DEV Community: Karl N. Redman

A message from the FUTURE: 2016

Karl N. Redman — Mon, 28 Oct 2019 09:51:00 +0000

TL;DR:
Sally is the computer.

Randomize your XFCE / LightDM Login Screen Wallpaper

Karl N. Redman — Tue, 25 Jun 2019 07:18:49 +0000

HowTo: randomize LightDM / XFCE v4.x login screen wallpaper images.

For fans of the lightweight and highly configurable linux desktop environment XFCE, with it's default Greeter managed by LightDM, it's a frustrating fact of reality that there is no built in way to randomize the login screen image. This simple perl script and directory layout will provide that service using a cron job. The process will copy a random image file from a directory that contains a pool of images into a 'publish' directory; as background.jpg. XFCE then uses that image as the background for the login screen once configured.

Overview:

The basic premise of this procedure will be:

populate some directory with images you want to use for login backgrounds.
setup a cron job to run a script to copy a random image from the source directory into a publish directory
XFCE will use that image as the login screen background image.

TL;DR:

An example project for configuring randomized login screens for XFC can be found here

Single vs Multi User system and encrypted `home` directories:

Single-user systems that do not use encrypted home directories you can use a directory within your home directory (i.e. /home/karl/Repositories/LoginWallpaper-XFCE-Variety).
Multi-user systems that use a program like variety will need to use a user-group level directory location. (i.e. /opt/LoginWallpaper-XFCE-Variety)
Single-user and Multi-user system that use encrypted home directories will need to use a user-group level directory location (i.e. /opt/LoginWallpaper-XFCE-Variety)

Generic setup

Examples will assume a multi-user system with encrypted home directories.

clone the example project to an appropriate directory:

# a global directory space
cd /opt
#
# clone the project
sudo git clone git@github.com:karlredman/LoginWallpaper-XFCE-Variety.git
#
# set permissions on the project directory
sudo chown -R root:users LoginWallpaper-XFCE-Variety

add images to the image directory `Variety`

Add whatever jpg images you deem worthy / safe to be on a login screen. For this example the images you add will go into /opt/LoginWallpaper-XFCE-Variety/Variety/.
Do verify that your images belong to users group -change as necessary

run the project script to seed the `publish` directory

(i.e. /opt/LoginWallpaper-XFCE-Variety/Greeter/background.jpg)

seed the file

cp $(/opt/LoginWallpaper-XFCE-Variety/mbin/print_random_file.pl /opt/LoginWallpaper-XFCE-Variety/Variety) /opt/LoginWallpaper-XFCE-Variety/Greeter/background.jpg

verify the file exists -troubleshoot...

ls /opt/LoginWallpaper-XFCE-Variety/background.jpb

Configure LightDM Greeter via the GUI interface

start the greeter config app using one of these methods:
- via the GUI menu for LightDM GTK Greeter settings
- name may vary based on linux desktop distribution
- via command line

  sudo lightdm-gtk-greeter-settings

Select Background->Image
Set Image to <path>/background.jpg
- i.e. /opt/LoginWallpaper-XFCE-Variety/Greeter/background.jpg
- the name background.jpg is required
Save the setting and exit the app

test and debug

If you are having issues try to run the command manually.

add cron job

open cron in your default editor:

cron -e

add the job (example)

# change to wallpaper every 10 min
*/10 * * * * cp $(/opt/LoginWallpaper-XFCE-Variety/mbin/print_random_file.pl /opt/LoginWallpaper-XFCE-Variety/Variety) /opt/LoginWallpaper-XFCE-Variety/Greeter/background.jpg

Using `Variety` as the image source:

Variety is a wallpaper background manager that is available for several linux distributions and is compatible with several desktop managers. One of the options variety offers is the ability to manage your login greeter image.

open variety preferences and navigate to customize tab
select Login Screen Support checkbox
specify the image directory (i.e. /opt/LoginWallpaper-XFCE-Variety/Variety)

I cut my hair for an interview -oh! No!

Karl N. Redman — Tue, 18 Jun 2019 12:27:28 +0000

In 2001 my hair had grown down to the middle of my back.After 911 I was laid off from Veriscape (they lost $12 million in funding -but you are welcome for my automatons for your robocalls). I had insanely long, flowing hair at that time -it took me 4 years to grow it that long.

After 911 the software engineering market go really tight, very conservative -in dress, in manner, 'brain bench' tests, etc. BTW, I ranked in the top 3% of all c++ 'brain bench' tests bake then -heh, the concept is laughable these days. But it was a big deal back then. So I decided to start consulting!

It wasn't hard to get c++ interviews after 911 for me. I had the 'street cred' in the Chicago financial community as a 24/7,work hard, play hard, all out,no bullshit, developer. I felt that I wanted a break from the financial developer world so I sought out other venues.

My first venture into 'massive infrastructure' companies was with Motorola. It was a job I had never worked. It was developing 3g phones (not apps -phones) for 3g. For Motorola!. The job payed $55/hour with a 6mo. contract -to hire after that. Wow did I want to work for those guys.

The big issue was that I had a lot of hair. Like 3 feet+ of it. 2 days before the Motorola interview, I was sure I had the job...you can see, I'm sure, where this is going.. I got my hair cut. I got the standard business style cut thing for the times. The person cutting my hair literally cried the entire time. It was kind of traumatic for both of us -although not so much over the hair from my point of view...

And I didn't get the job. Not at all. Was totally turned down -hairless. But I looked damn good for the next job interview! -and scored that one.

So, today I cut my hair. I'd been growing it these past 3 years of my 'sabbatical from society (try to explain that in an interview!)'. And it seems somewhat important to me to mention that I really, really, don't like having long hair in the first place. I don't know how other people even tolerate it... because I just always wanted to cut it, but then I'd get busy and time would go buy and, more hair! And in some ways it defined me. So I miss it. And, I'm a huge anti-webber person but... Hair!

Anyway, I shaved the long hair last night so that I 'look' more presentable in upcoming interviews. Nobody cried this time but, funny this, I miss it!

Be a Python Jinja Master: Automated Front Matter Editing

Karl N. Redman — Tue, 11 Jun 2019 04:13:35 +0000

TL;DR

I'm pleased to announce the initial release of EditFrontMatter: A python module for editing markdown front matter using Jinja2 Templates.

See: EditFrontMatter Project Homepage
Several examples are included in the documentation

The Pitch:

Imagine that you could retain all of your various article and documentation content in markdown. And then cater that content to whichever static site generator¹, and theme thereof, without manually editing each and every file. With EditFrontMatter you can write simple scripts, leveraging the power of Jinja2, to add / edit / delete the front matter of your markdown files easily.

A demonstration:

This is a simple example that you will also find in the documentation as Example 1. Here, we to use a Jinja2 Template to edit the front matter of a markdown file. All fields which are not included in the template will be ignored during editing. Furthermore, while white-space and comments will be eliminated², the order of the original front matter is preserved³ -new fields will be appended.

