Skip to main content

Celery platform to create microservices

Project description

travis coverage pypi pyversion

A platform to create microservices available through celery and Rest (using gunicorn) APIs.

Micro API

Micro uses a very simple API to run, list and get information about plugins:

  • plugins(): list all available plugins.
  • info(plugin_name): show information about a specific plugin.
  • help(plugin_name): show the plugin help.
  • run(plugin_name, params): execute the given plugin.

To use this API with Celery you can use the Micro-dev package who provides two important libraries, the access to the Celery API and the PluginBase class who allow writing Micro plugins. To use it as API Rest you can use the requests python library.

API Celery example (using micro-dev)

>>> from micro.api.endpoints import Requests
>>> req = Requests(BROKER_URL, QUEUE_NAME)
>>> req.plugins.delay().wait()
'[{"name": "Example Plugin", "version": null, "description": "A very simple example plugin"}]'
>>>"Example plugin", name="Micro").wait()
{"msg": "Hello Micro!!!"}

API Rest example (using requests)

>>> import requests
>>> url = "http://localhost:8000/plugins"
>>> response = requests.request("GET", url)
>>> print(response.text)
[{"name": "Example Plugin", "version": null, "description": "A very simple example plugin"}]
>>> url = "http://localhost:8000/run/Example%20Plugin"
>>> payload = '{"name": "Micro"}'
>>> headers = {'content-type': 'application/json'}
>>> response = requests.request("POST", url, data=payload, headers=headers)
>>> print(response.text)
{"msg": "Hello Micro!!!"}

Micro plugins

Write Micro plugins is very simple all that you need is to create a file called and a class which is the plugin itself.

  • the file defines the plugin metadata
  • the plugin class must:
    • inherit from micro.plugin.pluginbase.PluginBase
    • implement the run method
    • return Python dictionaries on that run method
from micro.plugin.pluginbase import PluginBase
from micro.plugin.pluginbase import PluginDescription

class ExamplePlugin(PluginBase):
    def __init__(self):
        print("This is an example plugin")

    # This is the method executed by Micro
    # Must return a Python dictionary
    def run(self, name):
        return {"msg": "Hello " + name + "!!!"}

# This description is required by Micro
plugin = PluginDescription(
    name="Example Plugin",
    author="Jhon Doe",
    description="A very simple example plugin",
    long_description="This plugin is a very simple example, "
                     "for that reason, we don't have a long description",
    plugin_help="Params: name type string; A name to greet"

Each plugin needs to have its own folder inside of the plugins directory (check the section “Configuring Micro” for details)

The plugin directory should look like this:


IMPORTANT: All the plugins must provide its own libraries inside of its own namespace in order to avoid overwritting files. The general recomendation is to use a base directory with the same name of the plugin as the example above shows



$ pip install micro

Development version:

$ git clone
$ cd micro
$ pip install -e .

or direct from repository:

$ pip install git+ssh://


Parameters priority

Micro can be configurated through CLI, environment variables, config file and/or default values (in that order).

Command line (CLI)

These arguments are the highest priority for Micro, so these overwrite any other parameters set by any other method. The CLI arguments that can be used are:

$ micro -h
usage: micro [-d] [-h] [-v] [-c CONFIG_FILE] [-b BIND] [-B BROKER_URL]
             [-C] [-cw WORKERS] [-G] [-gw WORKERS] [-H HOSTNAME]
             [-ln LOG_FILE_NAME] [-lp LOG_FOLDER_PATH] [-ll LOG_LEVEL]
             [-pp PID_FOLDER_PATH] [-p PLUGIN_PATH] [-q TASK_QUEUES]

optional arguments:
  -d, --default-values  show default values and exit
  -h, --help            show this help message and exit
  -v, --version         show program's version and exit
  -c CONFIG_FILE, --config-file CONFIG_FILE
                        path to the config file
                        env: MICRO_CONFIG_FILE
                        default: None
  -b BIND, --bind BIND  Gunicorn bind, HOST:PORT
                        env: MICRO_BIND
  -B BROKER_URL, --broker-url BROKER_URL
                        Celery broker URL
                        env: MICRO_BROKER_URL
                        default: None
  -C, --celery          plugins available through Celery
                        env: MICRO_CELERY
                        default: None
  -cw WORKERS, --celery-workers WORKERS
                        Celery number of workers
                        env: MICRO_CELERY_WORKERS
                        default: 1
  -G, --gunicorn        plugins available through API Rest
                        env: MICRO_GUNICORN
                        default: None
  -gw WORKERS, --gunicorn-workers WORKERS
                        Gunicorn number of workers
                        env: MICRO_GUNICORN_WORKERS
                        default: 1
  -H HOSTNAME, --hostname HOSTNAME
                        Celery worker's hostname
                        env: MICRO_HOSTNAME
                        default: micro
  -ln LOG_FILE_NAME, --log-file-name LOG_FILE_NAME
                        Micro's log file name
                        env: MICRO_LOG_FILE_NAME
                        default: micro.log
  -lp LOG_FOLDER_PATH, --log-folder-path LOG_FOLDER_PATH
                        path to the Micro's log folder
                        env: MICRO_LOG_FOLDER_PATH
                        default: /var/log/micro
  -ll LOG_LEVEL, --log-level LOG_LEVEL
                        Micro's log level
                        env: MICRO_LOG_LEVEL
                        default: WARNING
  -pp PID_FOLDER_PATH, --pid-folder-path PID_FOLDER_PATH
                        path to the Micro's PID folder
                        env: MICRO_PID_FOLDER_PATH
                        default: /var/run/micro
  -p PLUGIN_PATH, --plugin-path PLUGIN_PATH
                        path to the plugins folder
                        env: MICRO_PLUGIN_PATH
                        default: None
  -q TASK_QUEUES, --task-queues TASK_QUEUES
                        Celery task queues
                        env: MICRO_TASK_QUEUES
                        default: None

