DEV Community: DevOps Pass AI

Making UI for Helm in ~100 lines of code

DevOps Pass AI — Tue, 30 Jul 2024 14:55:19 +0000

Hey folks, pretty often I can hear from colleagues "We're using custom solution/API/tool for ..." Which usually means someone create some tool, but usually forgot to add any convenient interface except shell :D

Let's try on example of Helm use DevOps Pass plugins feature and create UI for it.
In the same way you can create UI for any API/CLI tool you have in your stack.

📝 Requirements

What we want to achieve:

List installed Helm releases in current Kubernetes context
List and manage Helm repos
List and install Helm chart from list

Ok, so we need three doc's to list entities and few more actions for that docs.

Lets go!

⛑️ Add Helm App

First of all let's add Helm into the list of DOP applications.

Go to "DevOps Pass AI" application (add if not added from apps list) and click "Add App" action.

Fill necessary fields. Dont forget about github repo name - helm/helm - https://github.com/helm/helm, this field will help to generate "Install Helm" action

Click create and reload plugins (click on your avatar at the right top). After that you will be able to add it from applications list.

📦 Helm repos

Now lets create doc types for Helm releases, Helm repos and Helm charts, use "Add document type" action from DOP app:

It will generate for you Python files list this:

import subprocess
import json
import yaml
import requests
import cdx

def error(msg: str):
    return [
            {
                'name': 'ERROR',
                'icon': 'assets/icons/general/error.png',
                'error': '```

