{"@attributes":{"version":"2.0"},"channel":{"title":"Docker Pirates ARMed with explosive stuff","link":"https:\/\/blog.hypriot.com\/index.xml","description":"Recent content on Docker Pirates ARMed with explosive stuff","generator":"Hugo -- gohugo.io","language":"en-us","lastBuildDate":"Thu, 25 Jul 2019 06:31:00 +0200","item":[{"title":"Releasing HypriotOS 1.11.0: Docker 19.03.0 CE from Raspberry Pi Zero to 4 B","link":"https:\/\/blog.hypriot.com\/post\/releasing-HypriotOS-1-11\/","pubDate":"Thu, 25 Jul 2019 06:31:00 +0200","guid":"https:\/\/blog.hypriot.com\/post\/releasing-HypriotOS-1-11\/","description":"

We’re proud to announce our 1.11.0 release of HypriotOS - the fastest way to get Docker up and running on any Raspberry Pi.<\/strong><\/p>\n\n

\"Raspberry<\/p>\n\n

<\/p>\n\n

Features of HypriotOS<\/h2>\n\n

Latest Docker Engine 19.03.0 Community Edition<\/strong> <\/br>\nYou can use the latest features of the freshly-baked Docker Engine 19.03.0 that is still warm. It includes the Swarm Mode, which allows high availability of services in a multi-node cluster within just a few simple commands.<\/p>\n\n

Up to date with Raspbian Lite Buster<\/strong> <\/br>\nYou can run HypriotOS with an up-to-date OS and Linux kernel which is in sync with the current Raspbian Lite Buster running a Linux kernel 4.19.58.<\/p>\n\n

Support for the complete Raspberry Pi family<\/strong> <\/br>\nYou can run HypriotOS on every model of the Raspberry Pi family - we’re supporting Pi 1, 2, 3, the 3 B+, Zero and even the Compute Module and the new Raspberry Pi 4 B.<\/p>\n\n

Easy flashing and configuration<\/strong> <\/br>\nWe improved our flash tool<\/a>, which puts HypriotOS onto a SD card that is ready to boot from with a single command. With additional command line options you can customize HypriotOS during the flash operation to have the best out-of-the-box first-boot experience.\nHypriotOS includes cloud-init<\/a>, which makes the first boot of your Raspberry Pi customizable and you even are able connecting it to your Wi-Fi network during boot.\nAfter booting, you can find the Raspberry Pi at your network with a simple ping black-pearl.local<\/code> \u2013 no more searching for IP addresses required thanks to the integrated Avahi service discovery.<\/p>\n\n

Enhanced security out of the box<\/strong> <\/br>\nWe think that security should be shipped out-of-the-box. We make HypriotOS more secure without you even noticing it. For instance, there is no built-in “root” user. Also, the default user “pirate” (password “hypriot”) is can be customized or removed before the first boot. Just look at the file \/boot\/user-data<\/code>. You can add your public SSH key, disable password logins and specify a different user account before you even boot your Raspberry Pi. WiFi can be customized and enabled to have Docker up and running through the air without attaching a keyboard and monitor.<\/p>\n\n

Smaller than Raspbian Lite<\/strong> <\/br>\nEven though HypriotOS 1.11.0 is fully packed with the complete and latest Docker tool set, it now comes at a size smaller than the tiniest version of Raspbian (“Raspbian Lite”).<\/p>\n\n

Please see all details in the release notes<\/a>.<\/p>\n\n

Quick start<\/h2>\n\n

Download our flash tool<\/a><\/strong><\/p>\n\n

curl -O https:\/\/raw.githubusercontent.com\/hypriot\/flash\/2.3.0\/flash\nchmod +x flash\nsudo mv flash \/usr\/local\/bin\/flash\n<\/code><\/pre>\n\n
Now run this command to flash HypriotOS 1.11.0<\/strong><\/p>\n\n
flash https:\/\/github.com\/hypriot\/image-builder-rpi\/releases\/download\/v1.11.0\/hypriotos-rpi-v1.11.0.img.zip\n<\/code><\/pre>\n\n
Afterwards, put the SD card into the Raspberry Pi and power it. That’s all to get HypriotOS up and running!<\/strong><\/p>\n\n
Next steps<\/h3>\n\n
If you want to connect to the Raspberry Pi, run<\/p>\n\n
ssh pirate@black-pearl.local\n<\/code><\/pre>\n\n
with password “hypriot”.<\/p>\n\n
Flash with Wi-Fi settings for Pi Zero \/ Pi 3 \/ Pi 4<\/h3>\n\n
If you want the Raspberry Pi Zero (or Pi 3 or Pi 4) to connect directly to your Wi-Fi after boot, change the hostname of the Raspberry Pi and more, edit \/boot\/user-data<\/code> of the SD card and have a look at our FAQ<\/a>. Alternatively, checkout the parameters of the Hypriot flash tool<\/a> that also allows you to define your own cloud-init user-data template file which will be copied onto the SD image for you:<\/p>\n\n
flash -n myHOSTNAME -u wifi.yml https:\/\/github.com\/hypriot\/image-builder-rpi\/releases\/download\/v1.11.0\/hypriotos-rpi-v1.11.0.img.zip\n<\/code><\/pre>\n\n
Feedback, please<\/h2>\n\n
As always, use the comments below to give us feedback and share it on Twitter or Facebook.<\/p>\n\n
Please send us your feedback on our Gitter channel<\/a> or tweet your thoughts and ideas on this project at @HypriotTweets<\/a>.<\/p>"},{"title":"NVIDIA Jetson Nano - Docker optimized Linux Kernel","link":"https:\/\/blog.hypriot.com\/post\/nvidia-jetson-nano-build-kernel-docker-optimized\/","pubDate":"Sat, 04 May 2019 03:57:21 +0200","guid":"https:\/\/blog.hypriot.com\/post\/nvidia-jetson-nano-build-kernel-docker-optimized\/","description":"
Despite the fact that the NVIDIA Jetson Nano DevKit comes with Docker Engine preinstalled and you can run containers just out-of-the-box on this great AI and Robotics enabled board, there are still some important kernel settings missing to run Docker Swarm mode, Kubernetes or k3s correctly.<\/p>\n\n
\"jetson-nano-board-docker-whale.jpg\"<\/p>\n\n
So, let’s try to fix this…<\/p>\n\n
<\/p>\n\n
Analyzing the Linux Kernel<\/h3>\n\n
In my last blogpost Verify your Linux Kernel for Container Compatibility<\/a>, I already shared all the details how you can easily verify the Linux kernel for all Container related kernel settings. So, this first part of analyzing the capabilities of the stock Linux kernel 4.9.x provided by NVIDIA is already done and documented. And this was an easy task as well, so everyone who’s interested in these details can repeat the task at his\/her own device.<\/p>\n\n
Let’s recap what we did found. Especially there is one important setting which will be used for networking. This feature called “IPVLAN”, is required for Docker Swarm mode and it’s also used for networking in Kubernetes and k3s.<\/p>\n\n
Building your own Linux Kernel<\/h3>\n\n
Anyway, even when we’d like to include or change only a single kernel setting, we have to customize the kernel configuration and have to compile and build our own Linux kernel. This is typically a common task for a desktop Linux system, but can be pretty ugly and cumbersome if you have to build the kernel for an Embedded Device.<\/p>\n\n
When we look back to all the other NVIDIA Jetson boards, like the TK1, TX1 and TX2, this requires a second Linux machine, running Ubuntu 14.04 or 16.04 on an Intel CPU. Then setting up a complete build system for cross-compiling and all these stuff. Honestly, this is a well-known approach for an Embedded Developer, but the good thing now for the Jetson Nano DevKit this is not required any more.<\/p>\n\n
Here the good news: you can customize and build your own Linux kernel directly on the Jetson Nano DevKit! You only need an internet connection and some time to perform all steps on your own. BTW, and this is another chance to learn something new.<\/p>\n\n
Preparing the Build Environment<\/h3>\n\n
Before we’re able to compile the Linux kernel on the Jetson Nano, we have to make sure we do have all required build tools installed. Here is all it takes, with a fast internet connection this is done within a few minutes only.<\/p>\n\n
$ sudo apt-get update\n$ sudo apt-get install -y libncurses5-dev\n<\/code><\/pre>\n\n
Download Linux Kernel Sources for Jetson Nano<\/h3>\n\n
Next, we’ll need to find and download the sources for the Linux kernel for the Jetson Nano DevKit directly from the NVIDIA website. The current version as writing this blogpost is NVIDIA Linux4Tegra Release r32.1 or short L4T 32.1. Just follow this link https:\/\/developer.nvidia.com\/embedded\/linux-tegra<\/a> and select at topic “32.1 Driver Details” the referenced download link for “Jetson Nano”, “SOURCES” and “BSP Sources”.<\/p>\n\n
We can also directly download this package to the Jetson Nano. But please be aware that this download link can change over time, so verify it carefully.<\/p>\n\n
$ cd\n$ mkdir -p nano-bsp-sources\n$ cd nano-bsp-sources\n$ wget https:\/\/developer.download.nvidia.com\/embedded\/L4T\/r32_Release_v1.0\/jetson-nano\/BSP\/Jetson-Nano-public_sources.tbz2\n$ ls -alh Jetson-Nano-public_sources.tbz2\n-rw-rw-r-- 1 pirate pirate 133M Mar 16 06:46 Jetson-Nano-public_sources.tbz2\n<\/code><\/pre>\n\n
Now extract the kernel source package “kernel_src.tbz2” from the downloaded file.<\/p>\n\n
