My sd cards with ubuntu inside don't boot anymore

Sorry, no, wifi cannot work if you know how the boot flow works on jetson nano.

Just in case you don’t know. What you are doing now (flashing with sdkmanger) is a very common thing that happens on every jetson platform (nano/tx2/xavier/orin) everyday. This is a basic operation for any jetson devices.

ok but is not there an alternative procedure to follow when the flashing of the SDK fails like in my case ? The error that I’m getting is hardware ? can’t be fixed anymore if the USB controller is not recognized anymore ?

Actually I don’t think your board is broken… for now, most cases here are the host side issue or cable issue… For example, your micro usb cable needs to have data line. It cannot be cheap cables which only has power line for charging.

Please make sure the board is in recovery mode and the “lsusb” command on your host can see nvidia device or not.

Didn’t put the board in recovery mode is the most common mistake for newbies.

I tried a lot of USB cables. No way. Most of the time I did not get any error. Just the board is not recognized. Let’s make a thing. Can you suggest to me a cable which will work for sure ? Please point me on the right USB cable with the relative shop where I can buy it. (amazon is good),thanks.

Then I don’t think it is usb issue… could you share a photo of how is your connection now?

Also, your host machine could not be a VM. Just a reminder.

If you really concern about your usb cable, you can use other devices to test if it can detect by your host… this can prove your cable has data line…

Just as below user is doing here…

→ I have verified that my micro usb to usb cable is functioning by plugging it into another device and checking lsusb

I have connected my phone (samsung note 3) throught the USB port of my PC using the same usb cable that I’ve used on the jetson nano and it has been recognized without reporting errors :

[29461.592899] usb 1-6: new high-speed USB device number 14 using xhci_hcd
[29461.741974] usb 1-6: New USB device found, idVendor=18d1, idProduct=4ee7, bcdDevice= 2.32
[29461.741979] usb 1-6: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[29461.741983] usb 1-6: Product: SAMSUNG_Android
[29461.741986] usb 1-6: Manufacturer: SAMSUNG

so,is the cable working ? can we exclude it ? Now I’m going to post a pic of how I have closed the pins that should put the board in recovery mode.

Here you go :

you can see what are the pins that I have covered with the jumpers.

What I do,with those pins covered is the following workflow :

  1. Connect the micro USB cable to the micro USB port on the Jetson Nano, without pluggin the other end to the PC yet

  2. connect the Jetson Nano to the DC jack J25

  3. connect the other end of the micro USB cable to the PC

  4. remove the jumper wire from the FRC pins. (I have also tried without doing that)

actually the board is not recognized at all.

Is your power supply from the DC jack or it is still from micro usb?

If it it still from micro usb, please make it use the DC jack as power supply.

Also, is the dmesg on your host side still same as previous one after you connect the micro usb cable to host?

Could you take photos of not ony jetson side but the whole connection between jetson and your host?

the power cable is connected to the adapter ; the usb cable to the USB (2.0) port of the PC ; and the jumpers cover the respective pins. Nothing happens on ubuntu 18.04 when I issue the command

lsusb | grep -i nvidia

on the terminal. The yellow light turns on only when I connect the power cable,not when I connect the USB cable.

I mean did host side dmesg (kernel log) still spew lots of error when you connect the micro usb cable?

nope. Between the dmsg messages there is nothing referred to the jetson nano. It is not detected.

dmesg

[   46.906696] mc: Linux media interface: v0.10
[   47.340864] usbcore: registered new interface driver snd-usb-audio
[   47.368929] intel_rapl_common: Found RAPL domain package
[   47.368930] intel_rapl_common: Found RAPL domain core
[   47.368930] intel_rapl_common: Found RAPL domain uncore
[   47.797583] ucsi_ccg 0-0008: con1: failed to register alternate modes
[   47.945917] nvidia-uvm: Loaded the UVM driver, major device number 510.
[   51.833449] ZFS: Loaded module v0.8.3-1ubuntu12.14, ZFS pool version 5000, ZFS filesystem version 5
[   80.580272] e1000e: eno1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx
[   80.580360] IPv6: ADDRCONF(NETDEV_CHANGE): eno1: link becomes ready
[  124.049522] systemd-journald[519]: File /var/log/journal/2e6059c368cb44469ea7714dbf0e15cb/user-1000.journal corrupted or uncleanly shut down, renaming and replacing.