' + msg + '

```'
            }
        ]

def list():
    # Run the command and capture the output
    #
    # output = subprocess.run(['docker', 'compose', 'ls', '--format', 'json'], text=True, check=True, capture_output=True)
    # # Parse the output to extract environment information
    # repos = json.loads(output.stdout)

    # Kick API and return result
    #
    # url = f'https://some.url.com/'
    # token = cdx.settings.get('github.token')
    # headers = {
    #     'Authorization': f"Bearer {token}",
    # }
    # resp = requests.get(url, headers=headers)
    # if resp.status_code == 200:
    #     return resp.json()
    # else:
    #     return error(f"Something went wrong: {resp.content}")

    return error("Not implemented!")

There is an example for HTTP API call or shell command run.

First of all Helm has wonderful feature - output in JSON format, so we can use it to receive currently installed repos, releases and existing charts.

Repos, simple-stupid, just return as-is:

import json
import subprocess
import re
import cdx

def list():
    try:
        # Run the command and capture the output
        output = subprocess.run(['helm', 'repo', 'list', '-o', 'json'], text=True, check=True, capture_output=True)
        return json.loads(output.stdout)

    except subprocess.CalledProcessError as e:
        return [{'name': 'ERROR', 'icon': 'assets/icons/general/error.png', 'error': f"{e}\n{e.output}\n{e.stderr}"}]
    except FileNotFoundError as e:
        return [
            {
                'name': 'ERROR',
                'icon': 'assets/icons/general/error.png',
                'error': f"Can't find 'helm' in PATH, looks like its not installed, please install first"
            }
        ]

Helm releases. Icon mappings used to make output a bit prettier and if helm name is one of predefined - add icon for it.

import subprocess
import json
import cdx

def get_icon(chart_name):
    icon_mapping = {
        "gitea": "apps/gitea.png",
        ....
    }

    # Extract the base image name from the full image name
    chart_name = chart_name.split("/")[-1]

    matching_keys = [key for key in icon_mapping.keys() if key in chart_name]

    # Use the mapping, or default to "unknown.png" if not found
    return icon_mapping.get(matching_keys[0], None) if matching_keys else 'apps/helm.png'

def list():
    try:
        # Run the command and capture the output
        output = subprocess.run(['helm', 'list', '-o', 'json'], text=True, capture_output=True, check=True)
        ret = []
        # Parse the output to extract environment information
        releases = json.loads(output.stdout)
        for release in releases:
            release['icon'] = 'assets/icons/' + get_icon(release['name'])
            ret.append(release)

        return releases

    except subprocess.CalledProcessError as e:
        return [{'name': 'ERROR', 'icon': 'assets/icons/general/error.png', 'error': f"{e}\n{e.output}\n{e.stderr}"}]
    except FileNotFoundError as e:
        return [
            {
                'name': 'ERROR',
                'icon': 'assets/icons/general/error.png',
                'error': f"Can't find 'helm' in PATH, looks like its not installed, please install first"
            }
        ]

Lets list available Helm charts:

import subprocess
import json
import cdx

def get_icon(chart_name):
    icon_mapping = {
        "gitea": "apps/gitea.png", 
        # ...
    }

    # Extract the base image name from the full image name
    chart_name = chart_name.split("/")[-1]

    matching_keys = [key for key in icon_mapping.keys() if key in chart_name]

    # Use the mapping, or default to "unknown.png" if not found
    return icon_mapping.get(matching_keys[0], None) if matching_keys else 'apps/helm.png'

def list():
    try:
        # Run the command and capture the output
        output = subprocess.run(['helm', 'search', 'repo', '-o', 'json'], text=True, check=True, capture_output=True)
        ret = []
        # Parse the output to extract environment information
        repos = json.loads(output.stdout)
        for repo in repos:
            repo['icon'] = 'assets/icons/' + get_icon(repo['name'])
            ret.append(repo)

        return repos

    except subprocess.CalledProcessError as e:
        return [{'name': 'ERROR', 'icon': 'assets/icons/general/error.png', 'error': f"{e}\n{e.output}\n{e.stderr}"}]
    except FileNotFoundError as e:
        return [
            {
                'name': 'ERROR',
                'icon': 'assets/icons/general/error.png',
                'error': f"Can't find 'helm' in PATH, looks like its not installed, please install first"
            }
        ]

Lets create application action for repo adding:

🏆 Result

As a result we getting UI where you can see Helm releases installed in current context, Helm repos and Helm charts available in that repos:

Next actions - add install, release test actions for charts/releases and that's it - https://github.com/devopspass/plugins/tree/main/helm

helm_repo

Refresh info
Remove repo

helm_repo_chart

Install Chart
Open Chart Source

helm_release

Test Release
Remove Release

With just a few hundred of code lines you've created UI for Helm, when some companies even trying to sell solutions like that...

You can do the same for any API/CLI you're using daily.

🌟 Support Us, Contact Us

If you like this post, support us, download, try and give us feedback!

Give us a star 🌟 on GitHub or join our community on Slack.

DevOps Fun #2: how long command will sleep?

DevOps Pass AI — Mon, 22 Jul 2024 12:36:14 +0000

In previous aticle: If you're cool enough to watch video in console?

Recently saw article on one LinedIn channel explaining Linux shell pipes and again saw the same mistake - 99% engineers doesn’t know how pipes actually works!

Absolutely sure you will be “not absolutely sure” how long that command sleep:

sleep 5 | sleep 5 | sleep 5 | sleep 5

You may “suspect” that it will be 20 seconds, but it’s not true.

It’s surprising people, that ALL commands in pipe running simultaneously! And command will sleep 5 secs.

You can check it with ps -ef | grep slee[p] while sleep is running.

That’s the reason why you see intermediate result when grepping large file and not waiting for command to finish, in case of sequential run of commands in pipe.

Support Us, Contact Us

If you like this post, support us, download, try and give us feedback!

Give us a star 🌟 on GitHub or join our community on Slack.

DevOps fun #1: if you're cool enough to watch video in console?

DevOps Pass AI — Tue, 16 Jul 2024 08:53:38 +0000

Hey folks, its time for a good content!

If you're cool enough to watch your favourite anime in console?

Story

Ok, some story first. In 1997 (sic!) released first version of aa-lib, library which been allowing to display images In console for MS-DOS and Unix like systems. That day world has been changed!

Actually not, but I have a feeling that Wachowski bro... (you know) saw aa-lib demo called bb, before thay started with Matrix:

If you want to check it locally, on Ubuntu/Debian (if not use Docker):

sudo apt install bb
bb

For MacOS you can build it from sources - https://github.com/artyfarty/bb-osx or run Ubuntu in Docker and run command above, finally it all working in console!

AA-project home could be found here - https://aa-project.sourceforge.net/

How to play video in console?!

Back to Naruto...the best player in the world - mplayer, has not very well known that days feature - do video output in different output destinations, for example console:

What? CACA? AA, BB now CACA? Yes :D

libCACA is a next iteration of AA, because AA had no colors, group enthusiasts around 2010 used the same idea, but added console colors.

CACA will help you to output video like on the first image of article.

CACA_GEOMETRY=200x70  mplayer naruto-shippuden_Ep1.mp4 -vo caca

As small bonus, you can have a cool tool to make ASCII versions of your photos and images:

# Install CACA Utils
sudo apt install caca-utils

# decrease console font size with Ctrl+Mouse_Scroll
img2txt -W 300 favicon.512x512.png

Support Us, Contact Us

If you like this post, support us, download, try and give us feedback!

Give us a star 🌟 on GitHub or join our community on Slack.

DevOps Pass AI plugins

DevOps Pass AI — Tue, 09 Jul 2024 13:28:21 +0000

Right now DOP has bunch of plugins which simplifies your daily activities with DevOps stack and navigation in it.

You can search in your entire stack in one app:

And make necessary actions with this entities just in a few clicks:

Ansible

Install Ansible, Molecule and tools for various platforms via Docker without changing local file system.

AWS

Manage AWS profiles, refresh credentials, swith defaults profiles in a few clicks.
Import all your AWS profiles from LandingZone in a few clicks

Confluence

Search in Confluence like on your local! Fast and furious.

DevopsPass AI

Add new applications, documents, actions and many more.
Plugins used for DOP extinsion.

Docker

Manage Docker containers, images, Docker Compose stacks.
Install, delete, stop and so on.

Git

Some common integrations for many git providers, like GitHub, GitLab, etc.

GitHub

Search you git repos and open them in IDE in a few clicks
You can clone in into your workspace in one click

Golang

Nothing special, just install Golang on your local

Helm

Manage Helm releases, repos and charts
Install Helm and charts in a few clicks

Homebrew

Install HomeBrew...what else you need?!

Java

Just configure Maven source...
JABBA version manager is commming soon

Jenkins

Generate Jenkinsfiles for various cases