$ tar xvf Jetson-Nano-public_sources.tbz2 public_sources\/kernel_src.tbz2\n$ mv public_sources\/kernel_src.tbz2 ~\/\n$ cd\n$ ls -alh ~\/kernel_src.tbz2\n-rw-r--r-- 1 pirate pirate 117M Mar 13 08:45 \/home\/pirate\/kernel_src.tbz2\n<\/code><\/pre>\n\n
You may now free some disk space and remove all the downloads, as we don’t need it any more.<\/p>\n\n
$ rm -fr ~\/nano-bsp-sources\/\n<\/code><\/pre>\n\n
Last step, please extract the kernel source tree.<\/p>\n\n
$ cd\n$ tar xvf .\/kernel_src.tbz2\n<\/code><\/pre>\n\n
Compile the default Linux Kernel<\/h3>\n\n
Cool, we have now all the Linux kernel source tree for the Jetson Nano DevKit downloaded and extracted.<\/p>\n\n
As the next step, I’d recommend to first compile the default unmodified kernel in order to verify that we do have all the build dependencies installed and this way, we’ll also get familiar with the kernel compiling.<\/p>\n\n
Before we can start the compile job, we have to make sure to use the correct kernel configuration file. This file “.config” is missing in the provided kernel source tree, but we can directly get it from our running Linux kernel on the Jetson Nano. This .config file can be found as kernel file at “\/proc\/config.gz” in a compressed form.<\/p>\n\n
$ cd ~\/kernel\/kernel-4.9\n$ zcat \/proc\/config.gz > .config\n<\/code><\/pre>\n\n
Now, let’s verify the content of the Linux kernel .config file.<\/p>\n\n
pirate@jetson-nano:~\/kernel\/kernel-4.9$ head -10 .config\n#\n# Automatically generated file; DO NOT EDIT.\n# Linux\/arm64 4.9.140 Kernel Configuration\n#\nCONFIG_ARM64=y\nCONFIG_64BIT=y\nCONFIG_ARCH_PHYS_ADDR_T_64BIT=y\nCONFIG_MMU=y\nCONFIG_DEBUG_RODATA=y\nCONFIG_ARM64_PAGE_SHIFT=12\n...\n<\/code><\/pre>\n\n
As you can see, it’s a Kernel Configuration for Linux kernel version 4.9.140 and for ARM 64-bit architecture.<\/p>\n\n
Start the kernel compile job. As we do have 4x cores available on the Nano, we’d like to keep the CPU busy and using 5x parallel compile tasks.<\/p>\n\n
$ make prepare\n$ make modules_prepare\n\n# Use 5x parallel compile tasks\n# Compile kernel as an image file\n$ time make -j5 Image\n...\nreal\t28m13,235s\nuser\t91m48,700s\nsys\t7m46,240s\n\n# List newly compiled kernel image\n$ ls -alh arch\/arm64\/boot\/Image\n-rw-rw-r-- 1 pirate pirate 33M May  4 00:14 arch\/arm64\/boot\/Image\n\n# Compile all kernel modules\n$ time make -j5 modules\n...\nreal\t29m15,621s\nuser\t92m41,176s\nsys\t8m18,404s\n<\/code><\/pre>\n\n
The Nano CPU’s are pretty busy while compiling the kernel and kernel modules.<\/p>\n\n
\"jetson-nano-board-compile-kernel.jpg\"<\/p>\n\n
The build\/compile job will take around 60 minutes in total, but the good thing is, all happens directly on your Jetson Nano DevKit. No other expensive equipment is required at all, just an internet connection and some of your time.<\/p>\n\n
Install our newly built Linux Kernel and Modules<\/h3>\n\n
After these pretty long compile jobs for generating our own Linux kernel and kernel modules, we are ready to install the kernel and verify if it’s able to boot correctly. Therefore we should make a backup of the old kernel first, then install the new kernel and also install all newly built kernel modules.<\/p>\n\n
Before we install the new kernel and boot the Jetson Nano, let’s check the default Linux kernel version. Then we can compare it later to our own kernel.<\/p>\n\n
pirate@jetson-nano:~$ uname -a\nLinux jetson-nano 4.9.140-tegra #1 SMP PREEMPT Wed Mar 13 00:32:22 PDT 2019 aarch64 aarch64 aarch64 GNU\/Linux\n<\/code><\/pre>\n\n
Here is also a ASCIINEMA recording of a check-config.sh<\/code> verification done with the default kernel.\n\"asciicast\"<\/a><\/p>\n\n
As we can see, we do have a Linux kernel version “4.9.140-tegra”. This one was compiled at “Wed Mar 13 00:32:22 PDT 2019” and it’s the default kernel provided by NVIDIA for the Jetson Nano.<\/p>\n\n
Now, install our new kernel and kernel modules.<\/p>\n\n
# Backup the old kernel image file\n$ sudo cp \/boot\/Image \/boot\/Image.original\n\n# Install modules and kernel image\n$ cd ~\/kernel\/kernel-4.9\n$ sudo make modules_install\n$ sudo cp arch\/arm64\/boot\/Image \/boot\/Image\n\n# Verify the kernel images\npirate@jetson-nano:~\/kernel\/kernel-4.9$ ls -alh \/boot\/Image*\n-rw-r--r-- 1 root root 33M May  4 00:55 \/boot\/Image\n-rw-r--r-- 1 root root 33M May  4 00:49 \/boot\/Image.original\n<\/code><\/pre>\n\n
Now, reboot the Nano and check the kernel again.<\/p>\n\n
pirate@jetson-nano:~$ uname -a\nLinux jetson-nano 4.9.140 #1 SMP PREEMPT Sat May 4 00:12:56 CEST 2019 aarch64 aarch64 aarch64 GNU\/Linux\n<\/code><\/pre>\n\n
As you can see, our newly compiled kernel is working. The kernel version has changed to “4.9.140”, note the missing trailing “-tegra” which indicates this build is a custom build. And the compile date\/time has also changed to “Sat May 4 00:12:56 CEST 2019”.<\/p>\n\n
Hint:<\/strong> Please remember, every time you do change a kernel setting and compile a new kernel, you have to install the kernel image file AND the kernel modules.<\/p>\n\n
Customizing the Linux Kernel Configuration<\/h3>\n\n
When it comes to the point to modify or customize the Linux kernel configuration, then this can get pretty complicated when you don’t know where to start. First of all, it’s a very bad idea to edit the .config file directly with an editor. Please, NEVER DO THIS - seriously!<\/p>\n\n
The correct way to customize the kernel .config file is, to use the right tooling. One tool which is already built-in and available even in your bash shell (works also via ssh), is the tool menuconfig<\/code>. Therefore we already installed the build dependency “libncurses5-dev” at the beginning.<\/p>\n\n
I don’t want to go into all details on how to use menuconfig<\/code>, therefore here are the basic commands to start it and then I did recorded an ASCIINEMA to change the setting for “IPVLAN”. I think then you’ll should get a good idea how this works.<\/p>\n\n
# Backup the kernel config\n$ cd ~\/kernel\/kernel-4.9\n$ cp .config kernel.config.original\n\n$ make menuconfig\n<\/code><\/pre>\n\n
ASCIINEMA recording on how to include the “IPVLAN” kernel setting.\n\"asciicast\"<\/a><\/p>\n\n
Finally let’s re-compile the kernel and the kernel modules and install them, like we did before.<\/p>\n\n
$ cd ~\/kernel\/kernel-4.9\n\n# Prepare the kernel build\n$ make prepare\n$ make modules_prepare\n\n# Compile kernel image and kernel modules\n$ time make -j5 Image\n$ time make -j5 modules\n\n# Install modules and kernel image\n$ sudo make modules_install\n$ sudo cp arch\/arm64\/boot\/Image \/boot\/Image\n<\/code><\/pre>\n\n
Reboot the Nano and check the kernel again.<\/p>\n\n
Fast Forward - Fully Container Optimized Kernel Configuration<\/h3>\n\n
As you have learned here in this tutorial, you’re now able to apply more and more settings to your kernel configuration. But this will take some time for sure.<\/p>\n\n
In order to save you a lot of time and efforts, I’ve already optimized the Linux kernel in all details. Here you can find a public Gist at Github<\/a> with my resulting kernel .config. You can download it directly to your Nano and compile your own Linux kernel with this configuration.<\/p>\n\n
# Download the fully container optimized kernel configuration file\n$ cd ~\/kernel\/kernel-4.9\n$ wget https:\/\/gist.githubusercontent.com\/DieterReuter\/a7d07445c9d62b45d9151c22b446c59b\/raw\/6decc91cc764ec0be8582186a34f60ea83fa89db\/kernel.config.fully-container-optimized \n$ cp kernel.config.fully-container-optimized .config\n\n# Prepare the kernel build\n$ make prepare\n$ make modules_prepare\n\n# Compile kernel image and kernel modules\n$ time make -j5 Image\n$ time make -j5 modules\n\n# Install modules and kernel image\n$ sudo make modules_install\n$ sudo cp arch\/arm64\/boot\/Image \/boot\/Image\n<\/code><\/pre>\n\n
Now, reboot the Nano and check the kernel again.<\/p>\n\n
pirate@jetson-nano:~$ uname -a\nLinux jetson-nano 4.9.140 #2 SMP PREEMPT Sat May 4 02:17:23 CEST 2019 aarch64 aarch64 aarch64 GNU\/Linux\n\npirate@jetson-nano:~$ ls -al \/boot\/Image*\n-rw-r--r-- 1 root root 34381832 May  4 03:13 \/boot\/Image\n-rw-r--r-- 1 root root 34048008 May  4 00:49 \/boot\/Image.original\n<\/code><\/pre>\n\n
ASCIINEMA recording of the final run of check-config.sh<\/code> with the fully optimized kernel for running Containers on the Jetson Nano DevKit.\n\"asciicast\"<\/a><\/p>\n\n
Result: An almost perfect Linux kernel to run Containers on the NVIDIA Jetson Nano!<\/strong><\/p>\n\n
\"jetson-nano-board-docker-running2.jpg\"<\/p>\n\n
Conclusion<\/h3>\n\n
As you could learn with this short, but highly technical tutorial, you’re able to compile your own customized Linux kernel directly on the Jetson Nano DevKit without the need of an additional and maybe expensiv development machine. All can be done within an hour or two, and you have now the ability to change kernel settings whenever you want to. Just customize the kernel .config file, compile a new kernel and kernel modules and install it on your Nano.<\/p>\n\n