Original markdown file with yaml front matter (example1.md)

---
title: "EditFrontMatter Class Example 1"
description: "Edit some fields in this front matter"
catagories: [programming, python, markdown]

deleteme: this will be deleted

tags: [front matter, administration, testing]

# comments and spaces will be eliminated (see docs)

author: "Karl N. Redman"
creatordisplayname: "Karl N. Redman"
creatoremail: "karl.redman@example.com"
date: 2019-05-23T17:43:45-05:00
lastmodifierdisplayname: "Karl N. Redman"
lastmodifieremail: "karl.redman@gmail.com"
lastmod: 2019-05-23T17:43:45-05:00
toc: false
type: "page"
hasMath: false
draft: false
weight: 5
---

# EditFontMatter Class Example 1

Edit several fields of front matter.

## Fields affected in this example:

* toc
  * note: uses local template variable
  * pre: false
  * post: true
* draft:
  * note: uses jinja2 filter (callback)
  * pre: false
  * post: true
* hasMath
  * note: uses program variable
  * pre: true
  * post: false
* stuff:
  * note: uses program variable to create field
  * pre: did not exist
  * post: (list) ['one', 'two', 'three']
* deleteme:
  * note: removed from final result
  * pre: this will be deleted
  * post: N/A

Jinja2 template that will update the front matter data of the source markdown file (template1.j2). Note that the toc field processing is overridden by the template. All local set Jinja variables take precedence over other processing methods.

{% set toc = "true" %}

toc: {{ toc }}
draft: {{ false | canPublish }}
hasMath: {{ hasMath }}
stuff: {{ addedVariable }}

A python program to edit the markdown file with the Jinja2 template (example1.py:

This program/script uses a Jinja2 template filter and a callback function, specified by a call to EditFrontMatter.add_JinjaFilter() to provide data for the draft field. In addition, specific key, value variable pairs are provided when EditFrontMatter.run() is called. The values from run() will be used to add or edit the key, value pairs. Lastly the instance (list) variable EditFrontMatter().keys_toDelete specifies which keys to delete in the front matter.

#!/usr/bin/env python3
# -*- coding: utf-8 -*-


from editfrontmatter import EditFrontMatter
import os


def canPublish_func(val):
    # do some processing....
    return True


def main():

    # generic path - overridden by env var `TEST_DATA_DIR`
    DATA_PATH = "../data/"

    if "TEST_DATA_DIR" in os.environ:
        DATA_PATH = os.path.abspath(os.environ.get("TEST_DATA_DIR")) + "/"

    # set path to input file
    file_path = os.path.abspath(DATA_PATH + "example1.md")

    # initialize `template_str` with template file content
    template_str = ''.join(open(os.path.abspath(DATA_PATH + "template1.j2"), "r").readlines())
    print(template_str)

    # instantiate the processor
    proc = EditFrontMatter(file_path=file_path, template_str=template_str)

    # set fields to delete from yaml
    proc.keys_toDelete = ['deleteme']

    # add a filter and callback function
    proc.add_JinjaFilter('canPublish', canPublish_func)

    # populate variables and run processor
    proc.run({'toc': 'no effect', 'hasMath': "false",
              'addedVariable': ['one', 'two', 'three']})

    # dump file
    print(proc.dumpFileData())


if __name__ == '__main__':
    main()

Final Output:

---
title: "EditFrontMatter Class Example 1"
description: "Edit some fields in this front matter"
catagories:
- programming
- python
- markdown
tags:
- front matter
- administration
- testing
author: Karl N. Redman
creatordisplayname: Karl N. Redman
creatoremail: karl.redman@example.com
date: 2019-05-23 22:43:45
lastmodifierdisplayname: Karl N. Redman
lastmodifieremail: karl.redman@gmail.com
lastmod: 2019-05-23 22:43:45
toc: true
type: page
hasMath: false
draft: true
weight: 5
stuff:
- one
- two
- three
---

# EditFontMatter Class Example 1

Edit several fields of front matter.

## Fields affected in this example:

* toc
  * note: uses local template variable
  * pre: false
  * post: true
* draft:
  * note: uses jinja2 filter (callback)
  * pre: false
  * post: true
* hasMath
  * note: uses program variable
  * pre: true
  * post: false
* stuff:
  * note: uses program variable to create field
  * pre: did not exist
  * post: (list) ['one', 'two', 'three']
* deleteme:
  * note: removed from final result
  * pre: this will be deleted
  * post: N/A

My Personal Use Case:

Allow me paint a picture of my workflow: I maintain a group of repositories of exclusively markdown content. This content is split up between topics ranging form a personal journal to various projects and articles in various states of completion. The number of documents I've written in the last several years is now over 1000. In order to facilitate my workflow, I want all the content to be accessible in one place while having the ability to preview these documents and, ultimately, publish them using various methods. To that end, all of my markdown files do contain a minimal about of front matter. As I publish these content files I needed some way to batch process the content files and modify their metadata.

I use vimwiki for editing all of my markdown content. Vimwiki allows me to easily index various topics, keep topic related diaries, and otherwise edit content using Vim: my 'go to' editor of choice. All of my various projects and topic related content are tracked in individual git repositories and imported to a 'working tree' as submodules. This makes adding, archiving, and modifying topics very simple.

My personal development ecosystem uses a static site generated by Hugo to host content (my knowledge base - I use docdock theme by the way). This static site is plugged into my ecosystem's build system (drone). When I edit any document within the working three, and push that content to my git repository, my Gitea instance picks up the change and triggers a drone build.

In turn, drone builds my searchable knowledge base and publishes it to be hosted by my ecosystem's web server. From a git push to the content showing up through my web services, the total time is < 1 minute. And that includes the processing EditFrontMatter does to set the front matter fields needed for any number of specific themes and other purposes (i.e. like changing the 'edit this page' link in front matter to point to the corresponding content's Gitea edit page of the document itself: online editing of the content is a built in bonus!).

With the help of EditFrontmatter, I can easily change the 'edit this page' URL pragmatically when I decide to move, for instance, a project from my personal knowledge base to a publishing entity (such as Github). I no longer worry about having to manually edit a bunch of content files in order to change the Hugo theme, or change the static site generator being used.

The scripts that I have in place use a set of Jinja Templates that perform these actions for me. And it takes mere seconds to produce accurate results instead minutes of error prone tedious editing. While this all might seem complicated, it really isn't. Most of, if not all, of these steps are things we do as developers anyway. This is my attempt at streamlining the process since it's a well defined workflow.

Conclusion:

In previous articles I've mentioned that many of the, said, articles are interrelated. I've also eluded to the fact that I am working on a much larger, and more complicated, set of documentation for a project that I'm working on -building a complete development ecosystem you can use anywhere. While EditFrontmatter is a small class/module (less than 150 lines of code), the documentation is 1000's of lines of text. And that is the nature of documentation -the product is always much shorter than the documentation relative to lines of code.

The power of markdown is that I can continue writing the content of my project's documentation without having to necessarily worry about how I will eventually publish it -or where I will publish it. In fact, because EditFrontMatter provides the ability to easily manipulate the metadata of the content I create, I can review changes in near real-time in a variety of publishing entities. I'm not 'vendor locked' and neither is my project. Furthermore, depending on the build system I use, I can easily include content from one document to the next within the markdown itself -promoting the 'write it once' concept.