Install default Plugins set for your jenkins
List pipelines and its status, available for you
Fast navigation between jobs

Jira

Just search JIRA stories FAST!

k3s

Install K3s on your local and enjoy Kubernetes in Docker.

Kubernetes

Manage K8s/OSE contexts, switch between them
Manage namespaces
Use integration with your favourite K8s client

nodejs

Manage installed version of NodeJS
Select default

ollama

Install ollama locally
Manage locally available LLMs models

openshift

Install CLI client in a few cliks, all other functions are in Kubernetes plugin ;)

Python

Manage installed Python's versions via Conda
Manage environments, requirements.txt and pip

Sailpoint

Just keep list of available applications by hand,
List your current requests for Sailpoint access

Servers

Servers...list, SSH, RDP...what else?

systemd

Pretty often but annoying thing - generate SystemD units and timers for your apps.

Terraform

Manage version of Terraform, Terragrunt and OpenTofu

Vault

Keep list of secret engines and auth methods by hand
Easy navigation to them

VSCode

Install VSCode, just it.

WSL

Manage WSL installations, distros and runtimes.

How to run Ansible on Windows?

DevOps Pass AI — Wed, 12 Jun 2024 10:54:54 +0000

If you need consistant way to run Ansible on Mac, Linux and Windows you'll face a surprise! Ansible is not supported on Windows - https://docs.ansible.com/ansible/latest/os_guide/windows_faq.html#can-ansible-run-on-windows

Sad, right? But if you're doing tool like DevOps Pass, you need it.

So I was inspired by OpenTofu's tenv and created my own Ansible Version Manager, which allows you to run Ansible from Windows host, like on Mac or Linux.

Whats behind?

It uses Docker under the hood and allowing you to run Ansible tools from your local without installation of Python (on Mac/Linux) and on Windows.

It is running Docker container, pushing all your local environment variables related to Ansible, Molecule, SSH, AWS, Azure and GCP configuration.

🚀 Installation

MacOS / Linux

# Linux
curl -sL $(curl -s https://api.github.com/repos/devopspass/dop-avm/releases/latest | grep "https.*linux_amd64" | awk '{print $2}' | sed 's/"//g') | tar xzvf - dop-avm
# MacOS
curl -sL $(curl -s https://api.github.com/repos/devopspass/dop-avm/releases/latest | grep "https.*darwin_amd64" | awk '{print $2}' | sed 's/"//g') | tar xzvf - dop-avm

sudo mv dop-avm /usr/local/bin/
sudo sh -c "cd /usr/local/bin/ && dop-avm setup"

Windows

Download latest binary for Windows - https://github.com/devopspass/dop-avm/releases/

tar xzf dop-avm*.tar.gz
md %USERPROFILE%\bin
move dop-avm.exe %USERPROFILE%\bin\
setx PATH "%USERPROFILE%\bin;%PATH%"

cd %USERPROFILE%\bin\
dop-avm setup

DevOps Pass AI

In DOP you can add Ansible app and run action Install Ansible Version Manager, it will download and install dop-avm.

🤔 How it works?

dop-avm copying own binary with different names, which will be used later by user:

ansible
ansible-playbook
ansible-galaxy
ansible-vault
ansible-doc
ansible-config
ansible-console
ansible-inventory
ansible-adhoc
ansible-lint
molecule

When you're running any of this command, it will run Docker container devopspass/ansible:latest and binary inside (source Dockerfile in repo).

AVM will pass environment variables from host machine:

ANSIBLE_*
MOLECULE_*
GALAXY_*
AWS_*
GOOGLE_APPLICATION_CREDENTIALS

Plus volumes (if exist):

.ssh
.aws
.azure
.ansible

And services, like SSH-agent and Docker socket.
As a result you can run Ansible on Windows, MacOS and Linux via Docker without installation of Python and Ansible on your local, especially it's useful for Windows where it's not possible to run Ansible at all.

🐳 Use another Docker container

Probably you may have Docker container built in your organization, which is used in pipelines or recommended for local, you can use it by specifying DOP_AVM_IMAGE_NAME environment variable. Be sure that all necessary binaries, like molecule are inside, when you're running it.

Support Us, Contact Us

If you like this post, support us, download, try and give us feedback!

Give us a star 🌟 on GitHub or join our community on Slack.

What is DevOps…from business point of view?

DevOps Pass AI — Wed, 15 May 2024 13:30:24 +0000

If you see “DevOps” in our application name, that means we have to explain that term properly and we’ll do, but from a bit different angle. Lets see what DevOps is from business perspective.

Common name

First of all you should know, that DevOps have no clear definition. Yep.

Most often, DevOps is characterized by key principles: shared ownership, workflow automation, and rapid feedback.

And also, you’ve probably already heard “mantra”:

DevOps is not a role.

You can check great article on DevOpsKube website - What is Devops? What Does it really mean?

If you want summary:

DevOps is a culture or a philosophy aiming to bridge the gap between the development and operations teams to improve productivity and collaboration by automating infrastructure, code deployments, and continuous monitoring of applications.

Whats wrong with that defenition? It’s technical. And absolutely ignores business!

Pretty often engineers forgetting that “all that IT” is not about IT, it’s about business. We’re forgetting that all that wonderful stuff we’re doing, its not “because we can”, its because some business, which doing money, need it.

Business don’t care

It will confuse engineers if I would say that all that “automation”, “operations”, “development”, “monitoring” are not necessary from business perspective of view.

Business DONT CARE about it, because business is about selling tickets or making sport bets or selling financial services, etc. Its not about IT, at all.

You heard or faced such a situations when some process in a company is all manual and/or require a lot of people to do it or takes too long? You feel that process could be automated and if will save a lot of man-hours, but no one want’s or plan to automate it. And its OK.

Yep, you heard it right, its OK. Until its time to optimize costs or process slowness affects business needs, its ok. Business don’t value automation only because its automation.

What is DevOps from business perspective?

Lets try to make definition of DevOps in non-technical way as much as possible.

*DevOps *— is a service of development and adoption of tooling, which helps business to maintain software products on all stages of life-cycle in a self-service manner.

Sounds not much simpler, right? Lets explain each word here.

DevOps

After series of large personal data leaks, business realized, that they have to spent more resources on security hardening, as a result DevOps turned into DevSecOps… Last few years new areas appearing, where “Ops” should be connected with something: DataOps, FinOps etc…

So lets stop on DevOps term.

Automation is not a goal

Pretty often, when people trying to define what DevOps is, they falling into ‘enumeration explanation’: “you should do CI/CD, configuration management, use Cloud Copmuting, Ifrastructure as a Code, endless-list-here”.

As a result we have a mess in DevOps term definition… Not all projects need Cloud Computing (hello mainframes), when you’re saying “CD” you mean Continuous Delivery or Deployment (?), not all projects need infrastructure (hello Kubernetes ) and so on.

In the middle of all of that lists stays AUTOMATION. It requires special article about it, but yes folks, DevOps is not about automation, not each automation is done in DevOps way, not every DevOps activity could or worth to be automated.

DevOps is a service?!

Yes. DevOps engineers (Platform Team, DevX, SRE, etc.) are providing service, clients for that teams are entire company or software product team, which maintaining application. Or any other person who want to do Ops-related tasks, without knowledge in various “-Ops” areas.

What’s confusing me in many DevOps departments, that people thinking “they doing job”, folks NO!, you are providing SERVICE. You are serving to a business needs.

That small twist in mind blocking many good engineers from being successfull in their career. Soft skills in DevOps became more valuable in many cases than technical ones.

What does “development of tooling” mean?

Tooling its set of simple user interfaces, which helps people to make complicated things related to software life-cycle, without deep knowledge in SRE/DevOps practicies (or other technical areas). Yes, tooling is about “hiding complexity” by simple UI. “Business like pictures”, not because its stupid, but because technical details are not about business.

By other had this new tools creating new idioms for people who are not deep in technical details, but now can communicate and collaborate about it.

What does “adoption” mean?

Usually we have company-wide DevOps tools which requires adoption on an product team level, for example we have Splunk for the application logs search or AppDynamics for apps monitoring and troubleshooting, but not all applications/teams using it, such tools have to be adopted first, teams trained to use them.

What does “self-service” means?

Self-service means, that person or team using your tooling own process from end to end. DevOps team still responsible for that tool, means it should work as expected, but how that tool used they don’t care. Usage is absolute responsibility of product team, if tool works correctly, all consequences is your problem.

That twist in mind dramatically changes rules in collaboration between DevOps team and product team. Pretty often Ops teams suffering from situations when Ops have to do “all the magic” for developer and his issue will be fixed. Doesn’t matter if developer forgot to comment proxy usage or did not configured networking correctly for his server, if there is a tool which he can use to fix issue himself. “Sorry man, you own it!”

