Err:1 http://security.ubuntu.com/ubuntu xenial-security/main amd64 libglib2.0-dev amd64 2.48.2-0ubuntu4.1
404 Not Found [IP: 141.30.62.25 80]
Sorry I have no idea how to fix this environment issue. It is required to fix the ‘sudo apt-get update’ issue on your host machine before running SDK Manager.
I also have trouble installing JetPack 4.2.2 using the sdkmanager. The error messages are copied below.
The Flash Jetson OS operation was successful, but error happens when CUDA was installing. I check with lsusb and I am able to see the TX2. I also ping both the host and the target from each other, and they are reachable. What is the problem that is preventing me from installing all the Jetson SDK components?
Our internal team reported back that the build operates as expected after installation and alternatives are updated too. The following errors can be safely ignored. The issue will be fixed in future release. Thanks.
“11:43:40 ERROR : NVIDIA Nsight Systems : alternative path is not absolute as it should be:”
Hi Edward,
I found the issue.
The problem was that the NVidia SDKManager seems to mess with apt under certain circumstances, as in that apt tries to look for updates on the ARM architecture as well and there is no such path in the predefined package path. I changed the /etc/apt/sources.list file by adding [arch=amd64,i386] to every path in that file and then it worked.
The solution is described in detail here: https://askubuntu.com/questions/996103/16-04-update-packages-not-found
Apparently I am not the only one who ran into this issue along with the Jetpack install so maybe its worth for the dev team to have a look into this issue.
This is an environment issue, which is different from the ARM architecture issue. Historically JetPack has the ARM architecture issue, but that issue has been fixed after CUDA 10.0.145 version. When the issue is caused by ARM architecture, it reports error like
Err:8 Index of /ubuntu xenial-security/main arm64 Packages
404 Not Found [IP: 91.189.88.162 80]
Notice that it specifies arm64 in above error.
However, the error on your machine is:
Err:1 Index of /ubuntu xenial-security/main amd64 libglib2.0-dev amd64 2.48.2-0ubuntu4.1
404 Not Found [IP: 141.30.62.25 80]
Notice the error mentions amd64 instead of arm64, so it’s a different one.
Yes it’s strange, but adding [arch=amd64,i386] to every path in /etc/apt/sources.list solved the problem for me.
My host PC is used for nothing else but installing Jetpack, and I never had any of such issues with JP3.3, 4.2, and 4.2.1 before.
Indeed, it’s been repeatedly stated in my case and apparently many others. Whatever special sauce went into packaging 4.2.2 probably broke it for some of us folks, since <= 4.2.1 seems to work. At the moment, I’ve given up yak shaving and just sticking with 4.2.1.
Hello, I am trying to install Jetpack 4.2.2 into a AGX, however the installation failed. The OS was successfully installed into the AGX but all the Jetson SDK components failed to install, except for the NVDIA Container Runtime. CUDA was the first to fail, everything else follows. I have the following error codes. What should I do to install all the Jetson SDK Components? I am able to ping each other, and the host and AGX are connected by USB only; the host has the IP of 192.168.55.10, and the target AGX has 192.168.55.1. Host PC is running Ubuntu 16.04. Thanks!
19:52:55 ERROR : CUDA Toolkit for L4T : command scp -q -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no /tmp/tmp_NV_L4T_CUDA_TARGET_POST_INSTALL_COMP.sh agxic1@192.168.55.1:~; ssh -t -q -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no agxic1@192.168.55.1 "~/tmp_NV_L4T_CUDA_TARGET_POST_INSTALL_COMP.sh && rm -f ~/tmp_NV_L4T_CUDA_TARGET_POST_INSTALL_COMP.sh" finished with error
19:52:55 ERROR : CUDA Toolkit for L4T : command terminated with error
FYI, after a flash, on more recent releases, you have to actually have a keyboard/monitor installed on the TX2, and then during the first boot, create the account the scp command uses. scp (and ssh) would fail like this if there is no first time login locally from the TX2 on its first boot. It is after that you can do the extra package installs. Did you do the first time login?
One hint that this is the case would be if the host PC can “ping 192.168.55.1”, but “ssh” to that address has no logins.