Lastly, if you are curious about how I generated the Sphinx Read The Docs Theme documentation to be hosted via Github Pages check out the About Theme section on the EditFrontMatter home page. And if you would like to read about why I'm using reST and python to to build static pages for Hugo, check out the FAQ.

Now go be a Python Jinja Master!

Popular site generators that support markdown front matter: Directly - Gatsby, Hexo, Hugo, Jeykyll. Indirectly - Netlify CMS, Next.js, Nuxt. And Many more ... ↩
Future versions of this module may provide extra functionality, like preserving comments, via ruamel.yaml; depending upon frequency of requests for such features. ↩
EditFrontMatter uses oyaml to preserve the order front matter. (see footnote 2) ↩

Private Network + Traefik + Keycloak + Gatekeeper Overview (diagram)

Karl N. Redman — Sun, 28 Apr 2019 10:00:14 +0000

A high level network overview of Traefik, Keycloak, and Gatekeeper working together

This is yet another artifact [although ugly] from a project I'm working on. This diagram depicts a basic lab infrastructure with Traefik, Keyclaok, and Keycloak Gatekeeper working together behind a local DNS (dnsmasq). Details for how all of this actually fits together are forthcomming. For now, hopefully, it will be helpful for those who are wondering what the layout looks like when configuring these various components to work together. Note the whoami that is a service illustrated in a previous post.

Original PlantUML source + Image File: here
Original source edited with:
- vim iamcco/markdown-preview.nvim: markdown preview plugin for (neo)vim
- VSCode PlantUML - Visual Studio extension

Keycloak v5 + Gatekeeper v5: Flowcharts - Easily Create and Restrict an Isolated (IODC) Client Service by Group-Role

Karl N. Redman — Sun, 28 Apr 2019 08:31:12 +0000

This is another artifact from an upcoming series of articles that I'm writing for creating a home/laptop development laboratory ecosystem. I'm sharing these artifacts now because the series that I'm writing won't be released for at least a month (It's a huge undertaking -for me).

These flow charts demonstrate the simplest form (IMHO) of creating Keycloak client services behind an authentication/authorization proxy (Keycloak Gatekeeper). I imagine that if you are reading this article you are already frustrated with just creating a simple client in Keycloak and/or trying to figure out how to restrict authorization so that only specific users of a group are allowed to log in via Keycloak. Hopefully this is enough of a clue to help people until I am able to publish the article series as a whole.

Artifact Specifics:

Keycloak v5
Keycloak Gatekeeper v5
Original PlantUML source + Image Files: here
Original source edited with:
- vim iamcco/markdown-preview.nvim: markdown preview plugin for (neo)vim
- VSCode PlantUML - Visual Studio extension
NOTE: While current (v6.0.0) Keycloak documentation specifies that groups can be used directly in Keycloak Gatekeeper only role based authorization is available in V5.0.0.

Flowchart Caveates:

whoami: The example client service being created
auth_user: A User created to log in the whoami service
whoami_group: A User Group mapped to a client role for authorization by Keycloak Gatekeeper
https://whoami.example.com: The URL of our client service

Relevant Documentation References:

Securing Applications and Services Guide

The Keycloak (v.5.0.0) UI mind map you didn't ask for

Karl N. Redman — Wed, 24 Apr 2019 14:18:30 +0000

I'm working on a future set of articles for creating a home/laptop lab and needed a reference diagram. I figured that I might as well share. The diagram is meant to be a mid-level view to show where everything is located in the UI.

Keycloak version: v5.0.0
Required fields: denoted by * suffex
Original PlantUML source + Image files here

Navigate your vscode like it's 1999 (the vim way)

Karl N. Redman — Fri, 15 Mar 2019 13:51:53 +0000

If you are a vim person it can be frustraiting to work in vscode without vim keystrokes. The fantastic plugin vscodevim only goes so far. If you, like me, are used to M-h,M-j,M-k,M-l (where M == 'Alt key') and vim-like tab movements things get freaking nuts. Note that these keys in default vim are the same as pressing escape [I HATE ever having to hit the esc key -ever].

This configuration demonstrates how to switch to vscode easily from vim. In addition a method for you to switch tabs in a vim-like way.Lastly, these settings allow you to navigate vscode with bookmarks in a way you would use marks in regular vim.

Wait there's more! Read though the code (below) and find many vim-like jems that will make vscode feel a bit closer to home.

Tested with:
- vscode v1.32.1
- vim 8.1 how to compile
- MX Linux 18.1
- Debian Stretch 9.8

Vim configuration (at your discretion)

Open you current vim document in vcode

Add this following line to your ~/.vimrc
This (default) means \ov will open the current file in vscode from vim

nnoremap <leader>ov :exe ':silent !code %'<CR>:redraw!<CR>

Use `ctrl+j` and `ctrl+k` to change tabs in vim (I prefer up/down -adjust as needed)

" switch tabs (same as gt & gT)
nnoremap <C-j> :tabprevious<CR>
nnoremap <C-k> :tabnext<CR>
"
" (bonus) move tabls right or left
map <C-h> :execute "tabmove" tabpagenr() - 2 <CR>
map <C-l> :execute "tabmove" tabpagenr() + 1 <CR>

VS Code changes

This get's really complicated to explain. I'm going to list the functions with examples here and then the code -edit as needed. I had to configure keybindings.json and settings.json -I haven't tried combining the configurations.

install plugin Vim - Visual Studio Marketplace
install plugin Numbered Bookmarks - Visual Studio Marketplace
config: ~/.config/Code/User/keybindings.json
config: ~/.config/Code/User/settings.json

features from `keybindings.json` (+noted settings.json)

focus active editor (vim windows)
- i.e. (similar to vim C-w shortcuts)

    {
        "key": "ctrl+w j",
        "command": "workbench.action.focusBelowGroup",
    },

focus window groups
- (similar to tmux panes)

{
    "key": "ctrl+w down",
    "command": "workbench.action.moveActiveEditorGroupDown",
},

do nothing for <ALT>+h,j,k,l
- be sure to disable menu keys in vscode settings: Uncheck "Settings->Window->Enable Menubar Mneumonics"
- Same as default vim behavior

   {
    "key": "alt+h",
    "command": "extension.vim_escape",
    "when": "editorTextFocus && vim.active && vim.mode == 'Insert'"
},

create mark
- see file (/.conf/Code/User/settings.json)

     "vim.normalModeKeyBindingsNonRecursive": [
        {
            "before": ["m","a"],
            "commands": [
                "numberedBookmarks.toggleBookmark1"
            ]
        }
    ]

jump to mark
- i.e. jump to mark a

{
    "key": "' a",
    "command": "numberedBookmarks.jumpToBookmark1",
    "when": "editorTextFocus && vim.active && vim.mode == 'Normal'"
},

The files / configs

At this point I think you get the idea. There are several other vim-like settins I've added into the config files but, like Vim, you'll have to read the code to find the gems.