This way you’re now able to optimize the kernel for all your needs. For running Containers on the Nano with the help of Docker, Kubernetes or k3s, you’re now well prepared and know how to do this by yourself.<\/p>\n\n
Once the Linux kernel is fully optimized with all important Container related kernel settings, you can run Docker Swarm mode, Kubernetes and k3s with all features on that great ARM board from NVIDIA.<\/p>\n\n
Finally, May the 4th be with You!\n\"may-the-4th-be-with-you.jpg\"<\/p>\n\n
Feedback, please<\/h3>\n\n
As always use the comments below to give us feedback and share it on Twitter or Facebook.<\/p>\n\n
Please send us your feedback on our Gitter channel<\/a> or tweet your thoughts and ideas on this project at @HypriotTweets<\/a>.<\/p>\n\n
Dieter @Quintus23M<\/a><\/p>"},{"title":"Verify your Linux Kernel for Container Compatibility","link":"https:\/\/blog.hypriot.com\/post\/verify-kernel-container-compatibility\/","pubDate":"Sun, 28 Apr 2019 08:48:50 -0700","guid":"https:\/\/blog.hypriot.com\/post\/verify-kernel-container-compatibility\/","description":"
Are you sure whether your Linux kernel is able to run Containers in an optimal way, or if there are still some missing kernel settings which will lead to some strange issues in the future?<\/p>\n\n
\"400px-NewTux.svg.png\"<\/p>\n\n
Normally you don’t have to bother about this question. When you’re using Docker and Containers on a modern Linux system or on a public cloud offering, this has been already optimized by the Linux distribution or your cloud provider. But when you start using Containers on Embedded Devices you should verify this carefully.<\/p>\n\n
So, let’s learn how easy it is to verify it by yourself…<\/p>\n\n
<\/p>\n\n
How can I verify the Linux Kernel for Container Compatibility?<\/h3>\n\n
Typically this is really an easy task, as soon as you know the right tools.<\/p>\n\n
For running Containers you’ll need some basic settings applied to your Linux kernel. Some settings are mandatory and some others are optional and only used for specific use cases. But let’s see how we can use the right tools.<\/p>\n\n
At the Docker open source project you can find a great bash script which does all these tests on your Linux kernel configuration and tells you within a few seconds all the required details. The script is able to read the kernel config live from a running kernel or directly from a kernel .config file as well. Now you can imagine you can verify the container compatibility also from a remote device.<\/p>\n\n
Download check-config.sh<\/code> script<\/h4>\n\n
Let’s download the bash script check-config.sh<\/a> directly from the Moby project (yes, this is the new name for the Docker open source project).<\/p>\n\n
$ wget https:\/\/github.com\/moby\/moby\/raw\/master\/contrib\/check-config.sh\n$ chmod +x check-config.sh\n<\/code><\/pre>\n\n
Verify the Linux Kernel directly<\/h4>\n\n
If you have your Linux system available you can download and run the script directly without any parameters.<\/p>\n\n
$ .\/check-config.sh\n<\/code><\/pre>\n\n
Then you’ll get a detailled output with all kernel settings which are important for running containers.<\/p>\n\n
If you want to verify a kernel from a remote system, you could also first extract the Linux kernel config on this system and analyse it later.<\/p>\n\n
# extract the .config from a running kernel\n$ zcat \/proc\/config.gz > kernel.config\n\n$ ls -al kernel.config\n-rw-rw-r-- 1 pirate pirate 165739 Apr 28 07:26 kernel.config\n<\/code><\/pre>\n\n
Hint:<\/strong> On some Linux systems like Raspbian for the Raspberry Pi, the kernel .config is only available as a kernel module. Then you have to load the module first, using the command sudo modprobe configs<\/code>.<\/p>\n\n
Verify the Linux Kernel from a .config file<\/h4>\n\n
The kernel .config is a readable configuration file which is used to compile a new Linux kernel. Typically it will get embedded into your new kernel and therefore you can read it from the running kernel. It’s available as a file at \/proc\/config.gz<\/code> in a compressed form, so we have to use zcat<\/code> to extract the .config file in clear text.<\/p>\n\n
$ zcat \/proc\/config.gz | head -10\n#\n# Automatically generated file; DO NOT EDIT.\n# Linux\/arm64 4.9.140 Kernel Configuration\n#\nCONFIG_ARM64=y\nCONFIG_64BIT=y\nCONFIG_ARCH_PHYS_ADDR_T_64BIT=y\nCONFIG_MMU=y\nCONFIG_DEBUG_RODATA=y\nCONFIG_ARM64_PAGE_SHIFT=12\n...\n<\/code><\/pre>\n\n
Next, let’s run the check-config.sh<\/code> script again and read all kernel configs from the file.<\/p>\n\n
$ .\/check-config.sh kernel.config\n<\/code><\/pre>\n\n
Verify the Linux Kernel on NVIDIA Jetson Nano DevKit<\/h3>\n\n
As a real life example let’s now verify the Linux kernel of the brand-new Jetson Nano DevKit from NVIDIA. I already wrote a blogpost about how to install Linux for the Nano board, see here NVIDIA Jetson Nano Developer Kit - Introduction<\/a>.<\/p>\n\n
First we’ll check the Linux kernel version and we can see, it’s a current LTS kernel 4.9.<\/p>\n\n
pirate@jetson-nano:~$ uname -a\nLinux jetson-nano 4.9.140-tegra #1 SMP PREEMPT Wed Mar 13 00:32:22 PDT 2019 aarch64 aarch64 aarch64 GNU\/Linux\n<\/code><\/pre>\n\n
Now, let’s run the check-config.sh<\/code> script on the Nano and determine all the Container related kernel settings. We’ll get the complete output as colored text. From the screenshots here we can clearly see which of the required and optional kernel settings are already applied for the Nano’s Linux kernel.<\/p>\n\n
In the first section “Generally Necessary” all the mandatory kernel settings are listed, and for the Nano this is completely green, all is perfect.<\/p>\n\n
\"kernel-checkconfig-nano1.jpg\"<\/p>\n\n
Then in the second section “Optional Features” we can see that most Container related settings are applied, but a few are missing.<\/p>\n\n
Not all of these are really important to have, but when we look into the “Network Drivers” I would recommend to include all in the kernel to avoid issues. For example, if you want to use Docker Swarm mode you have to know that CONFIG_IPVLAN<\/code> is mandatory - this kernel can’t run Swarm mode correctly!<\/p>\n\n
For “Storage Drivers” you can typically ignore the missing settings for aufs<\/code> and zfs<\/code> as long as you don’t required to use these, same is true for devicemapper<\/code>.<\/p>\n\n
\"kernel-checkconfig-nano2.jpg\"<\/p>\n\n
Here I’d also like to present the output as pure ASCII text so you can easily analyse (search, copy&paste) it later.<\/p>\n\n
pirate@jetson-nano:~$ .\/check-config.sh\ninfo: reading kernel config from \/proc\/config.gz ...\n\nGenerally Necessary:\n- cgroup hierarchy: properly mounted [\/sys\/fs\/cgroup]\n- CONFIG_NAMESPACES: enabled\n- CONFIG_NET_NS: enabled\n- CONFIG_PID_NS: enabled\n- CONFIG_IPC_NS: enabled\n- CONFIG_UTS_NS: enabled\n- CONFIG_CGROUPS: enabled\n- CONFIG_CGROUP_CPUACCT: enabled\n- CONFIG_CGROUP_DEVICE: enabled\n- CONFIG_CGROUP_FREEZER: enabled\n- CONFIG_CGROUP_SCHED: enabled\n- CONFIG_CPUSETS: enabled\n- CONFIG_MEMCG: enabled\n- CONFIG_KEYS: enabled\n- CONFIG_VETH: enabled (as module)\n- CONFIG_BRIDGE: enabled\n- CONFIG_BRIDGE_NETFILTER: enabled (as module)\n- CONFIG_NF_NAT_IPV4: enabled (as module)\n- CONFIG_IP_NF_FILTER: enabled (as module)\n- CONFIG_IP_NF_TARGET_MASQUERADE: enabled (as module)\n- CONFIG_NETFILTER_XT_MATCH_ADDRTYPE: enabled (as module)\n- CONFIG_NETFILTER_XT_MATCH_CONNTRACK: enabled (as module)\n- CONFIG_NETFILTER_XT_MATCH_IPVS: enabled (as module)\n- CONFIG_IP_NF_NAT: enabled (as module)\n- CONFIG_NF_NAT: enabled (as module)\n- CONFIG_NF_NAT_NEEDED: enabled\n- CONFIG_POSIX_MQUEUE: enabled\n\nOptional Features:\n- CONFIG_USER_NS: enabled\n- CONFIG_SECCOMP: enabled\n- CONFIG_CGROUP_PIDS: enabled\n- CONFIG_MEMCG_SWAP: enabled\n- CONFIG_MEMCG_SWAP_ENABLED: enabled\n    (cgroup swap accounting is currently enabled)\n- CONFIG_BLK_CGROUP: enabled\n- CONFIG_BLK_DEV_THROTTLING: enabled\n- CONFIG_IOSCHED_CFQ: enabled\n- CONFIG_CFQ_GROUP_IOSCHED: missing\n- CONFIG_CGROUP_PERF: enabled\n- CONFIG_CGROUP_HUGETLB: missing\n- CONFIG_NET_CLS_CGROUP: enabled\n- CONFIG_CGROUP_NET_PRIO: enabled\n- CONFIG_CFS_BANDWIDTH: enabled\n- CONFIG_FAIR_GROUP_SCHED: enabled\n- CONFIG_RT_GROUP_SCHED: enabled\n- CONFIG_IP_NF_TARGET_REDIRECT: missing\n- CONFIG_IP_VS: enabled (as module)\n- CONFIG_IP_VS_NFCT: enabled\n- CONFIG_IP_VS_PROTO_TCP: missing\n- CONFIG_IP_VS_PROTO_UDP: missing\n- CONFIG_IP_VS_RR: enabled (as module)\n- CONFIG_EXT4_FS: enabled\n- CONFIG_EXT4_FS_POSIX_ACL: enabled\n- CONFIG_EXT4_FS_SECURITY: enabled\n- Network Drivers:\n  - "overlay":\n    - CONFIG_VXLAN: enabled\n      Optional (for encrypted networks):\n      - CONFIG_CRYPTO: enabled\n      - CONFIG_CRYPTO_AEAD: enabled\n      - CONFIG_CRYPTO_GCM: enabled\n      - CONFIG_CRYPTO_SEQIV: enabled\n      - CONFIG_CRYPTO_GHASH: enabled\n      - CONFIG_XFRM: enabled\n      - CONFIG_XFRM_USER: enabled\n      - CONFIG_XFRM_ALGO: enabled\n      - CONFIG_INET_ESP: missing\n      - CONFIG_INET_XFRM_MODE_TRANSPORT: enabled\n  - "ipvlan":\n    - CONFIG_IPVLAN: missing\n  - "macvlan":\n    - CONFIG_MACVLAN: enabled (as module)\n    - CONFIG_DUMMY: enabled\n  - "ftp,tftp client in container":\n    - CONFIG_NF_NAT_FTP: enabled (as module)\n    - CONFIG_NF_CONNTRACK_FTP: enabled (as module)\n    - CONFIG_NF_NAT_TFTP: enabled (as module)\n    - CONFIG_NF_CONNTRACK_TFTP: enabled (as module)\n- Storage Drivers:\n  - "aufs":\n    - CONFIG_AUFS_FS: missing\n  - "btrfs":\n    - CONFIG_BTRFS_FS: enabled (as module)\n    - CONFIG_BTRFS_FS_POSIX_ACL: enabled\n  - "devicemapper":\n    - CONFIG_BLK_DEV_DM: enabled\n    - CONFIG_DM_THIN_PROVISIONING: missing\n  - "overlay":\n    - CONFIG_OVERLAY_FS: enabled (as module)\n  - "zfs":\n    - \/dev\/zfs: missing\n    - zfs command: missing\n    - zpool command: missing\n\nLimits:\n- \/proc\/sys\/kernel\/keys\/root_maxkeys: 1000000\n<\/code><\/pre>\n\n
