How to check process or pid running in nvidia-smi and stop it using a bash script?

#!/bin/bash
total=1
base=1
counter=0
sub='/temp/ninja/build/main'
TEMPFILE=/tmp/$$.tmp
nvidia-smi --query-compute-apps=gpu_bus_id,pid,process_name,used_memory --format=csv,noheader,nounits | while read line; do
        if [[ "$line" == *"$sub"* ]]; then
                $((counter = counter+1))
                echo $counter > $TEMPFILE
                echo "$line, cc: $counter"
        fi
done
echo $(cat $TEMPFILE)
if [ "$(cat $TEMPFILE)" -le "$total" ]; then
        docker stop ninja_server
        docker rm ninja_server
        docker stop ninja_client
        docker rm ninja_client
fi

~   

During docker build . or Dockerfile, CUDA driver version is insufficient for CUDA runtime version?

sudo vi /etc/docker/daemon.json

{
 "default-runtime": "nvidia",
  "runtimes": {
     "nvidia": {
         "path": "/usr/bin/nvidia-container-runtime",
         "runtimeArgs": []
     }
 },

}

sudo service docker restart

* you can try to print the nvidia-smi when building a docker image using Dockerfile 

inside your build.sh, add the following command, DOCKER_BUILDKIT=0 :-

DOCKER_BUILDKIT=0 docker build . -t image_name

How to setup Kubernetes in Ubuntu?

https://phoenixnap.com/kb/install-minikube-on-ubuntu

Home » DevOps and Development » How to Install Minikube on Ubuntu 18.04 / 20.04

Introduction

Minikube is an open source tool that allows you to set up a single-node Kubernetes cluster on your local machine. The cluster is run inside a virtual machine and includes Docker, allowing you to run containers inside the node.

This is an excellent way to test in a Kubernetes environment locally, without using up too much resources.

This tutorial will show you how to install Minikube on Ubuntu 18.04 or 20.04.

Prerequisites

• A system running Ubuntu 18.04 Bionic Beaver or Ubuntu 20.04
• A user account with sudo privileges
• Access to a terminal window / command line (Ctrl+Alt+T, search > terminal)

How to Install Minikube on Ubuntu

To install Minikube on Ubuntu, follow the steps outlined below. Besides installation instructions, you can also find some basic commands for working inside your local single-node cluster.

Step 1: Update System and Install Required Packages

Before installing any software, you need to update and upgrade the system you are working on. To do so, run the commands:

sudo apt-get update -y

sudo apt-get upgrade -y

Also, make sure to install (or check whether you already have) the following required packages:

sudo apt-get install curl

sudo apt-get install apt-transport-https

In the image below, the output informs that the packages are already installed.

Step 2: Install VirtualBox Hypervisor

As mentioned above, you need a virtual machine in which you can set up your single node cluster with Minikube. Depending on your preference, you can use VirtualBox or KVM.

This guide will show you how to install Minikube with VirtualBox.

1. To install VirtualBox on Ubuntu, run the command:

sudo apt install virtualbox virtualbox-ext-pack

2. Confirm the installation with y and hit Enter.

3. Next, the licence agreement appears on the screen. Press Tab and then Enter to continue.

4. The installer asks you to agree with the terms of the VirtualBox PUEL license by selecting Yes.

5. Wait for the installation to complete and then move on to the next step.

Note: For the VirtualBox hypervisor to work, hardware virtualization must be enabled in your system BIOS.

Step 3: Install Minikube

With VirtualBox set up, move on to installing Minikube on your Ubuntu system.

1. First, download the latest Minikube binary using the wget command:

wget https://storage.googleapis.com/minikube/releases/latest/minikube-linux-amd64

2. Copy the downloaded file and store it into the /usr/local/bin/minikube directory with:

sudo cp minikube-linux-amd64 /usr/local/bin/minikube

There will be no output if the command was executed correctly.

3. Next, give the file executive permission using the chmod command:

sudo chmod 755 /usr/local/bin/minikube

Again, there will be no output.

4. Finally, verify you have successfully installed Minikube by checking the version of the software:

minikube version

The output should display the version number of the software, as in the image below.