~/.config/Code/User/keybindings.json

{
    "vim.easymotion": true,
    "vim.sneak": true,
    "vim.incsearch": true,
    "vim.useSystemClipboard": true,
    "vim.useCtrlKeys": true,
    "vim.hlsearch": true,
    "vim.leader": "\\",
    "vim.handleKeys": {
        "<C-a>": false,
        "<C-f>": false
    },
    "vim.normalModeKeyBindingsNonRecursive": [
        {
            "before": ["<C-n>"],
            "commands": [":nohl"]
        },
        {
            "before": ["m","a"],
            "commands": [
                "numberedBookmarks.toggleBookmark1"
            ]
        },
        {
            "before": ["m","b"],
            "commands": [
                "numberedBookmarks.toggleBookmark2"
            ]
        },
        {
            "before": ["m","c"],
            "commands": [
                "numberedBookmarks.toggleBookmark3"
            ]
        },
        {
            "before": ["m","d"],
            "commands": [
                "numberedBookmarks.toggleBookmark4"
            ]
        },
        {
            "before": ["m","e"],
            "commands": [
                "numberedBookmarks.toggleBookmark5"
            ]
        },
        {
            "before": ["m","f"],
            "commands": [
                "numberedBookmarks.toggleBookmark6"
            ]
        },
        {
            "before": ["m","g"],
            "commands": [
                "numberedBookmarks.toggleBookmark7"
            ]
        },
        {
            "before": ["m","h"],
            "commands": [
                "numberedBookmarks.toggleBookmark8"
            ]
        },
        {
            "before": ["m","i"],
            "commands": [
                "numberedBookmarks.toggleBookmark9"
            ]
        },
        {
            "before": ["m","j"],
            "commands": [
                "numberedBookmarks.toggleBookmark0"
            ]
        },
    ],
    "breadcrumbs.enabled": true,
    "workbench.activityBar.visible": true,
    "workbench.statusBar.visible": true,
    "workbench.sideBar.location": "left",
    "zenMode.hideStatusBar": false,
    "zenMode.hideTabs": false,
    "window.enableMenuBarMnemonics": false,
    "window.menuBarVisibility": "default",
    "window.zoomLevel": 1,
    "telemetry.enableTelemetry": false,
    "telemetry.enableCrashReporter": false
}

~/.config/Code/User/settings.json

// Place your key bindings in this file to override the defaults
[
    {
        "key": "ctrl+w j",
        "command": "workbench.action.focusBelowGroup",
    },
    {
        "key": "ctrl+w k",
        "command": "workbench.action.focusAboveGroup",
    },
    {
        "key": "ctrl+w h",
        "command": "workbench.action.focusLeftGroup",
    },
    {
        "key": "ctrl+w l",
        "command": "workbench.action.focusRightGroup",
    },
    {
        "key": "ctrl+w down",
        "command": "workbench.action.moveActiveEditorGroupDown",
    },
    {
        "key": "ctrl+w up",
        "command": "workbench.action.moveActiveEditorGroupUp",
    },
    {
        "key": "ctrl+w left",
        "command": "workbench.action.moveActiveEditorGroupLeft",
    },
    {
        "key": "ctrl+w right",
        "command": "workbench.action.moveActiveEditorGroupRight",
    },
    {
        "key": "alt+h",
        "command": "extension.vim_escape",
        "when": "editorTextFocus && vim.active && vim.mode == 'Insert'"
    },
    {
        "key": "alt+j",
        "command": "extension.vim_escape",
        "when": "editorTextFocus && vim.active && vim.mode == 'Insert'"
    },
    {
        "key": "alt+k",
        "command": "extension.vim_escape",
        "when": "editorTextFocus && vim.active && vim.mode == 'Insert'"
    },
    {
        "key": "alt+l",
        "command": "extension.vim_escape",
        "when": "editorTextFocus && vim.active && vim.mode == 'Insert'"
    },
    {
        "key": "' a",
        "command": "numberedBookmarks.jumpToBookmark1",
        "when": "editorTextFocus && vim.active && vim.mode == 'Normal'"
    },
    {
        "key": "' b",
        "command": "numberedBookmarks.jumpToBookmark2",
        "when": "editorTextFocus && vim.active && vim.mode == 'Normal'"
    },
    {
        "key": "' c",
        "command": "numberedBookmarks.jumpToBookmark3",
        "when": "editorTextFocus && vim.active && vim.mode == 'Normal'"
    },
    {
        "key": "' d",
        "command": "numberedBookmarks.jumpToBookmark4",
        "when": "editorTextFocus && vim.active && vim.mode == 'Normal'"
    },
    {
        "key": "' e",
        "command": "numberedBookmarks.jumpToBookmark5",
        "when": "editorTextFocus && vim.active && vim.mode == 'Normal'"
    },
    {
        "key": "' f",
        "command": "numberedBookmarks.jumpToBookmark6",
        "when": "editorTextFocus && vim.active && vim.mode == 'Normal'"
    },
    {
        "key": "' g",
        "command": "numberedBookmarks.jumpToBookmark7",
        "when": "editorTextFocus && vim.active && vim.mode == 'Normal'"
    },
    {
        "key": "' h",
        "command": "numberedBookmarks.jumpToBookmark8",
        "when": "editorTextFocus && vim.active && vim.mode == 'Normal'"
    },
    {
        "key": "' i",
        "command": "numberedBookmarks.jumpToBookmark9",
        "when": "editorTextFocus && vim.active && vim.mode == 'Normal'"
    },
    {
        "key": "' j",
        "command": "numberedBookmarks.jumpToBookmark0",
        "when": "editorTextFocus && vim.active && vim.mode == 'Normal'"
    },
    {
        "key": "ctrl+1",
        "command": "-numberedBookmarks.jumpToBookmark1",
        "when": "editorTextFocus"
    },
    {
        "key": "ctrl+2",
        "command": "-numberedBookmarks.jumpToBookmark2",
        "when": "editorTextFocus"
    },
    {
        "key": "ctrl+3",
        "command": "-numberedBookmarks.jumpToBookmark3",
        "when": "editorTextFocus"
    },
    {
        "key": "ctrl+4",
        "command": "-numberedBookmarks.jumpToBookmark4",
        "when": "editorTextFocus"
    },
    {
        "key": "ctrl+5",
        "command": "-numberedBookmarks.jumpToBookmark5",
        "when": "editorTextFocus"
    },
    {
        "key": "ctrl+6",
        "command": "-numberedBookmarks.jumpToBookmark6",
        "when": "editorTextFocus"
    },
    {
        "key": "ctrl+7",
        "command": "-numberedBookmarks.jumpToBookmark7",
        "when": "editorTextFocus"
    },
    {
        "key": "ctrl+8",
        "command": "-numberedBookmarks.jumpToBookmark8",
        "when": "editorTextFocus"
    },
    {
        "key": "ctrl+9",
        "command": "-numberedBookmarks.jumpToBookmark9",
        "when": "editorTextFocus"
    },
    {
        "key": "ctrl+shift+tab",
        "command": "workbench.action.focusNextGroup"
    },
    {
        "key": "ctrl+j",
        "command": "workbench.action.previousEditorInGroup"
    },
    {
        "key": "ctrl+pagedown",
        "command": "workbench.action.nextEditorInGroup"
    },
    {
        "key": "ctrl+w shift+-",
        "command": "workbench.action.minimizeOtherEditors"
    },
    {
        "key": "ctrl+w =",
        "command": "workbench.action.evenEditorWidths"
    },
    {
        "key": "ctrl+w v",
        "command": "workbench.action.splitEditorRight"
    },
    {
        "key": "ctrl+w s",
        "command": "workbench.action.splitEditorDown"
    },
    {
        "key": "ctrl+w c",
        "command": "workbench.action.closeActiveEditor"
    },
    {
        "key": "ctrl+w shift+k",
        "command": "workbench.action.moveEditorToAboveGroup"
    },
    {
        "key": "ctrl+w shift+j",
        "command": "workbench.action.moveEditorToBelowGroup"
    },
    {
        "key": "ctrl+w shift+h",
        "command": "workbench.action.moveEditorToLeftGroup"
    },
    {
        "key": "ctrl+w shift+l",
        "command": "workbench.action.moveEditorToRightGroup"
    },
    {
        "key": "ctrl+b",
        "command": "-extension.vim_ctrl+b",
        "when": "editorTextFocus && vim.active && vim.use<C-b> && !inDebugRepl && vim.mode != 'Insert'"
    },
    {
        "key": "ctrl+k",
        "command": "-extension.vim_ctrl+k",
        "when": "editorTextFocus && vim.active && vim.use<C-k> && !inDebugRepl"
    },
    {
        "key": "ctrl+b b",
        "command": "workbench.action.toggleSidebarVisibility"
    },
    {
        "key": "ctrl+b",
        "command": "-workbench.action.toggleSidebarVisibility"
    },
    {
        "key": "ctrl+b a",
        "command": "workbench.action.toggleActivityBarVisibility"
    },
    {
        "key": "ctrl+b f",
        "command": "workbench.files.action.showActiveFileInExplorer"
    },
    {
        "key": "ctrl+b n",
        "command": "workbench.action.nextSideBarView"
    },
    {
        "key": "ctrl+b p",
        "command": "workbench.action.previousSideBarView"
    },
    {
        "key": "ctrl+b m",
        "command": "workbench.action.toggleMenuBar"
    },
    {
        "key": "ctrl+b s",
        "command": "workbench.action.toggleStatusbarVisibility"
    },
    {
        "key": "ctrl+b l",
        "command": "workbench.action.toggleSidebarPosition"
    },
    {
        "key": "ctrl+b o",
        "command": "outline.focus"
    },
    {
        "key": "ctrl+b e",
        "command": "workbench.files.action.focusFilesExplorer"
    },
    {
        "key": "ctrl+b r",
        "command": "search.action.focusSearchList"
    },
    {
        "key": "ctrl+' l",
        "command": "workbench.action.togglePanelPosition"
    },
    {
        "key": "ctrl+' '",
        "command": "workbench.action.togglePanel"
    },
    {
        "key": "ctrl+j",
        "command": "-workbench.action.togglePanel"
    },
    {
        "key": "ctrl+' c",
        "command": "workbench.debug.panel.action.clearReplAction",
        "when": "inDebugRepl"
    },
    {
        "key": "meta+l",
        "command": "-workbench.debug.panel.action.clearReplAction",
        "when": "inDebugRepl"
    },
    {
        "key": "ctrl+' c",
        "command": "workbench.action.closePanel"
    },
    {
        "key": "ctrl+' space",
        "command": "workbench.action.focusPanel"
    },
    {
        "key": "ctrl+' n",
        "command": "workbench.action.nextPanelView"
    },
    {
        "key": "ctrl+' p",
        "command": "workbench.action.previousPanelView"
    },
    {
        "key": "ctrl+' shift+-",
        "command": "workbench.action.toggleMaximizedPanel"
    }
]