Conclusion<\/h3>\n\n
With these easy steps we’ve covered in this short blogpost, you’re now able to verify if the Linux kernel you’re using is able to run Docker, containerd, Kubernetes or k3s in an optimal way.<\/p>\n\n
Just keep this in mind whenever you’re discovering some strange errors with your Container Run-Time on a new Linux system. This is especially important when you run Containers on an Embedded Device. We’ve discovered a lot of missing kernel settings in the early days with the Raspberry Pi. Even today the Raspberry Pi kernel which comes with the default Raspbian is not fully optimized to run Containers, therefore the image built from HypriotOS is still a better alternative when you wish to run Containers on these devices.<\/p>\n\n
And even when Docker runs out-of-the-box on a brandnew device like it does on the Jetson Nano, it’s always a good idea to verify the Linux kernel - prior to get into some strange errors.<\/p>\n\n
In a later blogpost we’ll optimize the stock Linux kernel for the brand-new NVIDIA Jetson Nano DevKit and we’ll show you how to build your own customized kernel for this great board.<\/p>\n\n
Feedback, please<\/h3>\n\n
As always use the comments below to give us feedback and share it on Twitter or Facebook.<\/p>\n\n
Please send us your feedback on our Gitter channel<\/a> or tweet your thoughts and ideas on this project at @HypriotTweets<\/a>.<\/p>\n\n
Dieter @Quintus23M<\/a><\/p>"},{"title":"Docker Engine on Intel Linux runs Arm Containers","link":"https:\/\/blog.hypriot.com\/post\/docker-intel-runs-arm-containers\/","pubDate":"Sat, 27 Apr 2019 10:48:50 -0700","guid":"https:\/\/blog.hypriot.com\/post\/docker-intel-runs-arm-containers\/","description":"
Did you read the latest news from Docker about their newly announced technology partnership together with Arm, “Docker and Arm Partner to Deliver Frictionless Cloud-Native Software Development and Delivery Model for Cloud, Edge, and IoT”<\/a>?<\/p>\n\n
\"arm-docker-logo.jpg\"<\/p>\n\n
This is really a great ground-breaking news, as it will enable an improved development workflow. Build and test all your Arm containers on your Intel-based laptop or workstation. These new Arm capabilities will be available in Docker Desktop<\/a> products from Docker, both for MacOs and Windows, and for Docker\u2019s commercial enterprise offerings. First technical details and roadmap will be announced next week at DockerCon 2019<\/a> in San Francisco, so please stay tuned.<\/p>\n\n
But wait, what about all the users who are directly working on a pure Linux environment? Well, here is the good news, the basic technology you need is already available and ready-to-use for you.<\/p>\n\n
Yes, you could use it right away! Let’s start it now…<\/p>\n\n
<\/p>\n\n
Run an Arm Container with Docker Engine on Intel<\/h3>\n\n
I don’t want to hold you back and bore you with a lot of background details, so I’m going to show you how easy it is today.<\/p>\n\n
First, let’s start with a default Docker Engine installed on an Intel-based Linux system. For the sake of simplicity I’d like to show it directly on the Katacoda Training Platform<\/a>, then you are able to replay it without spending too much time. It just takes you a few seconds to spin up a complete Linux Docker environment, just a few click away.<\/p>\n\n
Step 1:<\/strong> Start the Docker Tutorial<\/a><\/p>\n\n
Next you have to click on “START SCENARIO” and the Linux Docker environment is ready for you.<\/p>\n\n
\"katacoda-docker-tutorial-start.jpg\"<\/p>\n\n
Step 2:<\/strong> Run an Arm-based Container<\/p>\n\n
In the bottom right box you’ll find a Linux Terminal window where you can issue your CLI commands. It’s a real Linux bash shell you can control through your browser and Docker is already installed on this machine.<\/p>\n\n
As soon as we try to start an Arm-based Docker container,<\/p>\n\n
$ docker run -it dieterreuter\/alpine-arm64:3.9\n<\/code><\/pre>\n\n
we’ll get an cryptic error message exec user process caused "exec format error"<\/code>, which basically tells us that this Docker Image tries to start a binary\/executable which can’t be run on the provided Intel processor.<\/p>\n\n
\"katacoda-docker-tutorial-runarm1.jpg\"<\/p>\n\n
Step 3:<\/strong> Run the magic command to enable Arm\/Arm64 on Intel<\/p>\n\n
$ docker run --rm --privileged hypriot\/qemu-register\n<\/code><\/pre>\n\n
This registers a few Qemu emulators inside of our Linux kernel with the help of the binfmt<\/code> tool. This instructs the Linux loader to start the specific Qemu emulator program to run the binary\/executable if it’s not based on Intel. Here we register \/qemu-arm<\/code> for Arm 32-bit and \/qemu-aarch64<\/code> for Arm 64-bit.<\/p>\n\n
Just to be precise, these emulators will be registered through a privileged Docker container. This is possible for all Linux kernel versions 4.9 and later. The emulators will be uploaded into memory, registered in the kernel and stay there persistent until you reboot your machine. This means you don’t have to change anything inside of your Docker Images, all magic will be done by the Linux kernel on the host system!<\/p>\n\n
$ docker run --rm --privileged hypriot\/qemu-register\nUnable to find image 'hypriot\/qemu-register:latest' locally\nlatest: Pulling from hypriot\/qemu-register\nfc1a6b909f82: Pull complete\n247c87d40120: Pull complete\n1e300bd4bcdc: Pull complete\n79c54222eda0: Pull complete\n7d0efdace32f: Pull complete\nDigest: sha256:17931ba1f5362c6fbf7f364b32bec7e06e0c376571a9e3b2849dea18ce887c91\nStatus: Downloaded newer image for hypriot\/qemu-register:latest\n---\nInstalled interpreter binaries:\n-rwxr-xr-x    3 root     root       6192520 Apr 27 17:17 \/qemu-aarch64\n-rwxr-xr-x    4 root     root       5606984 Apr 27 17:17 \/qemu-arm\n-rwxr-xr-x    2 root     root       5987464 Apr 27 17:17 \/qemu-ppc64le\n---\nRegistered interpreter=qemu-aarch64\nenabled\ninterpreter \/qemu-aarch64\nflags: OCF\noffset 0\nmagic 7f454c460201010000000000000000000200b700\nmask ffffffffffffff00fffffffffffffffffeffffff\n---\nRegistered interpreter=qemu-arm\nenabled\ninterpreter \/qemu-arm\nflags: OCF\noffset 0\nmagic 7f454c4601010100000000000000000002002800\nmask ffffffffffffff00fffffffffffffffffeffffff\n---\nRegistered interpreter=qemu-ppc64le\nenabled\ninterpreter \/qemu-ppc64le\nflags: OCF\noffset 0\nmagic 7f454c4602010100000000000000000002001500\nmask ffffffffffffff00fffffffffffffffffeffff00\n---\n$\n<\/code><\/pre>\n\n
Step 4:<\/strong> Run an Arm-based Container successfully on Intel<\/p>\n\n
Now let’s start the same Arm-based Docker container again, but this time it actually works successfully.<\/p>\n\n
$ docker run -it dieterreuter\/alpine-arm64:3.9\n<\/code><\/pre>\n\n
\"katacoda-docker-tutorial-runarm2.jpg\"<\/p>\n\n
When we run the command uname -a<\/code> it tell us, we’re running a Linux kernel 4.14.29 on Arm 64-bit architecture indicated as aarch64<\/code>.<\/p>\n\n
SUCCESS, our Arm 64-bit Docker Container even works on Intel CPU’s !!!<\/p>\n\n
References:<\/strong> The source code of this magic registration Docker Image is fully open source and can be found at https:\/\/github.com\/hypriot\/qemu-register<\/a>. I also updated it today to the latest available Qemu 4.0.0 release.<\/p>\n\n
Conclusion<\/h3>\n\n
As you could see, it’s damn easy to configure your Intel-based Linux Docker Engine to run 32 or 64-bit Arm Containers in an emulation mode seemlessly. As long as you run a recent Linux kernel 4.9 or later it just works.<\/p>\n\n
Basically this is the same emulation technology how Docker Desktop<\/a> is doing this behind the scenes on MacOS and Windows. BTW, this binfmt feature is already built-in in Docker Desktop since about April 2017.<\/p>\n\n