I did do the first time login, but your suspicion seems to be correct. I am able to ssh from host to AGX by: ssh agxic1@192.168.55.1 ,and I am able to do scp test if I can copy files to the AGX. However, CUDA installation still failed. The detailed error messages are attached below:
19:41:42 ERROR : command terminated with error
19:43:34 ERROR : CUDA Toolkit for L4T : command scp -q -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no /tmp/tmp_NV_L4T_CUDA_TARGET_POST_INSTALL_COMP.sh agxic1@192.168.55.1:~; ssh -t -q -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no agxic1@192.168.55.1 "~/tmp_NV_L4T_CUDA_TARGET_POST_INSTALL_COMP.sh && rm -f ~/tmp_NV_L4T_CUDA_TARGET_POST_INSTALL_COMP.sh" finished with error
19:43:34 ERROR : CUDA Toolkit for L4T : command terminated with error
19:43:34 ERROR : install 'CUDA Toolkit for L4T' failure, command < cd '/home/leer2-16/Downloads/AGX_Jetpack/Target_422/JetPack_4.2.2_Linux_GA_P2888' ; '/opt/nvidia/sdkmanager/resources/app/output/installUtils/adapter' -a='install' -c='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' -i='NV_L4T_CUDA_TARGET_POST_INSTALL_COMP' -v='10.0' -l='/home/leer2-16/Downloads/AGX_Jetpack/Target_422/JetPack_4.2.2_Linux_GA_P2888' -L='/home/leer2-16/.nvsdkm/logs/JetPack_4.2.2_Linux_GA' -p='2629' -t='P2888' -e='/home/leer2-16/Downloads/AGX_Jetpack/sdkM_download_422/cuda-repo-l4t-10-0-local-10.0.326_1.0-1_arm64.deb' > terminated with error
19:45:45 ERROR : TensorFlow : Run commands failed at step Install: command scp -q -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no /tmp/tmp_NV_TENSORFLOW_TARGET_POST_INSTALL_COMP.sh agxic1@192.168.55.1:~; ssh -t -q -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no agxic1@192.168.55.1 "~/tmp_NV_TENSORFLOW_TARGET_POST_INSTALL_COMP.sh && rm -f ~/tmp_NV_TENSORFLOW_TARGET_POST_INSTALL_COMP.sh" finished with error
19:45:45 ERROR : TensorFlow : command terminated with error
19:45:45 ERROR : install 'TensorFlow' failure, command < cd '/home/leer2-16/Downloads/AGX_Jetpack/Target_422/JetPack_4.2.2_Linux_GA_P2888' ; '/opt/nvidia/sdkmanager/resources/app/output/installUtils/adapter' -a='install' -c='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' -i='NV_TENSORFLOW_TARGET_POST_INSTALL_COMP' -v='1.14' -l='/home/leer2-16/Downloads/AGX_Jetpack/Target_422/JetPack_4.2.2_Linux_GA_P2888' -L='/home/leer2-16/.nvsdkm/logs/JetPack_4.2.2_Linux_GA' -p='2629' -t='P2888' -e='/home/leer2-16/Downloads/AGX_Jetpack/sdkM_download_422' > terminated with error
19:45:45 ERROR : Unavailable components detected, please check log
19:45:45 ERROR : cuDNN on Target Skipped. Due to its dependencies fail to install: CUDA Toolkit for L4T,
19:45:45 ERROR : TensorRT on Target Skipped. Due to its dependencies fail to install: CUDA Toolkit for L4T,
19:45:45 ERROR : OpenCV on Target Skipped. Due to its dependencies fail to install: CUDA Toolkit for L4T,
19:45:45 ERROR : VisionWorks on Target Skipped. Due to its dependencies fail to install: CUDA Toolkit for L4T,
19:45:45 ERROR : DeepStream Skipped. Due to its dependencies fail to install: CUDA Toolkit for L4T,
19:45:45 ERROR : Multimedia API Skipped. Due to its dependencies fail to install: CUDA Toolkit for L4T,
I couldn’t say what the specific failure point is, but since you know ssh is working, then it implies that either the host was unable to download a file from the internet for copy to the Jetson, or else there was some other package conflict. A number of “dependencies fail” lines occurred, but that could be either because a prerequisite couldn’t be downloaded, or it could have been due to other package conflicts.
Overall you at least know now that ssh works to that address. Continue to test if it works before each try, and if it does, then networking and ssh/scp are not the issue.
You might try this from the Jetson prior to trying to install packages again:
sudo apt update
sudo apt --fix-broken install
Depending on what occurs, someone with more knowledge of the content servers at NVIDIA can probably help. Just FYI, different regions of the world may use different servers, and as a result a specific region of the world may have an issue that other regions do not.
Hello all. I was wondering if someone would please help me know how to fix a problem. After hours of trying to setup my tx2. I keep getting this. Reading package lists… Done
Building dependency tree
Reading state information… Done
Note, selecting ‘sdkmanager’ instead of ‘./sdkmanager_0.9.14-4961_amd64.deb’
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:
The following packages have unmet dependencies:
sdkmanager : Depends: libgconf-2-4 but it is not installable
Depends: libcanberra-gtk-module but it is not installable
E: Unable to correct problems, you have held broken packages.
Note that the “.deb” file is a package for installing SDK Manager. The actual executable (the program itself) is the “sdkmanager” command (available only if you’ve installed SDKM, such as via the “.deb” package).
On the host side you might try this before continuing. This would probably help with dependencies:
Thank you very much for your time. I installed it. Now the problem is that i can not agree to the terms to go to step 3. The box is missing. looks like im missing some of the sdk.