Then I think your board is not in recovery mode. What is the serial console log on jetson in this situation?

This is what happens when I attach the serial cable :

dmesg :

[ 3037.161488] usb 1-13.3: new full-speed USB device number 14 using xhci_hcd
[ 3037.263390] usb 1-13.3: New USB device found, idVendor=10c4, idProduct=ea60, bcdDevice= 1.00
[ 3037.263395] usb 1-13.3: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[ 3037.263399] usb 1-13.3: Product: CP2102 USB to UART Bridge Controller
[ 3037.263402] usb 1-13.3: Manufacturer: Silicon Labs
[ 3037.263405] usb 1-13.3: SerialNumber: 0001
[ 3037.369326] usbcore: registered new interface driver usbserial_generic
[ 3037.369333] usbserial: USB Serial support registered for generic
[ 3037.392955] usbcore: registered new interface driver cp210x
[ 3037.393007] usbserial: USB Serial support registered for cp210x
[ 3037.393176] cp210x 1-13.3:1.0: cp210x converter detected
[ 3037.395748] usb 1-13.3: cp210x converter now attached to ttyUSB0

at the same time I put the board in recovery mode. Nothing happens.

No, I mean what gets printed on the jetson’s serial console when you put the jetson into recovery mode + power on?

I am not asking you to share me another host dmesg for connection usb-ttl cable.

I’m not sure to have understood how can I look what’s written on the “jetson’s serial console when I put the board into recovery mode + power on”. This is the tutorial that I followed :

as you can see he says to look between the dmesg messages.

After sometime new messages are appeared on the dmesg log :


[ 4137.170034] usb 1-13.1: new full-speed USB device number 15 using xhci_hcd
[ 4137.330071] usb 1-13.1: new high-speed USB device number 16 using xhci_hcd
[ 4137.430779] usb 1-13.1: New USB device found, idVendor=0955, idProduct=7f21, bcdDevice= 1.02
[ 4137.430784] usb 1-13.1: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[ 4137.430787] usb 1-13.1: Product: APX
[ 4137.430790] usb 1-13.1: Manufacturer: NVIDIA Corp.
[ 4137.715638] usb 1-13.1: USB disconnect, device number 16

as you can see the board had been detected and connected but also disconnected after some time. Are you there ?

Maybe one of my jumper does not fit correctly on the pins ? (I think the black one,because its size is a little different from the yellow one). Can you point me on a shop where I can buy some jumpers of the right size ?

Hi,

You need to understand what is the subject we are talking about here.

The reason I asked you to buy a usb-ttl cable is to dump the log “from jetson”.

So when I said I need a serial console log from jetson. I mean you need to open minicom /dev/ttyUSB0, dump and share what you saw from this “serial console”.

Dumping a dmesg from your “host” is just to make sure this serial console is working fine. It is not a log from jetson.
I already knew you can dump minicom log in previous conversations, so sharing me a dmesg from your host with usb-ttl cable case does not help here.


Please tell me if you are okay to understand above. You need to understand why the website asks you to try that but not just blindly follow.

Also, if you can see this log in your host dmesg, then it may be the timing to run sdkmanager…

[ 4137.430784] usb 1-13.1: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[ 4137.430787] usb 1-13.1: Product: APX
[ 4137.430790] usb 1-13.1: Manufacturer: NVIDIA Corp.
[ 4137.715638] usb 1-13.1: USB disconnect, device number 16

Does lsusb show device with this? If not, please again, share me the serial console log of the whole process… I mean log from /dev/ttyUSB0…