Step 4: Install Kubectl

To deploy and manage clusters, you need to install kubectl, the official command line tool for Kubernetes.

1. Download kubectl with the following command:

curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/amd64/kubectl"

2. Install the kubectl

sudo install -o root -g root -m 0755 kubectl /usr/local/bin/kubectl

3. Verify the installation by checking the version of your kubectl instance:

kubectl version --client

kubectl version --client --output=yaml

Step 5: Start Minikube

Once you have set up all the required software, you are ready to start Minikube.

Run the following command:

minikube start

First, the system downloads the Minikube ISO file from an online source and the localkube binary. Then, it creates a virtual machine in VirtualBox within which it starts and configures a single node cluster.

Managing Kubernetes with Minikube

Common Minikube Commands

To see the kubectl configuration use the command:

kubectl config view

To show the cluster information:

kubectl cluster-info

To check running nodes use the following command:

kubectl get nodes

To see a list of all the Minikube pods run:

kubectl get pod

To ssh into the Minikube VM:

minikube ssh

To exit out of the shell run:

exit

To stop running the single node cluster type:

minikube stop

To check its status use:

minikube status

To delete the single node cluster:

minikube delete

To see a list of installed Minikube add-ons:

minikube addons list

Access Minikube Dashboard

Minikube comes with a dashboard add-on by default. The web dashboard provides a way to manage your Kubernetes cluster without actually running commands in the terminal.

To enable and access the Minikube dashboard via terminal, run the following command:

minikube dashboard

Once you exit the terminal, the process will end and the Minikube dashboard will shut down.

Alternatively, you can access the dashboard directly via browser.

To do so, acquire the dashboard’s IP address:

minikube dashboard --url

Access your Minikube dashboard by browsing to your dashboard’s IP address.

Conclusion

By following this article, you should have successfully installed and configured Minikube on Ubuntu 18.04 or 20.04. You can now test and master the art of Kubernetes on your local machine with the help of a single Minikube node.

Check out our guide on how to install Minikube on CentOS if you are using this version of Linux distribution.

How to git push origin main in ubuntu?

Github is no longer support the password authentication.

Therefore we need to use ssh authentication as suggested by github.

1. open a terminal

https://docs.github.com/en/authentication/connecting-to-github-with-ssh/generating-a-new-ssh-key-and-adding-it-to-the-ssh-agent

ssh-keygen -t ed25519 -C "your_email@example.com"

> Enter a file in which to save the key (/home/you/.ssh/algorithm): [Press enter] 

> Enter passphrase (empty for no passphrase): [Press enter]

> Enter same passphrase again: [Press enter]

eval "$(ssh-agent -s)"

ssh-add ~/.ssh/id_ed25519  

cat ~/.ssh/id_ed25519.pub

2. Follow this page, to add the ssh key into github:-

https://docs.github.com/en/authentication/connecting-to-github-with-ssh/adding-a-new-ssh-key-to-your-github-account 

3. Authenticate SSH connection

https://docs.github.com/en/authentication/connecting-to-github-with-ssh/testing-your-ssh-connection

ssh -T git@github.com

> The authenticity of host 'github.com (IP ADDRESS)' can't be established.
> RSA key fingerprint is SHA256:nThbg6kXUpJWGl7E1IGOCspRomTxdCARLviKw6E5SY8.
> Are you sure you want to continue connecting (yes/no)?

> Hi username! You've successfully authenticated, but GitHub does not
> provide shell access.

4. Set the remote URL

git remote set-url origin git@github.com:username/repo_name.git 

 

5. Git Push the files into github repo

git add .

git commit -m "what you have done"

git push origin main

 

(if you want to pull again the repo, make sure you have downloaded the repo before, cd to that repo folder in your computer and pull it)

git pull origin main

 

(otherwise you need to git clone the repo again)

git clone xxx.git

 

Done!

How to create a user manual using Figma?

 https://www.figma.com/

how to debug the cpp in docker container with visual studio code?

   ctrl+shift+p -> cmake configure, cmake build, cmake debug, cmake clean, etc

 

 

Error1:

 

*make sure you have installed c++ or c++ extension pack inside docker