envmgr |pypi| |travis| |appveyor| |dockerautomated| |dockerpulls|
=================================================================
A cross-platform CLI client for `Environment
Manager <https://github.com/trainline/environment-manager>`__
.. figure:: https://github.com/trainline/envmgr-cli/raw/master/example.gif
:alt: envmgr example
Install
-------
::
pip install envmgr-cli
See `Configuration`_ for further install instructions.
Usage
-----
*envmgr* is designed to provide an intuitive, human readable interface
around the lower level `Environment Manager Python
Library <https://github.com/trainline/python-environment_manager/>`__
All ``envmgr`` commands are exposed behind a set of verbs (*get* a
status, *schedule* some downtime, *wait* for an ASG, etc). Verbs are
always the first value provided to ``envmgr`` and there is only ever one
verb per command.
docopt
^^^^^^
The cli interface is described in `docopt <http://docopt.org/%5D>`__.
The easiest way to discover and understand the different usage patterns
available is to simply run
::
envmgr --help
Output
^^^^^^
By default, ``envmgr`` commands will output a human friendly response
useful for testing single commands at a time. To help scripting or
chaining results together, all commands also accept a ``--json``
argument which will return the raw JSON response from Environment
Manager:
::
>> envmgr schedule asg my-asg on in prod
Scheduled 1 instance in my-asg to: ON
>> envmgr schedule asg my-asg on in prod --json
{"ChangedInstances": ["i-0afe2276909859130"], "ChangedAutoScalingGroups": ["my-asg"]}
Examples
--------
*In the examples below, assume that 'prod-1' is an environment, 'AwesomeService' is a
service and 'my-asg' is an ASG, all of which are already registered in
Environment Manager.*
.. code-block:: bash
# Get the health status of all instances of AwesomeService, in all slices:
envmgr get AwesomeService health in prod-1
# Get the the active slice information for the AwesomeService service in prod-1 environment:
envmgr get AwesomeService active slice in prod-1
# Get the status of the my-asg ASG in the prod-1 environment. Status is calculated as an aggregate of all instances in the ASG:
envmgr get asg my-asg status in prod-1
# Get the schedule value set on the my-asg ASG in the prod-1 environment. Note this will tell you what the schedule is configured to - not the current state according to the schedule:
envmgr get asg my-asg schedule in prod-1
# Get the current status of the deployment with ID a2fbb0c0-ed4c-11e6-85b1-2b6d1cb68994:
envmgr get deploy status a2fbb0c0-ed4c-11e6-85b1-2b6d1cb68994
# Block and wait until the deployment with ID a2fbb0c0-ed4c-11e6-85b1-2b6d1cb68994 either succeeds or fails:
envmgr wait-for deploy a2fbb0c0-ed4c-11e6-85b1-2b6d1cb68994
# Block and wait until all instances in the ASG my-asg are ready fordeployment (In Service):
envmgr wait-for asg my-asg in prod-1
# Block and wait until the service AwesomeService is running with all healthchecks passing:
envmgr wait-for healthy AwesomeService in prod-1
# Set the schedule of the ASG my-asg in prod-1 to be off permanently until further notice:
envmgr schedule asg my-asg off in prod-1
# Publish the file build-22.zip as version 1.2.9 of AwesomeService:
envmgr publish build-22.zip as AwesomeService 1.2.9
# Deploy the published version 1.2.9 of AwesomeService into the prod-1 environment:
envmgr deploy AwesomeService 1.2.9 in prod-1
# Toggle the upstreams for AwesomeService in the prod-1 environment:
envmgr toggle AwesomeService in prod-1
# Get the Windows patch status for servers belonging to A-Team in prod-1:
envmgr get A-team patch status in prod-1
Configuration
-------------
Authentication
^^^^^^^^^^^^^^
All calls to Environment Manager require authentication, which can be
provided in 1 of 2 ways.
Either export your credentials as environment variables:
::
ENVMGR_USER=myusername
ENVMGR_PASS=mypa$$word
Or provide a ``--user`` and ``--pass`` value to each commad:
::
envmgr get MyService health in prod --user="sarah" --pass="pa$$word"
*Note: It's recommended to only use this method in CI environments.*
Host Config
^^^^^^^^^^^
The hostname of your Environment Manager instance is configured
similarly to your credentials:
Export your hostname as an environment variable:
::
ENVMGR_HOST=environmentmanager.acme.com
Or provide the hostname with each command:
::
envmgr get MyService health in prod --host=environmentmanager.acme.com
Development
-----------
To install all test dependencies and run all tests, simply run:
::
python setup.py test [--adopts -v]
For convenience this is also available via the included `makefile`:
::
make test
Docker
------
If you want, you can use our automated container builds
Usage
^^^^^
::
docker run -it --rm \
-e ENVMGR_USER=user
-e ENVMGR_PASS=password
-e ENVMGR_HOST=foo.bar
trainline/envmgr-cli:latest envmgr {YOUR_ARGS}
Example
^^^^^^^
::
~$ docker run -it --rm trainline/envmgr-cli envmgr --version
1.9.1
Build
^^^^^
::
docker build -t {YOUR_NAME}/envmgr-cli .
.. |appveyor| image:: https://ci.appveyor.com/api/projects/status/w50g5yb1fh4qh3rq/branch/master?svg=true
:target: https://ci.appveyor.com/project/duncanhall/envmgr-cli/branch/master
.. |travis| image:: https://travis-ci.org/trainline/envmgr-cli.svg?branch=master
:target: https://travis-ci.org/trainline/envmgr-cli
.. |pypi| image:: https://img.shields.io/badge/python-2.7%2C%203.4%2C%203.5%2C%203.6-blue.svg
:target: https://pypi.python.org/pypi/envmgr-cli
.. |dockerautomated| image:: https://img.shields.io/docker/automated/trainline/envmgr-cli.svg
:target: https://hub.docker.com/r/trainline/envmgr-cli
.. |dockerpulls| image:: https://img.shields.io/docker/pulls/trainline/envmgr-cli.svg
:target: https://hub.docker.com/r/trainline/envmgr-cli