Drive OS 6.0.6 Flash fails

Please provide the following info (tick the boxes after creating this topic):
Software Version
[*] DRIVE OS 6.0.6
DRIVE OS 6.0.5
DRIVE OS 6.0.4 (rev. 1)
DRIVE OS 6.0.4 SDK
other

Target Operating System
[*] Linux
QNX
other

Hardware Platform
DRIVE AGX Orin Developer Kit (940-63710-0010-300)
DRIVE AGX Orin Developer Kit (940-63710-0010-200)
DRIVE AGX Orin Developer Kit (940-63710-0010-100)
DRIVE AGX Orin Developer Kit (940-63710-0010-D00)
DRIVE AGX Orin Developer Kit (940-63710-0010-C00)
[*] DRIVE AGX Orin Developer Kit (not sure its number)
other

SDK Manager Version
1.9.2.10884
[*] other 1.9.3.10904

Host Machine Version
[*] native Ubuntu Linux 20.04 Host installed with SDK Manager
native Ubuntu Linux 20.04 Host installed with DRIVE OS Docker Containers
native Ubuntu Linux 18.04 Host installed with DRIVE OS Docker Containers
other

With SDK manager last step fails

With looking at the logs I have no idea what is the issue,

SDKM_logs_DRIVE_OS_6.0.6_SDK_Linux_for_DRIVE_AGX_Orin_DevKits_2023-07-13_15-59-29.zip (203.1 KB)

Have you ever successfully flashed any devkits with sdkmanager or docker methods?
If you haven’t tried the docker method, when we are checking the sdkmanager logs you provide, please refer to Install DRIVE OS Linux Docker Containers from NGC and try with the docker method.

I found the below error snippet from your logs. Please provide “/home/sushant/.nvsdkm/driveinstaller.log” and execute “ls -l /home/sushant/nvidia/nvidia_sdk/DRIVE_OS_6.0.6_SDK_Linux_DRIVE_AGX_ORIN_DEVKITS/DRIVEOS//drive-linux/filesystem/targetfs/usr/bin/” to check if it exists or not.

