Abstract all the things
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.
ansible Minor changes for simplify add/remove commands May 29, 2016


Abstract All the Things


  • A bash shell
  • python2.7
  • virtualbox
  • vagrant
  • A slack channel with hubot enabled (optional)


Get the code

git clone git@github.com:dravetech/preso_abstract_all_the_things.git
mkvirtualenv -a preso_abstract_all_the_things preso_abstract_all_the_things
pip install -r requirements.txt

Start the lab

For this demo I used an online lab hosted by network.toCode(). The lab name is MULTI-VENDOR EOS (2) AND VMX (2). If you want to try it out network.toCode() has been kindly enough to provide the promotion code napalm6. The promotion code will let you run the lab to the first 100 users. Please, don't abuse it, you have 6 hours to try out this demo, which should be plenty of time if you make sure your environment is set correctly.

Note: Neither me or my employer are in any way affiliated with network.toCode()

Note 2: The lab will take a while to start so be patient.

Prepare chatops environment (optional)

This is optional, if you are not interested in trying the API or the slack integration you can move on to the next section.

  1. Go to the link https://www.slack.com/apps/A0F7XDU93-hubot. You will probably be redirected to https://$your_organization_name.slack.com/apps/A0F7XDU93-hubot. Make sure your organization is selected on the box on the upper right corner.
  2. Click on configure and configure your hubot name, icon, etc. Make sure you save the API Token.
  3. Create a channel on your slack organization.
  4. Go to your slack channel and invite your bot.

Configuring the environment

After the lab is started and before actually starting to work with the demo you will have to configure your environment. To do so gather the following information:

  1. Devices' public IP's. You can find them under "Console & IP Address Information" in your lab dashboard.
  2. Slack channel name (optional).
  3. Hubot's API Token (optional).

When you have all the needed information execute the script configure_environment.sh:

$ ./configure_environment.sh
Enter the IP vmx1:
Enter the IP vmx2:
Enter the IP eos-spine1:
Enter the IP eos-spine1:
Enter your slack channel name: abstract_everything
Enter your slack token: xoxb-37241085586-fake

Start ST2 (optional)

This part is only necessary if you want to test the slack integration and the API. To start the vagrant box execute:

VAGRANT_CWD=./vagrant vagrant up

Note the IP and the TOKEN shown on the screen after finishing booting up the VM.

Test everything works


Ansible should work by now:

$ cd ansible
$ ansible-playbook deploy_network.yml -e commit_changes=0
# Deploy Configuration ****************************************************************************************************
  * vmx.core2                               - changed=True --------------------------------------------------------------
    [edit interfaces ge-0/0/1 unit 0]
    -      family inet;
    +      family inet6 {
    +          address 2001:db8:c4f3:3::/127;
    +      }
    [edit interfaces ge-0/0/2 unit 0]
    -      family inet;
    +      family inet6 {

Slack (optional)

Slack should work as well.

API (optional)

And the API too:

$ curl -k -H "Content-Type: application/json" -X POST https://$IP/api/v1/webhooks/deploy_fabric -H "St2-Api-Key: $St2-Api-Key" --data '{}'

Replace the $IP and the $St2-Api-Key with the info you gathered when starting the VM.

Time to play

Now it's time for you to play. Feel free to play with ansible, invoke, slack, whatever. It's all yours now : )

Destroying ST2 (optional)

When you are done don't forget to destroy the VM we used for the slack/API integration. You can do that by just going to the vagrant folder on this project and executing vagrant destroy.