allmychanges.com

Project is closed!

Here is the screenshot of the final email I've got from the service:

If you want to resurrect AllMyChanges, please, contact me as sasha at svetlyak.ru.

Dev environment

In a development environment, everything is deployed in separate docker containers. In them you need to raise mysql, redis and will start in separate containers all management commands like runerver. The difference from the production environment is only in the fact that in the maiden inside the docker container the whole tree is thrown with source codes, and when they change, reload the code, the container is reassembled and do not need to restart.

If a working machine with Linux, then it is better to put docker on it, if not, then the docker must be launched in VirtualBox, the next part will be about how to do this on OSX.

Raise docker on OSX

First you need to install the docker client and docker-machine via homebrew:

$ brew install docker docker-machine

If you have not yet installed VirtualBox, then you need to install it.

Next, create a virtual machine in which docker will run:

$ docker-machine create --driver virtualbox amch

When the team completes, a new virtual box will appear in your VirtualBox with the name amch and docker will be installed on it.

To work with this docker daemon from the OSX console, you need to install some environment variables. What kind? This will tell the command:

$ docker-machine env amch
export DOCKER_TLS_VERIFY="1"
export DOCKER_HOST="tcp://192.168.99.100:2376"
export DOCKER_CERT_PATH="/Users/art/.docker/machine/machines/amch"
export DOCKER_MACHINE_NAME="amch"
# Run this command to configure your shell:
# eval "$(docker-machine env amch)"

In order not to install them manually, you can either register eval" $(docker-machine env amch)" in ~/.bashrc, or use autoenv and write it to the .env file inside the repository AllMyChanges.

In any case, you need to check if the docker client works:

$ docker version
Client: Docker Engine - Community
 Version:           19.03.6
 API version:       1.40
 Go version:        go1.12.16
 Git commit:        369ce74a3c
 Built:             Thu Feb 13 01:27:49 2020
 OS/Arch:           linux/amd64
 Experimental:      false

Server: Docker Engine - Community
 Engine:
  Version:          19.03.6
  API version:      1.40 (minimum version 1.12)
  Go version:       go1.12.16
  Git commit:       369ce74a3c
  Built:            Thu Feb 13 01:26:23 2020
  OS/Arch:          linux/amd64
  Experimental:     false
 containerd:
  Version:          1.2.13
  GitCommit:        7ad184331fa3e55e52b890ea95e65ba581ae3429
 runc:
  Version:          1.0.0-rc10
  GitCommit:        dc9208a3303feef5b3839f4323d9beb36df0a9dd
 docker-init:
  Version:          0.18.0
  GitCommit:        fec3683

If the client does not find the server, then instead of information about the server, it will display something like:

Cannot connect to the Docker daemon. Is the docker daemon running on this host?

The problem here is that the same environment variables are not set.

If the virtual machine is not running, then the client will wait longer and the message will be different:

Cannot connect to the Docker daemon. Is the docker daemon running on this host?

Then you need to start the virtual machine with the command:

docker-machine start amch

In principle, this line can also be written into the local .env file for autoenv.

Starting mysql and redis

The project used fabric to start different teams, but now everything Corresponds to invoke.

I usually put all sorts of python libs and utilities right inside the project tree, with the env folder created through virtualenv.

Create an environment:

$ virtualenv env
$ pip install -U pip
$ pip install -r requirements/host.txt

The environment must be activated:

$ source env/bin/activate

This command is also convenient to add to the .env file. It’s only better to add as:

if [ -e env ]; then
  source env/bin/activate
fi

After activating the environment, the inv command should be available to you, which runs the tasks described in tasks.py.

To create containers with databases, run:

$ inv start_databases

If everything went as it should, then the docker ps command should show us two running container:

CONTAINER ID        IMAGE               COMMAND                  CREATED             STATUS              PORTS               NAMES
6277dc3c8bf3        redis               "/entrypoint.sh redis"   7 seconds ago       Up 7 seconds        6379/tcp            redis.allmychanges.com
06aa8ed1bea8        mysql               "/entrypoint.sh mysql"   10 minutes ago      Up 15 seconds       3306/tcp            mysql.allmychanges.com

Now we need to run migrations

To do this, somehow run ./manage.py migrate inside the container with the code AllMyChanges. But we do not have a container, so we must first build it.

When assembling the image so that pip runs faster, we use wheels precompilation. To do this, run the command:

$ inv build-wheels

Then you need to build a docker image:

$ inv build-image

And create a base:

$ inv create-database
$ inv manage migrate

Migration has to be run twice because the first time there climbs out some Integrity error.

It remains to start the web server

$ inv runserver

And asynchronous worker

$ inv rqworker

Yes, open the result in a browser

In order to see what went up in the docker, you need to get the address virtual computers:

$ docker-machine ip amch
192.168.99.100

This url must be written in /etc/hosts, adding the following line:

192.168.99.100 dev.allmychanges.com

After that, you can open http://dev.allmychanges.com in your browser and it will appear AllMyChanges interface.

How to get data from production and pour local database

If there is a database dump from production, then uploading it to the local database is easy and simple. It is enough to put the dump in the dumps directory, so that the file name is dumps/latest.tar.bz2 and after that you need to run the command:

inv restore-db

Warning previous data will be erased.

How to create and run migrations

Create:

inv manage makemigrations

Run:

inv manage migrate

How to run tests

Very simple:

inv test

It is also useful to run not all tests, but only flawed ones:

inv test --failed