In the end it’s possible for a user to develop, build and test his\/her Arm-based Docker Containers easily on a Intel-based Linux machine. And with the upcoming new features built-in into the Docker Engine this multi-architecture development workflow will get better and better over time.<\/p>\n\n
For Linux users we can use at least all these basic features. And for Mac and Windows users, Docker will present an even better user experience, so stay tuned for DockerCon 2019<\/a> in San Francisco next week.<\/p>\n\n
Feedback, please<\/h3>\n\n
As always use the comments below to give us feedback and share it on Twitter or Facebook.<\/p>\n\n
Please send us your feedback on our Gitter channel<\/a> or tweet your thoughts and ideas on this project at @HypriotTweets<\/a>.<\/p>\n\n
Dieter @Quintus23M<\/a><\/p>"},{"title":"NVIDIA Jetson Nano - Upgrade Docker Engine","link":"https:\/\/blog.hypriot.com\/post\/nvidia-jetson-nano-upgrade-docker\/","pubDate":"Mon, 22 Apr 2019 08:44:52 +0200","guid":"https:\/\/blog.hypriot.com\/post\/nvidia-jetson-nano-upgrade-docker\/","description":"
In our last blogposts about the NVIDIA Jetson Nano Developer Kit - Introduction<\/a> and NVIDIA Jetson Nano - Install Docker Compose<\/a> we digged into the brand-new NVIDIA Jetson Nano Developer Kit<\/strong> and we know, that Docker 18.06.1-CE is already installed, but…<\/p>\n\n
\"jetson-desktop-login.jpg\"<\/p>\n\n
But, this isn’t the latest available version of the Docker Engine. So, I’d like to point you to a few different options on how to upgrade the Docker Engine to the very latest available version for the NVIDIA Jetson Nano.<\/p>\n\n
<\/p>\n\n
Check the current Docker Version<\/h3>\n\n
For this tutorial I’m starting again with a freshly flashed SD card image.<\/p>\n\n
Flashing from macOS just takes a few minutes with the Hypriot flash utility, which can be found here https:\/\/github.com\/hypriot\/flash<\/a>.<\/p>\n\n
$ time flash --device \/dev\/disk2 jetson-nano-sd-r32.1-2019-03-18.img\n\nIs \/dev\/disk2 correct? y\nUnmounting \/dev\/disk2 ...\nUnmount of all volumes on disk2 was successful\nUnmount of all volumes on disk2 was successful\nFlashing jetson-nano-sd-r32.1-2019-03-18.img to \/dev\/rdisk2 ...\n12.0GiB 0:03:40 [55.8MiB\/s] [=======================================================================================>] 100%\n0+196608 records in\n0+196608 records out\n12884901888 bytes transferred in 220.275160 secs (58494575 bytes\/sec)\nMounting Disk\n\nreal\t3m47.866s\nuser\t0m1.648s\nsys\t0m30.921s\n<\/code><\/pre>\n\n
Next we have to attach a computer monitor via HDMI cable, mouse and keyboard and of course an Ethernet cable in order to get an internet connect. Now connecting to a micro-USB power supply and follow the instruction on the screen to perform the initial setup of the NVIDIA Jetson Nano Developer Kit. This will take around 5 to 10 minutes and we do have a new Ubuntu 18.04 desktop running on that nice 64bit ARM Cortex-A57 developer board.<\/p>\n\n
Pro-Tip on using Docker client:<\/strong><\/p>\n\n
As we have seen in the last blogpost, we have to use sudo when we’re calling a docker command in the shell. This can be easily resolved, we only have to issue the following command ones.<\/p>\n\n
pirate@jetson-nano:~$ sudo usermod -aG docker pirate\n<\/code><\/pre>\n\n
Next, log out and log in again or just start a new shell and we don’t have to use sudo any more for our docker commands.<\/p>\n\n
Show the current version of the Docker Engine installed on the Nano.<\/p>\n\n
pirate@jetson-nano:~$ docker version\nClient:\n Version:           18.06.1-ce\n API version:       1.38\n Go version:        go1.10.1\n Git commit:        e68fc7a\n Built:             Fri Jan 25 14:35:17 2019\n OS\/Arch:           linux\/arm64\n Experimental:      false\n\nServer:\n Engine:\n  Version:          18.06.1-ce\n  API version:      1.38 (minimum version 1.12)\n  Go version:       go1.10.1\n  Git commit:       e68fc7a\n  Built:            Thu Jan 24 10:49:48 2019\n  OS\/Arch:          linux\/arm64\n  Experimental:     false\n<\/code><\/pre>\n\n
And here we see the version 18.06.1-ce<\/code> of the Docker Engine, which is installed on the Nano SD card image jetson-nano-sd-r32.1-2019-03-18.img<\/code>. And yes, this is by far not the latest nor the securest available version at all.<\/p>\n\n
Upgrade Docker Engine<\/h3>\n\n
As we’ve seen in the previous section, the installed Docker Engine version is 18.06.01-ce. Now let’s verify if there is already a newer version available.<\/p>\n\n
We can just use the apt utility from the Ubuntu package manager to determine the installed versions of any of the installed software packages. Here on Ubuntu 18.04 the Docker Engine is installed through the docker.io<\/code> package.<\/p>\n\n
pirate@jetson-nano:~$ apt list --installed | grep docker.io\n\nWARNING: apt does not have a stable CLI interface. Use with caution in scripts.\n\ndocker.io\/bionic-security,now 18.06.1-0ubuntu1.2~18.04.1 arm64 [installed,upgradable to: 18.09.2-0ubuntu1~18.04.1]\n<\/code><\/pre>\n\n
We can see the newer Docker version 18.09.2 is already available in the Ubuntu repository, so we’re going to upgrade the software package docker.io<\/code>.<\/p>\n\n
pirate@jetson-nano:~$ sudo apt-get --only-upgrade install docker.io\n<\/code><\/pre>\n\n
At the end we were able to upgrade the Docker Engine to the very latest version which is provided by the Ubuntu repository.<\/p>\n\n
pirate@jetson-nano:~$ docker version\nClient:\n Version:           18.09.2\n API version:       1.39\n Go version:        go1.10.4\n Git commit:        6247962\n Built:             Tue Feb 26 23:51:35 2019\n OS\/Arch:           linux\/arm64\n Experimental:      false\n\nServer:\n Engine:\n  Version:          18.09.2\n  API version:      1.39 (minimum version 1.12)\n  Go version:       go1.10.4\n  Git commit:       6247962\n  Built:            Wed Feb 13 00:24:14 2019\n  OS\/Arch:          linux\/arm64\n<\/code><\/pre>\n\n
Finally we have upgraded the Docker Engine to version 18.09.2. Please keep in mind, this version of the Docker Engine is provided by the Ubuntu repository. Maybe there is even a newer version available directly from the Docker open source project.<\/p>\n\n
Recommendation: Install official Docker Engine CE<\/h3>\n\n
For this step I can truely recommend to use the official Docker documentation for the Community Edition. For installing Docker Engine CE on Ubuntu you can directly follow the detailled steps at https:\/\/docs.docker.com\/install\/linux\/docker-ce\/ubuntu\/<\/a>.<\/p>\n\n
As you can see later, Docker Engine CE is also available for ARM 64bit on Ubuntu 18.04 LTS. It’s not too complicated to set it up, just follow the installation steps in this documentation and within a few minutes you have installed the official Docker Engine!<\/p>\n\n
Step 1:<\/strong> Uninstall old versions<\/p>\n\n
$ sudo apt-get remove docker docker-engine docker.io containerd runc\n<\/code><\/pre>\n\n
Step 2:<\/strong> Set up the repository<\/p>\n\n
# 1. Update the apt package index:\n$ sudo apt-get update\n\n# 2. Install packages to allow apt to use a repository over HTTPS:\n$ sudo apt-get install \\\n    apt-transport-https \\\n    ca-certificates \\\n    curl \\\n    gnupg-agent \\\n    software-properties-common\n\n# 3. Add Docker\u2019s official GPG key:\n$ curl -fsSL https:\/\/download.docker.com\/linux\/ubuntu\/gpg | sudo apt-key add -\n\n$ sudo apt-key fingerprint 0EBFCD88\npub   rsa4096 2017-02-22 [SCEA]\n      9DC8 5822 9FC7 DD38 854A  E2D8 8D81 803C 0EBF CD88\nuid           [ unknown] Docker Release (CE deb) <docker@docker.com>\nsub   rsa4096 2017-02-22 [S]\n\n# 4. Use the following command to set up the stable repository:\n#    here select the commands for "arm64"\n$ sudo add-apt-repository \\\n   "deb [arch=arm64] https:\/\/download.docker.com\/linux\/ubuntu \\\n   $(lsb_release -cs) \\\n   stable"\n<\/code><\/pre>\n\n
# Alternatively you can also select the "edge" channel for the very latest version\n$ sudo add-apt-repository \\\n   "deb [arch=arm64] https:\/\/download.docker.com\/linux\/ubuntu \\\n   $(lsb_release -cs) \\\n   edge"\n<\/code><\/pre>\n\n
Step 3:<\/strong> Install Docker CE<\/p>\n\n
# 1. Update the apt package index.\n$ sudo apt-get update\n\n# 2. Install the latest version of Docker CE and containerd\n$ sudo apt-get install docker-ce docker-ce-cli containerd.io\n\n# 3. Add user "pirate" to the group "docker", so we don't need sudo\n$ sudo usermod -aG docker pirate\n# Now logout and login again\n<\/code><\/pre>\n\n
Step 4:<\/strong> Show the installed Docker Engine version<\/p>\n\n
