The documentation you're currently reading is for version 3.5.0. Click here to view documentation for the latest stable version.
Vagrant & Virtual Appliance¶
Vagrant / OVA is a quick and easy way to try StackStorm in a VM.
It’s already pre-installed, configured and tested virtual machine image and so saves time avoiding time-consuming installation and configuration steps. Perfect for a quick platform overview, testing, demo or even using StackStorm in isolated from the internet air-gapped systems.
We highly recommend using a Vagrant box to get familiar with the StackStorm platform.
The source code is available as a GitHub repo: StackStorm/packer-st2
Starting a StackStorm Vagrant VM is easy:
vagrant init stackstorm/st2 vagrant up vagrant ssh
This will boot up a fresh StackStorm installation on Ubuntu 16.04 LTS.
You can access Web UI at
https://10.10.10.10/ or SSH to the box
vagrant ssh if you prefer st2 CLI.
As alternative to Vagrant box is Virtual appliance which is available for download as
image from the StackStorm/packer-st2 Github Releases
page. It might be especially helpful for the isolated from the internet air-gapped environments.
If using OVA in production environment, don’t forget to change the default credentials
and delete SSH authorized keys for
vagrant linux user.
Once the newer box version is released, Vagrant will warn you about the available update. To update the box:
vagrant box outdated vagrant box remove stackstorm/st2 vagrant up
Whether you want to pin StackStorm, it’s possible to use specific box version. While adding the box for the first time:
vagrant init stackstorm/st2 --box-version 2.7.1-20180507 vagrant up
Or directly in
Vagrant.configure("2") do |config| config.vm.box = "stackstorm/st2" config.vm.box_version = "2.7.1-20180507" end
The list of available box versions can be found at Vagrant Cloud.
Sometimes StackStorm does not run properly for some reason.
Discovering why at a infra level is the responsibility of
st2-integration-tests which will
perform StackStorm infrastructure/integration tests and report back with more detailed info.
This can save time for both you & community to avoid extensive troubleshooting steps.
If something went wrong, - just run