I had to go back to the old powershell extension, as I could not make the new xonsh version work

Technical Support
, , , ,
1

I tried everything I could to make the new extension work, but I either get a link error when linking the Boost log Libraries, or a good link, but a C++ exception when I run my code and try to log a message in Boost.
Using the same Boost compiled libraries (1_82), it links fine, but won’t run, with updated Boost libraries (1_88 and 1_89) I get an undefined symbol with __sscanf.
So, I had everything working and the extension was marked with don’t update. I should have backed up the directory, but didn’t. I restarted VSC serveral times to make sure that the extension didn’t update.
So, Monday morning I got a VSC update, which caused an extension renew and it went in deleted all my working updates and all the powershell tasks.
So, I got most it back, but I am running into a issue with the last step where I get a prompt asking for a password when it is trying to copy the exe files:

Executing task: scp -i /home/steve/GimbalS/.conf/id_rsa -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no -P 2230 -pr /home/steve/GimbalS/aarch64/debug torizon@192.168.6.242:~/app

Warning: Permanently added ‘[192.168.6.242]:2230’ (ED25519) to the list of known hosts.
torizon@192.168.6.242’s password:

  • The terminal process “scp ‘-i’, ‘/home/steve/GimbalS/.conf/id_rsa’, ‘-o’, ‘UserKnownHostsFile=/dev/null’, ‘-o’, ‘StrictHostKeyChecking=no’, ‘-P’, ‘2230’, ‘-pr’, ‘/home/steve/GimbalS/aarch64/debug’, ‘torizon@192.168.6.242:~/app’” terminated with exit code: 1.
  • Terminal will be reused by tasks, press any key to close it.

I tried using the verdin password, but it just comes back and asks again. It is passing the id that was made, so I am not sure why this is asking for a password. I am using a pub/private key setup to log in normally, but I know that doesn’t work for scp. Can you tell me what is wrong?

Thanks,

Steve

2

Hi, I’m the Toradex AI Assistant.

I might make mistakes or occasionally get you wrong. Our support team will step in if needed.

I’d be glad to help with your SSH connection issue! To provide the most accurate guidance for your specific setup, could you let me know which OS and version you’re running on your device?

You can use the tdx-info command to get detailed information about your hardware and OS configuration. For more details, see: tdx-info documentation.

This will help me provide the most appropriate troubleshooting steps for your specific platform configuration.

Best regards, Toradex AI Assistant.

Did you like my answer? :+1: :-1:

If I answered your question, please tick the “Solution” checkbox.

For follow-up questions to me, please use @ToradexAI in your post.

3

tdx-info:

Software summary

Bootloader: U-Boot

Kernel version: 6.6.94-7.3.0-g5a62539de8f9 #1-Torizon SMP PREEMPT Wed Jun 25 09:14:55 UTC 2025
Kernel command line: root=LABEL=otaroot rootfstype=ext4 quiet logo.nologo vt.global_cursor_default=0 plymouth.ignore-serial-consoles splash fbcon=map:3 ostree=/ostree/boot.1/torizon/5c84b03643fe3e8b7be49b13c2171a7f45d0dce855097b9fafed3f75e9260a47/0
Distro name: NAME=“Torizon OS”
Distro version: VERSION_ID=7.3.0-build.18
Distro variant: VARIANT=“Docker”
Hostname: verdin-imx8mp-15132709

Hardware info

HW model: Toradex Verdin iMX8M Plus WB on Verdin Development Board

Toradex version: 0058 V1.1A
Serial number: 15132709
Processor arch: aarch64

4

SSH Key Authentication Issue

Thanks for providing the system information! You’re running Torizon OS 7.3.0 on a Verdin iMX8M Plus with the Development Board(1). The issue you’re experiencing with the SCP command asking for a password despite using SSH keys is likely due to incorrect key permissions or path issues.

Troubleshooting Steps

1. Check SSH Key Permissions

The SSH private key file needs specific permissions. Run these commands:

chmod 600 /home/steve/GimbalS/.conf/id_rsa
chmod 644 /home/steve/GimbalS/.conf/id_rsa.pub

2. Verify SSH Key Authentication

Test SSH connection first before using SCP:

ssh -i /home/steve/GimbalS/.conf/id_rsa -p 2230 torizon@192.168.6.242

If this prompts for a password, the key authentication isn’t working properly.

3. Check Public Key on Target Device

Ensure your public key is properly installed on the Torizon device:

ssh -p 2230 torizon@192.168.6.242 "cat ~/.ssh/authorized_keys"

The output should contain the content of your /home/steve/GimbalS/.conf/id_rsa.pub file.

4. Alternative SCP Approach

Try using the -v flag for verbose output to see what’s happening:

scp -v -i /home/steve/GimbalS/.conf/id_rsa -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no -P 2230 -pr /home/steve/GimbalS/aarch64/debug torizon@192.168.6.242:~/app

5. Regenerate SSH Keys if Needed

If the keys were corrupted during the VS Code extension update, you may need to regenerate them:

ssh-keygen -t rsa -b 4096 -f /home/steve/GimbalS/.conf/id_rsa

Then copy the new public key to your Torizon device:

ssh-copy-id -i /home/steve/GimbalS/.conf/id_rsa.pub -p 2230 torizon@192.168.6.242

The password prompt suggests the SSH key authentication is failing, which is why it’s falling back to password authentication.

Best regards, Toradex AI Assistant.

Did you like my answer? :+1: :-1:

If I answered your question, please tick the “Solution” checkbox.

For follow-up questions to me, please use @ToradexAI in your post.

5

Yes, I was able to fix this issue, not really sure what the issue was, but I think that the known hosts file got marked as read only or something like that, as it wouldn’t accept any changes for the new key.

But now, I noticed that the bookworm3.0.8 is downloading a ~6GB file where it used to be ~3.5GB and of course it links properly, but I get a logging exception with this version now, where I didn’t before.

6

Hey @EvetsMostel1,

Glad the issue was resolved.

-Eric

7

@eric.tx
Except it isn’t. Got another update on VSC, which caused an update locally. Got everthing back but now when I download the 3.08 Bookworm, I get the same issue as I was getting with the xonsh version, mainly, I can’t run my program. Boost causes an exception, when I try to log a message. So now I am stuck.
I am going to see if I can get it to compile boost as the project build. Hopefully, then it will match whatever is going on.

8

@eric.tx

I was able to put in an exception handler in boost that finally allowed me to find out what was going on. It turns out that I create a folder via the docker-compose.yml using the volumes directory to allow in app debugging logs. These logs are persistent and can be looked at at a later time. In the old setup, this was not an issue, but in this new one, the log directory is created and owned by root and when boost tries to write a log, it can’t because it doesn’t have permission. I don’t know why there wasn’t any output from boost at least to standard error that showed this, but at least know what is happening. If I change the owner to torizon, then it works fine.
I am not sure why it only shows up in the later version.
Is there a way to create the volume under torizon in the docker-compose.yml?

Steve

9

@eric.tx

In the old system, you had to be root if you were using the CAN bus. Is this still the case? Since, the old system ran as root, it didn’’t seem to cause an issue, but the new system looks like it is running as “torizon”.