IMPORTANT: -G, –gunicorn or -C, –celery, or both, must be included otherwise Micro will finish without any message

Environment variables

The next priority in parameters for Micro are environment variables. The list of environment variables used are:

MICRO_CONFIG_FILE        # config file location: /path/to/config/config.json
MICRO_BIND               # Gunicorn socket bind (host:port)
MICRO_BROKER_URL         # broker url: ampq://user:pass@host:port//
MICRO_CELERY             # plugins available through Celery
MICRO_CELERY_WORKERS     # number of Celery workers (integer number)
MICRO_GUNICORN           # plugins available through API Rest (Gunicorn)
MICRO_GUNICORN_WORKERS   # number of Gunicorn workers (integer number)
MICRO_HOSTNAME           # workers hostname
MICRO_LOG_FILE_NAME      # log file name: micro.log
MICRO_LOG_FOLDER_PATH    # path to log folder: /path/to/log/folder
MICRO_LOG_LEVEL          # minimun log level to write: DEBUG, INFO, WARNING, ERROR, CRITICAL or FATAL
MICRO_PID_FOLDER_PATH    # path to Celery pid folder: /path/to/pid/folder
MICRO_PLUGIN_PATH        # path to plugin folder: /path/to/plugin/folder
MICRO_TASK_QUEUES        # queue name used

Config file

The lowest priority is the use of a JSON config file. The path to this config file must be set using -c, --config-file CLI arguments or MICRO_CONFIG environment variable.

Celery and Gunicorn configurations can be given through this config file as well.

Config file example:

    "gunicorn": {
        "bind": "",
        "workers": 1
    "celery": {
        "broker_url": "",
        "workers": 1,
        "hostname": "micro",
        "task_queues": ""
    "micro": {
        "log_file_name": "micro.log",
        "log_folder_path": "/var/log/micro",
        "log_level": "WARNING",
        "pid_folder_path": "/var/run/micro",
        "plugin_path": ""

A config file skeleton can be created using the following command: $ micro --default-values > config.json

Default values

The default values are the same shown above.

CORS support

Micro implements CORS support using Flask-CORS the configuration must be made inside of the Gunicorn key using the cors key, the possible values are listed in the CORS documentation

For example this configuration allow any client:

    "gunicorn": {
        "bind": "",
        "workers": 1,
        "cors": {"origins": "*"}



To download from Docker Hub:

$ docker pull humu1us/micro:<tag>

To check the available tags please visit Micro’s repository on Docker Hub


To build the container first move to the branch/tag to use and then use the following command:

$ docker build -t micro:<tag> .


All Micro environment variables are available with -e flag. For example to run Micro with Celery you can do:

$ docker run -d \
    -v /path/to/plugins:/etc/micro/plugins \
    -v /path/to/log:/var/log/micro \
    -e MICRO_BROKER_URL=amqp://guest:guest@my_host:5672// \
    -e MICRO_TASK_QUEUES=test \
    -e MICRO_HOSTNAME=my_host \
    -e MICRO_CELERY=1 \

MICRO_BROKER_URL and MICRO_TASK_QUEUES are the only mandatory environment variables to set when Celery will be used.

When Micro will be run with API Rest you have to bind the Gunicorn port:

$ docker run -d \
    -v /path/to/plugins:/etc/micro/plugins \
    -v /path/to/log:/var/log/micro \
    -e MICRO_BIND= \
    -p 5000:5000 \


Run all unit tests with:

$ python test

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for Micro, version 3.1.0
Filename, size File type Python version Upload date Hashes
Filename, size Micro-3.1.0-py3-none-any.whl (28.4 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size Micro-3.1.0.tar.gz (25.2 kB) File type Source Python version None Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page