Compile and install vim 8.1 from source (debian / MX Linux) with pyenv

Karl N. Redman — Fri, 15 Mar 2019 12:00:36 +0000

This article is intended for advanced users -but to aid those who want to run cutting edge vim things.
Vim 8.1 provides :terminal. This is BIG NEWS and has opened some pretty cool plugins such as markdown preview plugin for (neo)vim. Additionally, if you want deoplete.nvim to work properly you'll need to compile vim against python3 anyway. The most single/main user way I've found to make these plugins work is to compile Vim from source. This is a method that I feel is least obtrusive to a single systerm.
This HowTo will add vim v8.1 to your system under '/usr/local' and set it as default (with python 3.7.2 and other common necessities -adjust as needed)
This method follows the 'don't break debian' mantra as close as possible IMHO.
Disclaimer: !! Test in a VM or otherwise at your own risk !!!
Tested with:
- MX Linux v18.1
- Debian Stretch v9.8
references:

Install and setup pyenv

pyenv provides a localized python, versioned, installation at the user level.

reference: pyenv/pyenv-installer: This tool is used to install pyenv and friends.

curl https://pyenv.run | bash

config: add to your ~/.bashrc file and relogin (terminal or desktop env)

# Load pyenv automatically by adding
# the following to ~/.bashrc:

export PATH="/home/karl/.pyenv/bin:$PATH"
eval "$(pyenv init -)"
eval "$(pyenv virtualenv-init -)"

install a python version

# list available versions
pyenv install -l |less
# insta a version
pyenv install 3.7.2

set a python version for global

pyenv global 3.7.2

Script for vim compile and installation

#! /usr/bin/env bash

# Install vim from source on a debian based system where pyenv is used at thhe user level.
## The python version must already be installed via pyenv
## `pyenv install 3.7.2`
## It is not necessary remove debian installed vim packages


# fail on any error (fix things as needed)
set -e

# Convenience variables
FULL_VERSION="3.7.2"
MM_VERSION="3.7"
VIM_CONENSED_VER="81"
VIM_TAG="v8.1"
BUILT_BY="Karl N. Redman <karl.redman@gmail.com>"

# Install dependencies (note no python-dev here)
sudo apt install -y libncurses5-dev libgnome2-dev libgnomeui-dev \
libgtk2.0-dev libatk1.0-dev libbonoboui2-dev \
libcairo2-dev libx11-dev libxpm-dev libxt-dev \
ruby-dev lua5.1 liblua5.1-dev libperl-dev git

# Checkout vim
git clone https://github.com/vim/vim.git
cd vim

# make sure the repo is clean
make distclean

# (optional) Checkout the specific desired version
## umcomment to use specific tagged version
# git pull --tags
# git checkout tags/${VIM_TAG}

# Run `Configure`
./configure --with-features=huge \
            --enable-multibyte \
            --enable-rubyinterp=yes \
            --with-x \
            --enable-perlinterp=yes \
            --enable-luainterp=yes \
            --enable-gui=gtk2 \
            --enable-cscope \
            --prefix=/usr/local \
            --with-compiledby="${BUILT_BY}" \
            --enable-python3interp=yes \
            --with-python3-config-dir=$HOME/.pyenv/versions/${FULL_VERSION}/lib/python${MM_VERSION}/config-${MM_VERSION}m-x86_64-linux-gnu \
            --includedir=$HOME/.pyenv/versions/${FULL_VERSION}/include/

# Make and set the runtime directory (non default to avoid conflicts)
make VIMRUNTIMEDIR=/usr/local/share/vim/vim${VIM_CONENSED_VER}

## (optional) uncomment to remove current vim if desired
# sudo apt remove -y vim vim-runtime gvim vim-tiny vim-common vim-gui-common vim-nox

# install (to usr local)
sudo make install

# update alternatives to set our new build as default
sudo update-alternatives --install /usr/bin/editor editor /usr/local/bin/vim 1
sudo update-alternatives --set editor /usr/local/bin/vim
sudo update-alternatives --install /usr/bin/vi vi /usr/local/bin/vim 1
sudo update-alternatives --set vi /usr/local/bin/vim

echo "done."