pirate@jetson-nano:~$ docker version\nClient:\n Version:           18.09.5\n API version:       1.39\n Go version:        go1.10.8\n Git commit:        e8ff056\n Built:             Thu Apr 11 04:48:27 2019\n OS\/Arch:           linux\/arm64\n Experimental:      false\n\nServer: Docker Engine - Community\n Engine:\n  Version:          18.09.5\n  API version:      1.39 (minimum version 1.12)\n  Go version:       go1.10.8\n  Git commit:       e8ff056\n  Built:            Thu Apr 11 04:11:17 2019\n  OS\/Arch:          linux\/arm64\n  Experimental:     false\n<\/code><\/pre>\n\n
Step 5:<\/strong> Run the docker info<\/code> command<\/p>\n\n
pirate@jetson-nano:~$ docker info\nContainers: 0\n Running: 0\n Paused: 0\n Stopped: 0\nImages: 0\nServer Version: 18.09.5\nStorage Driver: overlay2\n Backing Filesystem: extfs\n Supports d_type: true\n Native Overlay Diff: true\nLogging Driver: json-file\nCgroup Driver: cgroupfs\nPlugins:\n Volume: local\n Network: bridge host macvlan null overlay\n Log: awslogs fluentd gcplogs gelf journald json-file local logentries splunk syslog\nSwarm: inactive\nRuntimes: runc\nDefault Runtime: runc\nInit Binary: docker-init\ncontainerd version: bb71b10fd8f58240ca47fbb579b9d1028eea7c84\nrunc version: 2b18fe1d885ee5083ef9f0838fee39b62d653e30\ninit version: fec3683\nSecurity Options:\n seccomp\n  Profile: default\nKernel Version: 4.9.140-tegra\nOperating System: Ubuntu 18.04.2 LTS\nOSType: linux\nArchitecture: aarch64\nCPUs: 4\nTotal Memory: 3.868GiB\nName: jetson-nano\nID: 4JER:EIWM:QFNF:6N2C:YUW3:YES2:RSP5:Z4D2:7PKI:YAOT:G5O7:5N25\nDocker Root Dir: \/var\/lib\/docker\nDebug Mode (client): false\nDebug Mode (server): false\nRegistry: https:\/\/index.docker.io\/v1\/\nLabels:\nExperimental: false\nInsecure Registries:\n 127.0.0.0\/8\nLive Restore Enabled: false\nProduct License: Community Engine\n<\/code><\/pre>\n\n
Step 6:<\/strong>  Start first Docker container<\/p>\n\n
pirate@jetson-nano:~$ docker run hello-world\nUnable to find image 'hello-world:latest' locally\nlatest: Pulling from library\/hello-world\n3b4173355427: Pull complete\nDigest: sha256:92695bc579f31df7a63da6922075d0666e565ceccad16b59c3374d2cf4e8e50e\nStatus: Downloaded newer image for hello-world:latest\n\nHello from Docker!\nThis message shows that your installation appears to be working correctly.\n\nTo generate this message, Docker took the following steps:\n 1. The Docker client contacted the Docker daemon.\n 2. The Docker daemon pulled the "hello-world" image from the Docker Hub.\n    (arm64v8)\n 3. The Docker daemon created a new container from that image which runs the\n    executable that produces the output you are currently reading.\n 4. The Docker daemon streamed that output to the Docker client, which sent it\n    to your terminal.\n\nTo try something more ambitious, you can run an Ubuntu container with:\n $ docker run -it ubuntu bash\n\nShare images, automate workflows, and more with a free Docker ID:\n https:\/\/hub.docker.com\/\n\nFor more examples and ideas, visit:\n https:\/\/docs.docker.com\/get-started\/\n<\/code><\/pre>\n\n
\"jetson-desktop-docker-ce.jpg\"<\/p>\n\n
Conclusion<\/h3>\n\n
As you could see, on the Jetson Nano DevKit there is already a version of the Docker Engine installed and it’s maintained by the Ubuntu project. But this is not the latest version available and it will not get updated fast enough to include all important security fixes in time.<\/p>\n\n
Therefore I’d like to strongly recommend to use the Docker Engine CE from the official Docker project. It’s well maintained and updated in time. All installation steps and options are also extremely well documented at https:\/\/docs.docker.com\/install\/linux\/docker-ce\/ubuntu\/<\/a>.<\/p>\n\n
And in case you need more help or having some technical questions about running Docker on an ARM 64bit system like the NVIDIA Jetson Nano, there is an arm<\/code> Slack channel available for you at the DockerCommunity Slack<\/a>.<\/p>\n\n
Feedback, please<\/h3>\n\n
As always use the comments below to give us feedback and share it on Twitter or Facebook.<\/p>\n\n
Please send us your feedback on our Gitter channel<\/a> or tweet your thoughts and ideas on this project at @HypriotTweets<\/a>.<\/p>\n\n
Dieter @Quintus23M<\/a><\/p>"},{"title":"NVIDIA Jetson Nano - Install Docker Compose","link":"https:\/\/blog.hypriot.com\/post\/nvidia-jetson-nano-install-docker-compose\/","pubDate":"Sat, 20 Apr 2019 09:51:52 +0200","guid":"https:\/\/blog.hypriot.com\/post\/nvidia-jetson-nano-install-docker-compose\/","description":"
In our last blogpost NVIDIA Jetson Nano Developer Kit - Introduction<\/a> we digged into the brand-new NVIDIA Jetson Nano Developer Kit<\/strong> and we did found out, that Docker 18.06.1-CE is already pre-installed on this great ARM board.<\/p>\n\n
\"Jetson-Nano-Upacked.jpg\"<\/p>\n\n
Today, I want to share some more details on how you can easily install Docker Compose on the Jetson Nano.<\/p>\n\n
<\/p>\n\n
Install Docker Compose<\/h3>\n\n
Sadly, there is no binary of Docker Compose<\/a> available we could install directly on an ARM 64bit board like the NVIDIA Jetson Nano. If we are looking at the official GitHub release page for Docker Compose https:\/\/github.com\/docker\/compose\/releases\/tag\/1.24.0<\/a> there are only binaries provided for Intel x86-64 based operating systems like Linux, macOS and Windows - but nothing for ARM 32bit or 64bit systems.<\/p>\n\n
This isn’t too bad, because Docker Compose is based upon Python and maybe it’s easier to install if Python is already available on our board. So, let’s check this out the easy way.<\/p>\n\n
Docker Compose is based upon Python<\/h3>\n\n
First, check out if Python is already available on the Jetson Nano. As we have a complete Ubuntu 18.04 LTS desktop operation system this is very likely. But let’s verify it directly on a shell running on the Nano. This can be done from the desktop terminal app or via a SSH shell.<\/p>\n\n
pirate@jetson-nano:~$ python --version\nPython 2.7.15rc1\n<\/code><\/pre>\n\n
OK, this looks pretty promising. We do have the Python version 2.7.15rc1 already installed on the Nano.<\/p>\n\n
Let’s try to install Docker Compose via Python PIP. So, first we need to install Python PIP itself.<\/p>\n\n
# first, we have to install Python PIP\n$ sudo apt-get update -y\n$ sudo apt-get install -y curl\n$ curl -sSL https:\/\/bootstrap.pypa.io\/get-pip.py | sudo python\n\n# verify the installed Python PIP version\npirate@jetson-nano:~$ pip --version\npip 19.0.3 from \/usr\/local\/lib\/python2.7\/dist-packages\/pip (python 2.7)\n<\/code><\/pre>\n\n
Now, let’s try to install Docker Compose via Python PIP. Here we’re going to pin the latest version of Docker Compose, which is 1.24.0 of the time of writing this blogpost.<\/p>\n\n
# install Docker Compose via pip\nexport DOCKER_COMPOSE_VERSION=1.24.0\nsudo pip install docker-compose=="${DOCKER_COMPOSE_VERSION}"\n<\/code><\/pre>\n\n
But this will throw a lot of error messages because it seems we do miss some required build dependencies. If you’re interested here are the full details.<\/p>\n\n
pirate@jetson-nano:~$ curl -sSL https:\/\/bootstrap.pypa.io\/get-pip.py | sudo python\nDEPRECATION: Python 2.7 will reach the end of its life on January 1st, 2020. Please upgrade your Python as Python 2.7 won't be maintained after that date. A future version of pip will drop support for Python 2.7.\nThe directory '\/home\/pirate\/.cache\/pip\/http' or its parent directory is not owned by the current user and the cache has been disabled. Please check the permissions and owner of that directory. If executing pip with sudo, you may want sudo's -H flag.\nThe directory '\/home\/pirate\/.cache\/pip' or its parent directory is not owned by the current user and caching wheels has been disabled. check the permissions and owner of that directory. If executing pip with sudo, you may want sudo's -H flag.\nCollecting pip\n  Downloading https:\/\/files.pythonhosted.org\/packages\/d8\/f3\/413bab4ff08e1fc4828dfc59996d721917df8e8583ea85385d51125dceff\/pip-19.0.3-py2.py3-none-any.whl (1.4MB)\n    100% |\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 1.4MB 843kB\/s\nCollecting setuptools\n  Downloading https:\/\/files.pythonhosted.org\/packages\/c8\/b0\/cc6b7ba28d5fb790cf0d5946df849233e32b8872b6baca10c9e002ff5b41\/setuptools-41.0.0-py2.py3-none-any.whl (575kB)\n    100% |\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 583kB 761kB\/s\nCollecting wheel\n  Downloading https:\/\/files.pythonhosted.org\/packages\/96\/ba\/a4702cbb6a3a485239fbe9525443446203f00771af9ac000fa3ef2788201\/wheel-0.33.1-py2.py3-none-any.whl\nInstalling collected packages: pip, setuptools, wheel\nSuccessfully installed pip-19.0.3 setuptools-41.0.0 wheel-0.33.1\npirate@jetson-nano:~$ pip --version\npip 19.0.3 from \/usr\/local\/lib\/python2.7\/dist-packages\/pip (python 2.7)\n\npirate@jetson-nano:~$ export DOCKER_COMPOSE_VERSION=1.24.0\npirate@jetson-nano:~$ sudo pip install docker-compose=="${DOCKER_COMPOSE_VERSION}"\nDEPRECATION: Python 2.7 will reach the end of its life on January 1st, 2020. Please upgrade your Python as Python 2.7 won't be maintained after that date. A future version of pip will drop support for Python 2.7.