Summary

Huh, it was a long text. Short summary:

DevOps is a service
DevOps engineers (of all kinds) are serving to a business
Automation is not an end in itself
Self-service is a key for a healthy communications between Dev’s and Ops’es

What were not covered and require additional articles:

When you have to automate something, when not
How to determine if something done “in a DevOps way” or not
DevOps from a solution became a problem. Why and How?

Support Us, Contact Us

If you like this post, support us, download, try and give us feedback!

Give us a star 🌟 on GitHub or join our community on Slack.

How to import all AWS profiles from AWS LandingZone

DevOps Pass AI — Thu, 02 May 2024 13:38:58 +0000

If your organization using multiple AWS accounts, you probably know URL’s like https://d-XXX.awsapps.com/start/#/ where you could login with your identity provider and and get credentials for AWS account you need.

But what if that account list is a bit long…what if each of 268 accounts has more than one role?! What if you want to switch between them or sync credentials for multiple? DevOps Pass AI tries to simplify it a bit for you.

Load AWS profiles from SSO

First of all it allows you to import all that profiles to your ~/.aws/profiles file. If you wish it can sync all credentials for all profiles to ~/.aws/credentials

All you need is paste your AWS LandingZone URL and click green button. It will open new window where you have to pass auth to LZ (via your corporate MS AD account, Okta, whatever). Once you logged in, script will grab all your accounts/roles and create profiles in ~/.aws/profiles and sync AWS credentials if requested.

Actions for AWS Profiles

Ok, now you have profiles in place, you have list of them on “AWS Profiles” tab, what else?

You can AWS Console for that ACCOUNT-ROLE
Refresh AWS Credentials in ~/.aws/credentials file
Make current account default and put its credentials to “[default]” account in ~/.aws/credentials

Refresh credentials will do almost the same action like for all accounts, but using cached credentials, if credentials doesnt work it will ask you to login to AWS LandingZone again.

Thats it! I hope it was useful, leave a comment!

Support Us, Contact Us

Give us a start, we’re kitties 🐱🐯

Give us a star 🌟 on GitHub or join our community on Slack.

Workshop: make your first AI app in a few clicks with Python+Ollama+llama3

DevOps Pass AI — Tue, 23 Apr 2024 06:52:55 +0000

Hey there, you probably heard that we’re living in AI era, so good time to play with it on your local.

DevOps Pass AI has pretty good integration with Ollama, so you can install Ollama, pull necessary model (fresh llama3 in our case) and make some app to play with it.

Installing Ollama and pulling image

First of all you have to add Ollama app in DevOps Pass AI:

Now you have to click “Install Ollama” in right left panel of application actions:

After that you can refresh list of docs below and switch to “Ollama remote models”, find there “llama3”, most fresh LLAMA model from Facebook and pull it:

Pull will require a bit of time, as model size is about 4.7Gb:

Ok, now you’re ready to play with LLM models, lets write some app.

Start Ollama endpoint

Before you write anything, dont forget to start Ollama to make its API endpoint available:

ollama serve

Keep that console running while you’re playing with LLM’s

Create new Python env for ollama

Lets keep our system Python clean and create new Conda env, for that you have to add Python app in DevOps Pass AI and click “Create Conda environment” action in right actions pane:

Call environment “ollama” and specify following requirements.txt

ollama==0.1.8
notebook==7.1.3

After that you have to activate Conda Shell (action in right pane, should be run only once), now you’re ready for app development.

Simple app DevOps helper

Lets write really simple script, which will receive prompt and return code.

We can use it for example to generate simple BASH scripts or Ansible playbooks or whatever. Create file app1.py:

import ollama
import sys

print(sys.argv)
stream = ollama.chat(
    model='llama3',
    messages=[
      {'role': 'user', 'content': 'Youre DevOps assistant tool, return only code and explain only if asked'},
      {'role': 'user', 'content': " ".join(sys.argv)}
    ],
    stream=True,
)

for chunk in stream:
  print(chunk['message']['content'], end='', flush=True)

Now you can activate conda env and run script, lets say we want to create new plugin for DevOps Pass AI, which will list Pods in Kubernetes current context and namespace:

# Activate Conda env ollama
micromamba activate ollama