That's it!

relogin (if necessary) and vim --version should report v8.1! Now you can try out :terminal to make sure it's working ... and take advantage of the new plugins that are a comming!

Resolve docker hostnames from host with DNS Proxy Server

Karl N. Redman — Sun, 03 Mar 2019 13:28:24 +0000

Last updated: 3/Mar/2019

This article is a continuation of my previous HowTo: Dnsmasq + NetworkManager + Private Network Setup. With just a few configuration changes we will add the capability to interact with docker containers by their respective hostnames. Even though the dns-proxy-server project affords us this new functionality I recommend that dns-proxy-server only be used for development purposes.

While it's awesome to have the ability to reference individual containers by hostname during development efforts the concept doesn't scale well. In addition the dns-proxy-server (DPS) suffers from some performance issues that should be taken into consideration before using it as a docker service.

Additional Features added to the previous HowTo:

Reference FQDN-like container hostnames from the docker host
Reference FQDN-like container hostnames from other containers
Reference FQDN-like container hostnames via the browser without port specifications
Consistent behavior whether your host system is online or offline
Works with both docker run and docker-compose containers in a mixed environment
NetworkManager functionality remains viable and effective
Set-it-and-forget-it configuration

Tested systems:

MX Linux v18.1
Ubuntu v18.10
Debian v9.8

Disclaimer:

I strongly urge that you initially test this procedure with a virtual machine. I am not responsible if you mess up your system, data, etc.. This information is provided as is -use at your own risk.
This article assumes that you have configured your target system as outlined in the article Dnsmasq + NetworkManager + Private Network Setup.
Note: The use of all-servers setting for dnsmasq may produce undesirable behavior or could be considered a security risk -depending on your standards. See: Man page of DNSMASQ for more information if you are concerned.

How To:

Allow for async nameserver lookups

directive all-servers to /etc/dnsmasq.conf
also sanity check the settings

all-servers         # async call to all nameservers
bind-dynamic        # could use bind-interfaces instead
listen-address=127.0.0.1
listen-address=10.127.127.1
address=/private.home/10.127.127.1
domain=private.home,10.127.127.0/24
domain=docker.devnet,172.17.0.0/24

Restart dnsmasq

sudo service dnsmasq restart

Add DPS configuration to NetworkManager

DPS will always bind to address 172.17.0.2

nmcli con modify 'Wired connection 1' ipv4.dns 172.17.0.2
nmcli con modify 'Wired connection 1 offline' ipv4.dns 172.17.0.2
#
nmcli con up 'Wired connection 1'

Create DPS config file

Create directories and file

sudo mkdir -p /opt/dns-proxy-server/conf
sudo touch /opt/dns-proxy-server/conf/config.json

reference for options: running dns-proxy-server
- fallback dns: 8.8.8.8
- become a secondary dns server
- set log level to: WARNING
- set logfile to /opt/dns-proxy-server/dps.log
- allow command line ping for container names
add DPS settings to the config file

{
    "remoteDnsServers": [ ["8.8.8.8"] ],
    "envs": [
        {
            "name": ""
        }
    ],
    "activeEnv": "",
    "lastId": 0,
    "webServerPort": 0,
    "dnsServerPort": 0,
    "defaultDns": false,
    "logLevel": "WARNING",
    "logFile": "/opt/dns-proxy-server/dps.log",
    "registerContainerNames": true
}

Obtain and build a docker image for testing

git clone https://github.com/karlredman/lighttpd-docker.git
#
cd lighttpd-docker
#
sudo docker build -t lighttpd .

Start DPS

docker run --rm --hostname dns.mageddo --name dns-proxy-server -p 5380:5380 -v /opt/dns-proxy-server/conf:/app/conf -v /var/run/docker.sock:/var/run/docker.sock -v /etc/resolv.conf:/etc/resolv.conf defreitas/dns-proxy-server

Start example containers

# start FQDN hostname docker container 1
sudo docker run -d -p 8081:80 -p 4441:443 --rm -t --name docker-container-1-name -h docker-container-1.docker.devnet --net docker.devnet  lighttpd

# start container 2
sudo docker run -d -p 8082:80 -p 4442:443 --rm -t --name docker-container-2-name -h docker-container-2.docker.devnet --net docker.devnet lighttpd

# start container 3
sudo docker run -d -p 8083:80 -p 4443:443 --rm -t --name docker-container-3-name -h docker-container-3.docker.devnet --net docker.devnet lighttpd

Test pings

note pings in offline mode will be slow due to round robin dns resolution

ping -c 4 localhost
ping -c 4 [your host name]
ping -c 4 [some system on your network]

ping -c 4 dns.mageddo
ping -c 4 docker-container-1.docker.devnet
ping -c 4 docker-container-3.docker.devnet
ping -c 4 docker-container-2-name       # fails offline mode

sudo docker exec -ti docker-container-1-name sh -c "ping -c 4 host.docker"
sudo docker exec -ti docker-container-1-name sh -c "ping -c 4 a-wildcard.private.home"
sudo docker exec -ti docker-container-2-name sh -c "ping -c 4 docker-container-3-name"
sudo docker exec -ti docker-container-3-name sh -c "ping -c 4 docker-container-1.docker.devnet"

View DPS web page (doesn't seem to work for me)

http://localhost:5380
http://dns.mageddo:5380

Test addresses via browser

Test docker-compose instance

start the container

cd lighttpd-docker
sudo docker-compose up -d

test ping to the compose container

ping -c 4 docker-container-4.docker.devnet
sudo docker exec -ti docker-container-4-name sh -c "ping -c 4 a-wildcard.private.home"
sudo docker exec -ti docker-container-4-name sh -c "ping -c 4 docker-container-3-name"
sudo docker exec -ti docker-container-4-name sh -c "ping -c 4 docker-container-1.docker.devnet"
#
sudo docker exec -ti docker-container-1-name sh -c "ping -c 4 docker-container-4.docker.devnet"

test browser access to the container
- http://localhost:8084/hostname.html
- http://docker-container-4.docker.devnet/hostname.html

That's It!!

Dnsmasq + NetworkManager + Private Network Setup

Karl N. Redman — Fri, 01 Mar 2019 13:51:00 +0000

This is a howto/method for minimal configuration which provides a local private network that will be available to the local host for both online and offline (internet/DHCP) connectivity. The goal is to provide a scalable and network switchable development environment for virtual machines and bare metal application systems. This method intends to provide the developer the ability to roam (via laptop, et al) and maintain a consistent network development environment. Note that this configuration will require that you become accustomed to using NetworkManager.

A TL;DR version of this article can be found here.