\nThe directory '\/home\/pirate\/.cache\/pip\/http' or its parent directory is not owned by the current user and the cache has been disabled. Please check the permissions and owner of that directory. If executing pip with sudo, you may want sudo's -H flag.\nThe directory '\/home\/pirate\/.cache\/pip' or its parent directory is not owned by the current user and caching wheels has been disabled. check the permissions and owner of that directory. If executing pip with sudo, you may want sudo's -H flag.\nCollecting docker-compose==1.24.0\n  Downloading https:\/\/files.pythonhosted.org\/packages\/51\/56\/5745e66b33846e92a8814466c163f165a26fadad8b33afe381e8b6c3f652\/docker_compose-1.24.0-py2.py3-none-any.whl (134kB)\n    100% |\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 143kB 2.3MB\/s\nCollecting cached-property<2,>=1.2.0 (from docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/3b\/86\/85c1be2e8db9e13ef9a350aecd6dea292bd612fa288c2f40d035bb750ded\/cached_property-1.5.1-py2.py3-none-any.whl\nCollecting docopt<0.7,>=0.6.1 (from docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/a2\/55\/8f8cab2afd404cf578136ef2cc5dfb50baa1761b68c9da1fb1e4eed343c9\/docopt-0.6.2.tar.gz\nCollecting jsonschema<3,>=2.5.1 (from docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/77\/de\/47e35a97b2b05c2fadbec67d44cfcdcd09b8086951b331d82de90d2912da\/jsonschema-2.6.0-py2.py3-none-any.whl\nCollecting backports.ssl-match-hostname>=3.5; python_version < "3.5" (from docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/ff\/2b\/8265224812912bc5b7a607c44bf7b027554e1b9775e9ee0de8032e3de4b2\/backports.ssl_match_hostname-3.7.0.1.tar.gz\nRequirement already satisfied: six<2,>=1.3.0 in \/usr\/lib\/python2.7\/dist-packages (from docker-compose==1.24.0) (1.11.0)\nCollecting ipaddress>=1.0.16; python_version < "3.3" (from docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/fc\/d0\/7fc3a811e011d4b388be48a0e381db8d990042df54aa4ef4599a31d39853\/ipaddress-1.0.22-py2.py3-none-any.whl\nCollecting requests!=2.11.0,!=2.12.2,!=2.18.0,<2.21,>=2.6.1 (from docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/ff\/17\/5cbb026005115301a8fb2f9b0e3e8d32313142fe8b617070e7baad20554f\/requests-2.20.1-py2.py3-none-any.whl (57kB)\n    100% |\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 61kB 2.8MB\/s\nCollecting texttable<0.10,>=0.9.0 (from docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/02\/e1\/2565e6b842de7945af0555167d33acfc8a615584ef7abd30d1eae00a4d80\/texttable-0.9.1.tar.gz\nCollecting dockerpty<0.5,>=0.4.1 (from docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/8d\/ee\/e9ecce4c32204a6738e0a5d5883d3413794d7498fe8b06f44becc028d3ba\/dockerpty-0.4.1.tar.gz\nCollecting enum34<2,>=1.0.4; python_version < "3.4" (from docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/c5\/db\/e56e6b4bbac7c4a06de1c50de6fe1ef3810018ae11732a50f15f62c7d050\/enum34-1.1.6-py2-none-any.whl\nCollecting PyYAML<4.3,>=3.10 (from docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/9e\/a3\/1d13970c3f36777c583f136c136f804d70f500168edc1edea6daa7200769\/PyYAML-3.13.tar.gz (270kB)\n    100% |\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 276kB 2.9MB\/s\nCollecting websocket-client<1.0,>=0.32.0 (from docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/29\/19\/44753eab1fdb50770ac69605527e8859468f3c0fd7dc5a76dd9c4dbd7906\/websocket_client-0.56.0-py2.py3-none-any.whl (200kB)\n    100% |\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 204kB 3.2MB\/s\nCollecting docker[ssh]<4.0,>=3.7.0 (from docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/48\/68\/c3afca1a5aa8d2997ec3b8ee822a4d752cf85907b321f07ea86888545152\/docker-3.7.2-py2.py3-none-any.whl (134kB)\n    100% |\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 143kB 3.5MB\/s\nCollecting functools32; python_version == "2.7" (from jsonschema<3,>=2.5.1->docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/c5\/60\/6ac26ad05857c601308d8fb9e87fa36d0ebf889423f47c3502ef034365db\/functools32-3.2.3-2.tar.gz\nCollecting urllib3<1.25,>=1.21.1 (from requests!=2.11.0,!=2.12.2,!=2.18.0,<2.21,>=2.6.1->docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/df\/1c\/59cca3abf96f991f2ec3131a4ffe72ae3d9ea1f5894abe8a9c5e3c77cfee\/urllib3-1.24.2-py2.py3-none-any.whl (131kB)\n    100% |\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 133kB 3.2MB\/s\nCollecting chardet<3.1.0,>=3.0.2 (from requests!=2.11.0,!=2.12.2,!=2.18.0,<2.21,>=2.6.1->docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/bc\/a9\/01ffebfb562e4274b6487b4bb1ddec7ca55ec7510b22e4c51f14098443b8\/chardet-3.0.4-py2.py3-none-any.whl (133kB)\n    100% |\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 143kB 3.5MB\/s\nCollecting idna<2.8,>=2.5 (from requests!=2.11.0,!=2.12.2,!=2.18.0,<2.21,>=2.6.1->docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/4b\/2a\/0276479a4b3caeb8a8c1af2f8e4355746a97fab05a372e4a2c6a6b876165\/idna-2.7-py2.py3-none-any.whl (58kB)\n    100% |\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 61kB 5.8MB\/s\nCollecting certifi>=2017.4.17 (from requests!=2.11.0,!=2.12.2,!=2.18.0,<2.21,>=2.6.1->docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/60\/75\/f692a584e85b7eaba0e03827b3d51f45f571c2e793dd731e598828d380aa\/certifi-2019.3.9-py2.py3-none-any.whl (158kB)\n    100% |\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 163kB 353kB\/s\nCollecting docker-pycreds>=0.4.0 (from docker[ssh]<4.0,>=3.7.0->docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/f5\/e8\/f6bd1eee09314e7e6dee49cbe2c5e22314ccdb38db16c9fc72d2fa80d054\/docker_pycreds-0.4.0-py2.py3-none-any.whl\nCollecting paramiko>=2.4.2; extra == "ssh" (from docker[ssh]<4.0,>=3.7.0->docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/cf\/ae\/94e70d49044ccc234bfdba20114fa947d7ba6eb68a2e452d89b920e62227\/paramiko-2.4.2-py2.py3-none-any.whl (193kB)\n    100% |\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 194kB 3.2MB\/s\nCollecting pyasn1>=0.1.7 (from paramiko>=2.4.2; extra == "ssh"->docker[ssh]<4.0,>=3.7.0->docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/7b\/7c\/c9386b82a25115cccf1903441bba3cbadcfae7b678a20167347fa8ded34c\/pyasn1-0.4.5-py2.py3-none-any.whl (73kB)\n    100% |\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 81kB 3.2MB\/s\nCollecting bcrypt>=3.1.3 (from paramiko>=2.4.2; extra == "ssh"->docker[ssh]<4.0,>=3.7.0->docker-compose==1.24.0)\n  Downloading https:\/\/files.pythonhosted.org\/packages\/ce\/3a\/3d540b9f5ee8d92ce757eebacf167b9deedb8e30aedec69a2a072b2399bb\/bcrypt-3.1.6.tar.gz (42kB)\n    100% |\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 51kB 8.4MB\/s\n  Installing build dependencies ... error\n  Complete output from command \/usr\/bin\/python \/usr\/local\/lib\/python2.7\/dist-packages\/pip install --ignore-installed --no-user --prefix \/tmp\/pip-build-env-qwOdIW\/overlay --no-warn-script-location --no-binary :none: --only-binary :none: -i https:\/\/pypi.org\/simple -- setuptools wheel "cffi>=1.1; python_implementation != 'PyPy'":\n  DEPRECATION: Python 2.7 will reach the end of its life on January 1st, 2020. Please upgrade your Python as Python 2.7 won't be maintained after that date. A future version of pip will drop support for Python 2.7.\n  The directory '\/home\/pirate\/.cache\/pip\/http' or its parent directory is not owned by the current user and the cache has been disabled. Please check the permissions and owner of that directory. If executing pip with sudo, you may want sudo's -H flag.\n  The directory '\/home\/pirate\/.cache\/pip' or its parent directory is not owned by the current user and caching wheels has been disabled. check the permissions and owner of that directory. If executing pip with sudo, you may want sudo's -H flag.\n  Collecting setuptools\n    Downloading https:\/\/files.pythonhosted.org\/packages\/c8\/b0\/cc6b7ba28d5fb790cf0d5946df849233e32b8872b6baca10c9e002ff5b41\/setuptools-41.0.0-py2.py3-none-any.whl (575kB)\n  Collecting wheel\n    Downloading https:\/\/files.pythonhosted.org\/packages\/96\/ba\/a4702cbb6a3a485239fbe9525443446203f00771af9ac000fa3ef2788201\/wheel-0.33.1-py2.py3-none-any.whl\n  Collecting cffi>=1.1\n    Downloading https:\/\/files.pythonhosted.org\/packages\/93\/1a\/ab8c62b5838722f29f3daffcc8d4bd61844aa9b5f437341cc890ceee483b\/cffi-1.12.3.tar.gz (456kB)\n  Collecting pycparser (from cffi>=1.1)\n    Downloading https:\/\/files.pythonhosted.org\/packages\/68\/9e\/49196946aee219aead1290e00d1e7fdeab8567783e83e1b9ab5585e6206a\/pycparser-2.19.tar.gz (158kB)\n  Installing collected packages: setuptools, wheel, pycparser, cffi\n    Running setup.py install for pycparser: started\n      Running setup.py install for pycparser: finished with status 'done'\n    Running setup.py install for cffi: started\n      Running setup.py install for cffi: finished with status 'error'\n      Complete output from command \/usr\/bin\/python -u -c "import setuptools, tokenize;__file__='\/tmp\/pip-install-XfNg4G\/cffi\/setup.py';f=getattr(tokenize, 'open', open)(__file__);code=f.read().replace('\\r\\n', '\\n');f.close();exec(compile(code, __file__, 'exec'))" install --record \/tmp\/pip-record-lpvT6b\/install-record.txt --single-version-externally-managed --prefix \/tmp\/pip-build-env-qwOdIW\/overlay --compile:\n      Package libffi was not found in the pkg-config search path.