16:00:24.447 - info: NV_DRIVE_FLASH_DRIVE_COMP@DRIVE_AGX_ORIN_DEVKITS: Pre flash script found! Calling scripts/linux_pre_flash.sh
16:00:25.014 - info: NV_DRIVE_FLASH_DRIVE_COMP@DRIVE_AGX_ORIN_DEVKITS: b’NV_WORKSPACE:‘b’NV_TARGETUSER:‘b’NV_TARGETPWD:‘b’NV_EULA_ACCEPTED:‘b’[host] [sudo] password for sushant: ‘b’\r\n’b’\r\nEULA has been accepted!\r\n’b"touch: cannot touch ‘/home/sushant/nvidia/nvidia_sdk/DRIVE_OS_6.0.6_SDK_Linux_DRIVE_AGX_ORIN_DEVKITS/DRIVEOS//drive-linux/filesystem/targetfs/etc/nvidia/oem-config/driveos_eula_accepted’: No such file or directory\r\n"b’\r\nExecuting /opt/nvidia/driveos/common/filesystems/build-fs/17/bin/build_fs.py for user account setup.\r\n’b"2023-07-13 16:00:24,793 [INFO] Reading Configuration File: ‘/opt/nvidia/driveos/common/filesystems//build-fs/17.1.11-67787297/build-fs.config’\r\n"b’2023-07-13 16:00:24,794 [INFO] No Associated filesystem builds requested in STDIN\r\n’b’2023-07-13 16:00:24,794 [INFO] Executing mount \r\n’b’2023-07-13 16:00:24,796 [INFO] Executing grep “binfmt_misc on /proc/sys/fs/binfmt_misc”\r\n’b’2023-07-13 16:00:24,798 [INFO] Executing update-binfmts --enable qemu-aarch64 \r\n’b’update-binfmts: warning: qemu-aarch64 already enabled in kernel.\r\n’b’2023-07-13 16:00:24,799 [INFO] No debian packages requested in CONFIG. Skipping MANIFEST generation.\r\n2023-07-13 16:00:24,799 [INFO] No debian packages requested in MANIFEST. Skipping package install.\r\n’b’Traceback (most recent call last):\r\n File “/opt/nvidia/driveos/common/filesystems/build-fs/17/bin/build_fs.py”, line 3905, in \r\n’b’ build_fs.build()\r\n File “/opt/nvidia/driveos/common/filesystems/build-fs/17/bin/build_fs.py”, line 545, in build\r\n self.ug_manager.create_users()\r\n File “/opt/nvidia/driveos/common/filesystems/build-fs/17/bin/build_fs.py”, line 1284, in create_users\r\n’b’ command, args, user_passwd, username = self.parse_user(user)\r\n File “/opt/nvidia/driveos/common/filesystems/build-fs/17/bin/build_fs.py”, line 1192, in parse_user\r\n if uid and self.if_user_id_exists(uid):\r\n File “/opt/nvidia/driveos/common/filesystems/build-fs/17/bin/build_fs.py”, line 1064, in if_user_id_exists\r\n’b’ shell_stream = self.executor.execute_for_arm64(\r\n File “/opt/nvidia/driveos/common/filesystems/build-fs/17.1.11-67787297/bin/executor.py”, line 136, in execute_for_arm64\r\n self.setup_arm64_chroot()\r\n File “/opt/nvidia/driveos/common/filesystems/build-fs/17.1.11-67787297/bin/executor.py”, line 153, in setup_arm64_chroot\r\n shutil.copy2(QEMU_PATH+'/qemu-aarch64-static',\r\n File “/usr/lib/python3.8/shutil.py”, line 435, in copy2\r\n copyfile(src, dst, follow_symlinks=follow_symlinks)\r\n File “/usr/lib/python3.8/shutil.py”, line 264, in copyfile\r\n with open(src, 'rb') as fsrc, open(dst, 'wb') as fdst:\r\n’b"FileNotFoundError: [Errno 2] No such file or directory: ‘/home/sushant/nvidia/nvidia_sdk/DRIVE_OS_6.0.6_SDK_Linux_DRIVE_AGX_ORIN_DEVKITS/DRIVEOS//drive-linux/filesystem/targetfs/usr/bin/’\r\n2023-07-13 16:00:24,854 [INFO] \r\nExecuting Cleanup Routine for Linux Build-FS on Exit.\r\n\r\n"b’2023-07-13 16:00:24,856 [INFO] \r\nExecuting Cleanup Routine for Build-FS on Exit.\r\n\r\n’b’Script scripts/linux_pre_flash.sh fails.\r\nDRIVEOS filesystem Username/Password setup has not been completed.\r\n’Extracting the PDK failed!
16:00:25.018 - info: NV_DRIVE_FLASH_DRIVE_COMP@DRIVE_AGX_ORIN_DEVKITS: driveinstaller cannot continue! Check the log file for details: /home/sushant/.nvsdkm/driveinstaller.log

Yes I did flash SDK manager before for 6.0.5, I had to struggle with different errors in the Log & Ubuntu version back then but finally it did flash. Now i am trying to re-flash it for 6.0.6.
ls -l command suggested by does run with 1179 rows the folder does exist.
driveinstaller.log (1.6 MB)
please find attached log requested.
I was more worried about the double slash /DRIVEOS// in the path but I guess it is not the problem. It is not so clear this time the exact error from the log this time.

I’m currently investigating this issue with our team. In the meantime, I recommend trying the flashing process using the Docker method. This might help resolve the problem. Thank you for your cooperation.

For now, I did update to 6.0.6 with docker flash. But do let me know if you are able to find the issue. Thanks.

1 Like

We will continue to investigate the issue and provide updates. Thanks.

Any solution? I have the same problem and I don’t know how to use the docker version.

If I do

docker run -it --rm sdkmanager

it doesn’t open the GUI and says Please supply SDK Manager CLI argument. --help' will retrieve the options.

If I do

docker run -it --rm sdkmanager --help

I see that there is a flash option, but if I do

docker run -it --rm sdkmanager --flash

it says Ignoring --flash. This argument is supported only with --cli. /opt/nvidia/sdkmanager/sdkmanager-gui: error while loading shared libraries: libxshmfence.so.1: cannot open shared object file: No such file or directory and I can’t find much more documentation on the internet on how to use the docker version.

Please create a new topic and then link it to this one if you think they are related.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.