# Run app
python app1.py "generate python function list() to list Kubernetes pods, return list of dicts, with name, status and age."

Not ideal, but could be good starting point!

App two: Chat

Ok, first example is nice, but doesnt support history, its one-shot script.

Lets improve it, we’ll write script which will start chat by name (lets say JIRA story id or any other), will keep history for context and you can return to chat at any time.

# Init Ollama:
import ollama
import os, json, sys

# Message function
def ask(msg: str):
    file_path = sys.argv[1] + '.json'
    if os.path.exists(file_path):
        with open(file_path, 'r') as f:
            messages = json.load(f)
    else:
        messages = [{'role': 'user', 'content': 'Youre DevOps assistant tool, return only code and explain only if asked'}]

    messages.append({"role": "user", "content": msg})

    stream = ollama.chat(
      model='llama3',
      messages=messages,
      stream=True,
    )
    resp = ""
    for chunk in stream:
      print(chunk['message']['content'], end='', flush=True)
      resp = resp + chunk['message']['content']
    messages.append({"role": "assistant", "content": resp})
    with open(file_path, 'w') as f:
        json.dump(messages, f)

# Check for arguments
if len(sys.argv) == 1:
    print(f"ERROR: Specify chat name as argument: '{sys.argv[0]} CHAT_NAME'")
    exit(1)

# Main loop
while True:
    print(">>> ", end="", flush=True)
    req = sys.stdin.readline()
    ask(req)

Now you can start chat for a new story:

You can interrupt it with Ctrl+C and start chat again with the same argument “STORY-1”.

Support Us, Contact Us

Give us a start on GitHub or join our community on Slack

Managing versions of OpenTofu, Terraform and TerraGrunt with tenv and DOP

DevOps Pass AI — Mon, 22 Apr 2024 08:53:06 +0000

Hey folks, its DevOps Pass AI again, and today we’ll tell you about new plugin released last week, to manage Terraform, Terragrunt and Tofu versions on your local.

Why not TFEnv?

First of all lets explain why not tfenv — https://github.com/tfutils/tfenv, which many of you mb already uses for Terraform versions management.

There is multiple reasons: first of all it doesnt support Windows and means not universal for all platforms and second one it doesnt support OpenTofu, which is weird if not say stupid…

What is tenv

tenv (https://github.com/tofuutils/tenv) is a part of OpenTofu utils, which allowing you to manage installed versions of Terraform, Tofu and Terragrunt. Also it supports Linux, MacOS and Windows in the same way and has bunch of packages for different platforms and distros.

And we did pretty nice integartion with DevOps Pass AI which allowing you to install tenv and manage tools from UI.

After adding Terraform app in DOP, you can install “tenv” and manage tools, make some version active or remove it.

TBH, that’s it, we did it simple for you ;)

Support Us, Contact Us

Give us a start on GitHub or join our community on Slack

K3s: smallest K8s on your local

DevOps Pass AI — Tue, 16 Apr 2024 08:32:01 +0000

Have you ever needed to play with Kubernetes on your local?
For K8s apps development or to test your changes in Helm chart or just for fun?

Recommended way for local Kubernetes is Minikube, but peronally I dont like it and let me explain why K3s + autok3s + K3d is your choise.

Why not Minikube

Minikube, in the same way as many other local Kubernetes providers is annoying me in one way - it makes your laptop/PC dirty. Creates some sophisticated networking configurations, creates SystemD units, doing some unknown changes on local filesystem, install/update some tools you dont wanted to and so on.

My solution - use K3s with K3d provider, which will setup Kubernetes in your Docker and change nothing on your local. You can create multi node cluster and remove it next minute. And NOTHING will be changed on your local!

How to

If you prepere video

If not lets move step by step.

First of all you have to add Docker and K3s applications in DevOps Pass:

You can install Docker from DOP and move on with K3s.

K3s app providing following actions:

Install autoK3s - start autok3s tool in your Docker
Setup k3d cluster - Setup local Kubernetes in Docker
Open autoK3s in Browser - open autok3s in your browser

autok3s is a tool, developed by Rancher, which allowing you to install K3s anywhere, currently supported AWS, Docker, GCP, etc. - https://github.com/cnrancher/autok3s

k3d is one of providers k3s can use, it setups Kubernetes cluster in your local Docker and you even can specify amount of nodes you want to run.

So install autok3s first, then click "Setup k3d cluster", wait until it appear "Running" in clusters list:

And done! You have Kubernetes cluster installed locally.

You can swithc to Kubernetes app and check if context exists and installed namespaces list:

MacOS, as usual...

As usual for MacOS there is one exception... Alike Windows or Linux you can not access Docker containers from local network, you have to install additional Brew package - https://github.com/chipmk/docker-mac-net-connect

It will allow you to connect your containers and K3s services/ingress'es from host machine:

# Install via Homebrew
$ brew install chipmk/tap/docker-mac-net-connect

# Run the service and register it to launch at boot
$ sudo brew services start chipmk/tap/docker-mac-net-connect

Thats it, now you can forget about it.
If you're curious what is it: its Brew service written in Go, which refreshing MacOS routing table, to allow you access Docker networks from host machine. More details you can find in repo readme.

Install Jenkins via Helm

Go to Helm app in DOP, click "Addd Helm repo" action in right top and add Bitnami Helm repo - https://charts.bitnami.com/bitnami

Then on "Helm Repo charts" tab find Jenkins chart:

Now "Actions -> Install Chaart", it will require some values, value should be in JSON format, i.e. if its string "..." if number 8083 if boolean true:

service.ports.https=8443
service.ports.http=8083
jenkinsUser="user"
jenkinsPassword="12345678"
agent.enabled=true

Click "Install" and wait for Jenkins pod to be ready (you can open "default" namespace in K9s in Kubernetes app or just wait for a few mins):

Now copy this to your console to get Jenkins URL:

export SERVICE_IP=$(kubectl get svc --namespace default jenkins --template "{{ range (index .status.loadBalancer.ingress 0) }}{{ . }}{{ end }}")
echo "Jenkins URL: http://$SERVICE_IP:8083/"

Removing K3s cluster

At the moment you dont need K8s cluster on your local, you can drop it. Navigate to K3s application, find cluster in the list at the bottom and find "Destroy" action.

Thats it! Not even a hint of K8s on your local, you can forget about it till you'll need it again.

Summary

I hope that workshop was interresting for you we've created K8s cluster on your local, installed some Helm charts and dropped it.

No more broken semi-configured networking rules or services, no more waste on your local.

If you likes this article, give us 👍 and start on GitHub 🤩 - https://github.com/devopspass/devopspass/

Pushing your app in Docker in a few clicks

DevOps Pass AI — Tue, 09 Apr 2024 16:44:20 +0000

Video

Step by Step

https://www.devopspass-ai.com/ - Video version in the end of article.

Imagine you're developing some application and its good time to make it public or at least test on some more production-similar environment.

Or for a local development you need some infra, like Redis, MongoDB, MySQL or Postgres. Nice candidate for that is to spin it up with Docker and Docker compose.

Add Docker app

To start with Docker, add Docker application you switch to Docker in left side.

At the bottom you can see three tabs:

Docker containers, to manage running/stopped containers
Docker images, to manage local images
Docker compose, to see running Docker Compose stacks

In the Right you can see Docker app actions:

Install Docker, to install Docker on MacOS, Linux, Windows
Generate Dockerfile, to generate Dockerfile for application
Start Docker container, just an action to start Docker container from DOP UI
Generate compose.yml, action to generate Docker Compose config file

Generating Dockerfile

Action Generate Dockerfile is used to generate Dockerfile for your application, we support various languages, including:

Python
Python Django
Python Flask
Golang
Java Spring Boot
Java Quarkus
NodeJS
NodeJS node-gyp
NodeJS NestJS
React
React pnpm
ASP.NET Core
ASP.NET Core Alpine
Ruby RoR
Ruby RoR with assets
Rust root
Rust non-priviledged
Dart
R Studio MSSQL
R Studio MySQL

Docker Compose

Once you have Dockerfile for app, you can create Docker Compose config:

Action asking if you need infra, like nginx, MySQL, MongoDB, PostgreSQL, Redis. For all backeds it will create volume, so you can be sure that your data will be in place after restart of laptop/PC.

After that you can spin stack up and play with it...

Swiss Army Knife for Kubernetes

DevOps Pass AI — Wed, 03 Apr 2024 09:53:10 +0000

Can you install all necessary Kubernetes tools in one click? How you switching between contexts and namespaces?
How you'll find README and value of Helm chart?
How would you test if Helm release is healthy?

And main question:

How you will install Kubernetes locally?!

DevOps Pass AI helping you to answer all that questions:

Install kubectl, k9s, k3s, Helm in one click
Manage namespaces and contexts in OpenShift and K8s distros
Install smallest Kubernetes on your local in one click (and destroy if you dont need it anymore)
Manage Helm releases, charts and repos with only your mouse!

Video

Intro to Kubernetes toolset in video tutorial