Features of this howto:

Reference Docker containers by name between containers.
Reference the host (private network) with wildcards from both the host and containers.
Work/develop offline or online (almost) seamlessly.
Reference internet hosts from docker containers (if connected -most likely DHCP).
Reference local host network hosts (if connected -most likely DHCP and there is a DNS running).
No extra configuration with /etc/hosts
Ability to implement (your extended configurations):
- DNS caching.
- DHCP on a private network local to the host.
- shared networking via NetworkManager from a host interface.
- various VMs on the private network with full local DNS support.

Caveats / Limitations

Your bare metal services will need to be configured for the private network (i.e. 10.127.127.1) in order to utilize 'docker container to host' DNS resolution.
This procedure does not provide a method to resolve container hostnames from the host.
This procedure does not provide a method to resolve container hostnames from the from containers (see below for an explanation).
- containers can, however, use Docker's internal DNS to resolve containers by container names.

Tested systems:

MX Linux v18.1
Ubuntu v18.10
Debian v9.8

Disclaimer:

I strongly urge that you initially test this procedure with a virtual machine. I find Virtualbox to be the easiest sandbox installation platform for desktop systems but your mileage may vary. Either way, use a spare bare metal system or a virtual machine for testing this method to see if it works for you and your environment. I am not responsible if you mess up your system, data, etc.. This information is provided as is -use at your own risk.

Also, I tend to write wordy howto's with extra information that you may not care about -as evidenced by this sentence. I'll try to keep that to a minimum here. ;)

Introduction:

TL;DR: Skip this part -it's a motivation statement...

Developing network-able services across the myriad virtual machine/container options (Docker, kubernetes, singularity, libvert, singularity, LXC, etc.) is challenging. An additional challenge is trying to develop these kinds of services while, as a developer, roaming between networks. For me the answer is to emulate the various network environments as closely as possible. One step toward this endeavor is to establish a basic platform whereby a single system (i.e. laptop) can roam between different network environments and maintain a consistent production-like network simulation -whether online, offline, at work, at a party, at your friends house, in a mall, on the road, or at home (etc.).

There are a lot of 'moving parts' when developing any software application. That number of moving parts increases by orders of magnitude for every system resource (i.e. basic network connectivity compounds any application -add any concept of scalability and the complexity of the application dramatically increases). My attempt here is to provide a minimalist configuration whereby the reader doesn't have to understand how everything works right now. Rather, as needs for various development environments and tools become relevant, this basic network structure (hopefully) should provide a general baseline for adding emulation capabilities for future development capabilities. In the end, this howto is really simple -yet the ability to develop in a private network with a roaming system (i.e. laptop) eludes many, many developers. The moving parts to building a proper development environment become very confusing and very frustrating to many of us. It's overwhelming and thus, we hack...

I came up with this solution so that I could develop across the many virtual machine/container options on a single machine (laptop) regardless of the environment/location and the network I'm working in. With a simple configuration of NetworkManager IP addressing + system level dnsmasq I can now create any service for any IP (on the private network) and ultimately proxypass it (with whatever proxypass tool) to my host network domain -allowing me to both provide shared services across networks and/or develop any number of combined virtual systems on an isolated system; all on a single system (laptop).

Hopefully this method is a step toward providing the basis of an environment which is helpful for isolated and corporate and cloud development alike.

Configure NetworkManager

Presumably a fresh installation of various linux flavors will configure NetworkManager at installation time. The following information assumes that you have installed linux, are able to connect to the internet via DHCP, and do not have other issues pending with your network.

Add private network ip to network manager

This section implements a static ip address for the local host to use with future networks and services. All future networks on the system will use this base network -base address as configured here. To leverage this network for virtual systems you will need to identify your vitual interfaces to use this network -or otherwise configure a bridge (beyond the scope of this article).

reference network manager - How to Add dnsmasq and keep systemd-resolved (18.04) - Ask Ubuntu
determine the connection to use (i.e. eth0, wlan0, etc.)
- command

nmcli connection show

* example output

NAME                UUID                                  TYPE            DEVICE
Wired connection 1  xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx  802-3-ethernet  eth0
br-0a159c4a8cf4     xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx  bridge          br-0a159c4a8cf4
br-b1bce84765b8     xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx  bridge          br-b1bce84765b8
docker0             xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx  bridge          docker0

add a 10 network address to our Network Manager connection

nmcli con mod 'Wired connection 1' ipv4.address 10.127.127.1/24
nmcli con mod 'Wired connection 1' connection.autoconnect yes

Set the (auto) connection priority for the connection
- This will force this connection to be attepted first followed by the offline (cloned) version configured below

nmcli connection modify 'Wired connection 1' connection.autoconnect-priority -998

update connection (i.e. save the changes)

nmcli con up 'Wired connection 1'

Note: do not add the new IP to your /etc/hosts file. A network/host name for this new address will be configured later through Dnsmasq.

Clone connection and configure

NetworkManager will hang (seemingly) indefinitely when you connect to a network that does not provide DHCP connectivity. Cloning the working connection and making some adjustments will provide the ability to work 'offline' while maintaining the private netowrk with the system dnsmasq instance and it's (if configured) DHCP capabilities.

When a network does not offer DHCP capability you will use this connection to gain access to your local private network.

Note: switching between NetworkManager connections may require that you restart dnsmasq: sudo service dnsmasq restart. This issue occurs mostly when switching from a fully DHCP authenticated internet connection to one that is not (i.e. a connection that is not DHCP/internet enabled). In the reverse I've found, rarely, a need to run sudo dhclient when switching from an isolated network to one that is DHCP enabled. Additionally, I've had to reboot in various situations when switching from DHCP networks to non and visa versa -milage may very (sorry for the ambiguity -that's the state of networking...).

Clone the connection

nmcli connection clone 'Wired connection 1' 'Wired connection 1 offline'

Change cloned connection settings in favor of isolation

nmcli connection modify 'Wired connection 1 offline'  ipv4.method manual

(Optional) Install resolvconf

this is a workaround for Network Manager that allows for finer grained resolver configuration

sudo apt install resolvconf

Install and configure Dnsmasq

Install dnsmasq

# uninstall network manager version of dnsmasq
sudo apt remove dnsmasq-base -y

# install actual dnsmasq
sudo apt install dnsmasq -y

enable dnsmasq service

sudo systemctl enable dnsmasq

Configure dnsmasq

Dnsmasq configuration `/etc/dnsmasq.conf`

address=/private.home/10.127.127.1
listen-address=127.0.0.1
listen-address=10.127.127.1
bind-dynamic       # could use bind-interfaces instead
domain=private.home,10.127.127.0/24
domain=docker.devnet,172.17.0.0/24

Restart Dnsmasq and inspect hosts file

restart dnsmasq service

sudo service dnsmasq restart

!!!! REBOOT !!!!!

