter now attached to ttyUSB6
[   53.164613] usb 1-1.5: FTDI USB Serial Device converter now attached to ttyUSB7
[   53.165180] usb 1-1.6: FTDI USB Serial Device converter now attached to ttyUSB8
[   53.165743] usb 1-1.6: FTDI USB Serial Device converter now attached to ttyUSB9
[   53.166033] usb 1-1.6: FTDI USB Serial Device converter now attached to ttyUSB10
[   53.166214] usb 1-1.6: FTDI USB Serial Device converter now attached to ttyUSB11
[   53.166395] usb 1-1.7: FTDI USB Serial Device converter now attached to ttyUSB12
[   53.166613] usb 1-1.7: FTDI USB Serial Device converter now attached to ttyUSB13
[   53.166799] usb 1-1.7: FTDI USB Serial Device converter now attached to ttyUSB14
[   53.166972] usb 1-1.7: FTDI USB Serial Device converter now attached to ttyUSB15
10

Hi @VickNV,

as it seems that the USB connection is the only one working, is there a way we can communicate with the Xavier or see what is going on there?

Thanks,
Thilo

11

Please you provide more complete dmesg output (at least around the ttyUSB context).

Have you done anything special prior to seeing the problem? Is there any HDMI signal detected by the monitor?

If you are able to flash, that means aurix access still works. Could you help try if “tegrareset” (https://docs.nvidia.com/drive/drive_os_5.1.6.1L/nvvib_docs/index.html#page/DRIVE_OS_Linux_SDK_Development_Guide/MCU%20Setup%20and%20Configuration/mcu_console_usage.html#wwpID0EEHA) helps?

Please also check “Figure 2-3. LED Locations” of DRIVE AGX Developer Kit Mechanical & Installation Guide (PDF) to see if anything is abnormal.

Thanks!

12

It turned out that minicom actually works! It was just that we got confused by the empty terminal output. You just have to press a random key to see something on the minicom shell.

The monitor still doesn’t detect any HDMI signal, however.

Thanks for your help.

13

Thanks for your reply. For monitor detecting no signal, you can try with other monitors and check dmesg on the target system.