juanje

Connect services with docker-compose at multirepo project

Sometimes you have a project with different services (or microservices) and each one of them has its own repository. At those cases, tests the whole project (and its interactions) can be challenging.

With Docker and Docker Compose you can run easily each service from its repo, but for making them to see each other you (usually) need to manually create a network with Docker, assume so facts (as directories names) or do some crazy network configurations.

TL;DR

You have the final configuration here.

juanje

# This is a simple example of how to run a basic service inside a container with Podman

# Podman
## Pull the Docker image
podman pull docker.io/redis

## Run the container as you would do with Docker
podman run -d --name redis_server -p 6379:6379 redis

# But Podman facilitate some extra ways:

juanje

I was tired of Chrome eating all my laptop resources so I decided to put some limit to it with cgroup.

As I was using Ubuntu 12.04 with support for cgroup, I installed the package cgroup-bin and add the following group to the file /etc/cgconfig.conf:

group browsers {
    cpu {
#       Set the relative share of CPU resources equal to 25%
        cpu.shares = "256";
 }

juanje

@ECHO ON

REM This batch script is almost identical to http://code.can.cd/winrm_setup.bat
REM To use this from powershell on windows in a one liner:
REM (New-Object System.Net.WebClient).DownloadFile('https://gist.github.com/juanje/8496054/raw/winrm_setup.bat','winrm_setup.bat') ; .\winrm_setup.bat
cmd.exe /c winrm quickconfig -q
cmd.exe /c winrm quickconfig -transport:http
cmd.exe /c winrm set winrm/config @{MaxTimeoutms="1800000"}
cmd.exe /c winrm set winrm/config/winrs @{MaxMemoryPerShellMB="300"}
cmd.exe /c winrm set winrm/config/service @{AllowUnencrypted="true"}

juanje

from dataclasses import dataclass

DOMAIN = 'example.com'

@dataclass
class User:
    user_id: str = None
    name: str = None
    email: str = None
    role: str = 'standard'

juanje

This is a little trick I use to spin up the packages instalation on Debian/Ubuntu boxes in Vagrant.

I add a simple function that checks if a directory named something similar to ~/.vagrant.d/cache/apt/opscode-ubuntu-12.04/partial (it may have another path in Windows or MacOS) and create the directory if it doesn't already exist.

def local_cache(basebox_name)
  cache_dir = Vagrant::Environment.new.home_path.join('cache', 'apt', basebox_name)
  partial_dir = cache_dir.join('partial')
  partial_dir.mkdir unless partial_dir.exist?
 cache_dir

juanje

input {
  tcp {
    type => "linux-syslog"
    port => 3333
  }

  file {
    type => "linux-syslog"
    path => [ "/var/log/auth.log" ]
  }

juanje

---
- name: Compose variables and merge lists
  hosts: localhost
  connection: local
  gather_facts: no

  vars:
    api_versions:
      - 1
      - 2

juanje

#!/bin/bash

URL="http://ipv4.download.thinkbroadband.com/50MB.zip"

function get_file() {
  # Declare these local variables as integer
  local -i count
  local -i max_retries
  local -i wait_time
  local filename

juanje

# Just a place to write down notes for old time Debian/Ubuntu user
# that has just moved into Fedora/RedHat world

# Show which package provides the file '/bin/ls'
# Similar to 'dpkg -S /bin/ls'
$ rpm -qf /bin/ls
coreutils-8.5-7.fc14.i686

# The same but showing only the package name (without version)
$ rpm -qf /bin/ls --queryformat '%{NAME}\n'