This is a cringe worthy necessity. Just reboot, it's easy. Accept that the moving parts are more than you want to deal with and that it's really less time to just reboot. I, personally, hate this statement but ... this is the state of things -accept it or spend some time finding a universal way around the reboot and report back here :D

your /etc/resolv.conf file will look similar to this:

# Dynamic resolv.conf(5) file for glibc resolver(3) generated by resolvconf(8)
#     DO NOT EDIT THIS FILE BY HAND -- YOUR CHANGES WILL BE OVERWRITTEN
nameserver 127.0.0.1
search your-isp.net

your /etc/hosts file will look something like this

127.0.0.1   localhost
127.0.0.1   this-laptop

# The following lines are desirable for IPv6 capable hosts
::1     localhost ip6-localhost ip6-loopback
fe00::0     ip6-localnet
ff00::0     ip6-mcastprefix
ff02::1     ip6-allnodes
ff02::2     ip6-allrouters

Test Dnsmasq

After reboot:

test local private network (should print as 10.127.127.1)

ping -c 4 private.home

test google DNS (from a DHCP connected network)

ping -c 4 google.com

test host hostame

ping -c 4 [your hostname here]

test another host on the network

ping -c 4 [a local network hosthame here]

Things should be working so far...

If you don't have responses that seem correct at this point -if something seems broken -then either this method doesn't work for you on your OS or you skipped a step (or I goofed and the documentation is wrong).

If you can't ping private.home and/or google.com by name then something is wrong. Please do ping 8.8.8.8 (google name server) explicitly to see if you have network access and work backwards from there. ...

Add Docker container DNS resolution

Important:

This section demonstrates docker container dynamic host name resolution for container to container and container to host resolution. Modern Docker container names can resolve across a bridge network -which we will create. Also, because we have established a private network, containers will be able to resolve a host's hostname back to our private 10.127.127.1 ip address (established earlier).

Note: resolving host to container hostnames is generally not required (or even considered a good idea) in production environments. The point of view is that virtual machines export ports to the host without regard for (or, more appropriately, in spite of) localhost. localhost in a container is not the host's localhost -i.e. the container's localhost is 127.0.0.1 within it's own namespace/address-space while the host's localhost is 127.0.0.1 relative to physical components of the host system. The container has it's own localhost separate from the host system.

install docker

do an sudo apt update
download and install the .deb file manually (because mx linux doesn't natively support /etc/apt/sources.list)
- reference: get docker ce for debian | docker documentation
- reference: toplevel download path
  1. download and install docker-cli
  2. download and install containerd.io
    - the docker daemon will start automatically
  3. download and install docker-ce
test docker

sudo docker run hello-world

Install Docker Compose

reference Install Docker Compose
- Releases · docker/compose

# be sure to get the latest version (that is compatible with docker -most likely latest...)
#
sudo curl -L "https://github.com/docker/compose/releases/download/1.23.2/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose
sudo chmod +x /usr/local/bin/docker-compose

Build an image for testing

This image is used for testing throughout this documentation where Docker is specified.

git clone https://github.com/karlredman/lighttpd-docker.git
#
cd lighttpd-docker
#
sudo docker build -t lighttpd .

Create common docker network for containers

This section creates a common network that all docekr run commands and docker-compose instances may communicate with one another -including the ability to resolve docker container names from within the respective container instances.
Specify a . (dot) delimited network name for finicky applications that require a FQDN (fully qualified domain name).
Note: use - in names instead of _ when possible (if needed) to maintain compatibility with various tools.

sudo docker network create docker.devnet

Add DNS to Docker daemon

file /etc/docker/daemon.json

{
    "dns": ["10.127.127.1", "8.8.8.8"]
}

restart docker daemon

sudo service docker restart

Testing host to Docker container DNS resolution

Start test containers

# start FQDN hostname docker container 1
sudo docker run -d -p 8081:80 -p 4441:443 --rm -t --name docker-container-1-name -h docker-container-1.docker.devnet --net docker.devnet lighttpd

# start container 2
sudo docker run -d -p 8082:80 -p 4442:443 --rm -t --name docker-container-2-name -h docker-container-2 --net docker.devnet lighttpd

# start container 3
sudo docker run -d -p 8083:80 -p 4443:443 --rm -t --name docker-container-3-name -h docker-container-3 --net docker.devnet lighttpd

Test DNS resolution

####### normal docker behavior with user-defined network 'docker.devnet'
#
# ping container 3 *container name* from container 2
sudo docker exec -ti docker-container-2-name sh -c "ping -c 4 docker-container-3-name"
#
# ping private network host wildcard from a container
## Note: if this part isn't working then you likely are being blocked by your own firewall
sudo docker exec -ti docker-container-1-name sh -c "ping -c 4 a-wildcard.private.home"

That's it!

I hope this workes for you...

DEV Community: Karl N. Redman

A message from the FUTURE: 2016

Randomize your XFCE / LightDM Login Screen Wallpaper

Overview:

TL;DR:

Single vs Multi User system and encrypted home directories:

Generic setup

clone the example project to an appropriate directory:

add images to the image directory Variety

run the project script to seed the publish directory

Configure LightDM Greeter via the GUI interface

test and debug

add cron job

Using Variety as the image source:

I cut my hair for an interview -oh! No!

Be a Python Jinja Master: Automated Front Matter Editing

TL;DR

The Pitch:

A demonstration:

My Personal Use Case:

Conclusion:

Private Network + Traefik + Keycloak + Gatekeeper Overview (diagram)

Keycloak v5 + Gatekeeper v5: Flowcharts - Easily Create and Restrict an Isolated (IODC) Client Service by Group-Role

Artifact Specifics:

Flowchart Caveates:

Relevant Documentation References:

The Keycloak (v.5.0.0) UI mind map you didn't ask for

Navigate your vscode like it's 1999 (the vim way)

Vim configuration (at your discretion)

Open you current vim document in vcode

Use ctrl+j and ctrl+k to change tabs in vim (I prefer up/down -adjust as needed)

VS Code changes

features from keybindings.json (+noted settings.json)

The files / configs

Compile and install vim 8.1 from source (debian / MX Linux) with pyenv

Install and setup pyenv

Script for vim compile and installation

That's it!

Resolve docker hostnames from host with DNS Proxy Server

Additional Features added to the previous HowTo:

Tested systems:

Disclaimer:

How To:

Allow for async nameserver lookups

Add DPS configuration to NetworkManager

Create DPS config file

Obtain and build a docker image for testing

Start DPS

Start example containers

Test pings

View DPS web page (doesn't seem to work for me)

Test addresses via browser

Test docker-compose instance

That's It!!

Dnsmasq + NetworkManager + Private Network Setup

Features of this howto:

Caveats / Limitations

Tested systems:

Disclaimer:

Introduction:

Configure NetworkManager

Add private network ip to network manager

Clone connection and configure

(Optional) Install resolvconf

Install and configure Dnsmasq

Configure dnsmasq

Dnsmasq configuration /etc/dnsmasq.conf

Restart Dnsmasq and inspect hosts file

!!!! REBOOT !!!!!

Test Dnsmasq

After reboot:

Things should be working so far...

Add Docker container DNS resolution

install docker

Install Docker Compose

Build an image for testing

Create common docker network for containers

Add DNS to Docker daemon

Testing host to Docker container DNS resolution

That's it!

Single vs Multi User system and encrypted `home` directories:

add images to the image directory `Variety`

run the project script to seed the `publish` directory

Using `Variety` as the image source:

Use `ctrl+j` and `ctrl+k` to change tabs in vim (I prefer up/down -adjust as needed)

features from `keybindings.json` (+noted settings.json)

Dnsmasq configuration `/etc/dnsmasq.conf`