\n      Perhaps you should add the directory containing `libffi.pc'\n      to the PKG_CONFIG_PATH environment variable\n      No package 'libffi' found\n      Package libffi was not found in the pkg-config search path.\n      Perhaps you should add the directory containing `libffi.pc'\n      to the PKG_CONFIG_PATH environment variable\n      No package 'libffi' found\n      Package libffi was not found in the pkg-config search path.\n      Perhaps you should add the directory containing `libffi.pc'\n      to the PKG_CONFIG_PATH environment variable\n      No package 'libffi' found\n      Package libffi was not found in the pkg-config search path.\n      Perhaps you should add the directory containing `libffi.pc'\n      to the PKG_CONFIG_PATH environment variable\n      No package 'libffi' found\n      Package libffi was not found in the pkg-config search path.\n      Perhaps you should add the directory containing `libffi.pc'\n      to the PKG_CONFIG_PATH environment variable\n      No package 'libffi' found\n      running install\n      running build\n      running build_py\n      creating build\n      creating build\/lib.linux-aarch64-2.7\n      creating build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/ffiplatform.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/vengine_gen.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/setuptools_ext.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/verifier.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/backend_ctypes.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/vengine_cpy.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/pkgconfig.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/recompiler.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/api.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/cparser.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/__init__.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/cffi_opcode.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/error.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/lock.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/model.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/commontypes.py -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/_cffi_include.h -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/parse_c_type.h -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/_embedding.h -> build\/lib.linux-aarch64-2.7\/cffi\n      copying cffi\/_cffi_errors.h -> build\/lib.linux-aarch64-2.7\/cffi\n      running build_ext\n      building '_cffi_backend' extension\n      creating build\/temp.linux-aarch64-2.7\n      creating build\/temp.linux-aarch64-2.7\/c\n      aarch64-linux-gnu-gcc -pthread -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -fno-strict-aliasing -Wdate-time -D_FORTIFY_SOURCE=2 -g -fdebug-prefix-map=\/build\/python2.7-ytvI67\/python2.7-2.7.15~rc1=. -fstack-protector-strong -Wformat -Werror=format-security -fPIC -DUSE__THREAD -DHAVE_SYNC_SYNCHRONIZE -I\/usr\/include\/ffi -I\/usr\/include\/libffi -I\/usr\/include\/python2.7 -c c\/_cffi_backend.c -o build\/temp.linux-aarch64-2.7\/c\/_cffi_backend.o\n      c\/_cffi_backend.c:15:10: fatal error: ffi.h: No such file or directory\n       #include <ffi.h>\n                ^~~~~~~\n      compilation terminated.\n      error: command 'aarch64-linux-gnu-gcc' failed with exit status 1\n\n      ----------------------------------------\n  Command "\/usr\/bin\/python -u -c "import setuptools, tokenize;__file__='\/tmp\/pip-install-XfNg4G\/cffi\/setup.py';f=getattr(tokenize, 'open', open)(__file__);code=f.read().replace('\\r\\n', '\\n');f.close();exec(compile(code, __file__, 'exec'))" install --record \/tmp\/pip-record-lpvT6b\/install-record.txt --single-version-externally-managed --prefix \/tmp\/pip-build-env-qwOdIW\/overlay --compile" failed with error code 1 in \/tmp\/pip-install-XfNg4G\/cffi\/\n\n  ----------------------------------------\nCommand "\/usr\/bin\/python \/usr\/local\/lib\/python2.7\/dist-packages\/pip install --ignore-installed --no-user --prefix \/tmp\/pip-build-env-qwOdIW\/overlay --no-warn-script-location --no-binary :none: --only-binary :none: -i https:\/\/pypi.org\/simple -- setuptools wheel "cffi>=1.1; python_implementation != 'PyPy'"" failed with error code 1 in None\n<\/code><\/pre>\n\n
After some investigation we’re able to identify the missing build dependencies and here are the required commands to install it.<\/p>\n\n
# install Docker Compose build dependencies for Ubuntu 18.04 on aarch64\n$ sudo apt-get install -y libffi-dev\n$ sudo apt-get install -y python-openssl\n<\/code><\/pre>\n\n
Install Docker Compose via Python PIP<\/h3>\n\n
TL;DR These are all commands you need to install Docker Compose on a freshly installed NVIDIA Jetson Nano.<\/p>\n\n
# step 1, install Python PIP\n$ sudo apt-get update -y\n$ sudo apt-get install -y curl\n$ curl -sSL https:\/\/bootstrap.pypa.io\/get-pip.py | sudo python\n\n# step 2, install Docker Compose build dependencies for Ubuntu 18.04 on aarch64\n$ sudo apt-get install -y libffi-dev\n$ sudo apt-get install -y python-openssl\n\n# step 3, install latest Docker Compose via pip\n$ export DOCKER_COMPOSE_VERSION=1.24.0\n$ sudo pip install docker-compose=="${DOCKER_COMPOSE_VERSION}"\n<\/code><\/pre>\n\n
Conclusion<\/h3>\n\n
Finally we have the latest available Docker Compose version 1.24.0 installed on the NVIDIA Jetson Nano. It just takes a few minutes.<\/p>\n\n
pirate@jetson-nano:~$ docker-compose version\ndocker-compose version 1.24.0, build 0aa5906\ndocker-py version: 3.7.2\nCPython version: 2.7.15rc1\nOpenSSL version: OpenSSL 1.1.0g  2 Nov 2017\n\npirate@jetson-nano:~$ docker-compose --version\ndocker-compose version 1.24.0, build 0aa5906\n<\/code><\/pre>\n\n
Another Pro-Tip on using Docker command line interface<\/h3>\n\n
As we have seen in the last blogpost, we have to use sudo when we’re calling a docker command in the shell. This can be easy resolved, we only have to issue the following command ones.<\/p>\n\n
pirate@jetson-nano:~$ sudo usermod -aG docker pirate\n<\/code><\/pre>\n\n
Next, log out and log in again or just start a new shell and we don’t have to use sudo any more for our docker commands.<\/p>\n\n
pirate@jetson-nano:~$ docker version\nClient:\n Version:           18.06.1-ce\n API version:       1.38\n Go version:        go1.10.1\n Git commit:        e68fc7a\n Built:             Fri Jan 25 14:35:17 2019\n OS\/Arch:           linux\/arm64\n Experimental:      false\n\nServer:\n Engine:\n  Version:          18.06.1-ce\n  API version:      1.38 (minimum version 1.12)\n  Go version:       go1.10.1\n  Git commit:       e68fc7a\n  Built:            Thu Jan 24 10:49:48 2019\n  OS\/Arch:          linux\/arm64\n  Experimental:     false\n<\/code><\/pre>\n\n
Feedback, please<\/h3>\n\n
As always use the comments below to give us feedback and share it on Twitter or Facebook.<\/p>\n\n
Please send us your feedback on our Gitter channel<\/a> or tweet your thoughts and ideas on this project at @HypriotTweets<\/a>.<\/p>\n\n
Dieter @Quintus23M<\/a><\/p>"},{"title":"NVIDIA Jetson Nano Developer Kit - Introduction","link":"https:\/\/blog.hypriot.com\/post\/nvidia-jetson-nano-intro\/","pubDate":"Fri, 19 Apr 2019 14:16:25 +0200","guid":"https:\/\/blog.hypriot.com\/post\/nvidia-jetson-nano-intro\/","description":"
Let me introduce the brand new NVIDIA Jetson Nano Developer Kit<\/strong>, which is basically a quad-core 64bit ARM Cortex-A57 CPU with 128 GPU cores - suitable for all kinds of maker ideas: AI, Robotics, and of course for running Docker Containers…<\/p>\n\n
\"Jetson-Nano-Box.jpg\"<\/p>\n\n
Unboxing<\/h3>\n\n
Let’s unbox the board and do the initial configuration…<\/p>\n\n
<\/p>\n\n
\"Jetson-Nano-Upacked.jpg\"<\/p>\n\n
When opening the box we’ll find the board itself and a short getting started flyer which heads us to the NVIDIA website https:\/\/NVIDIA.com\/JetsonNano-Start<\/a> where we can find all basic information about the new DevKit.<\/p>\n\n
Please note, there are no other parts included in the DevKit. So, in order to use the DevKit you’ll need some additional accessories:<\/p>\n\n