NVIDIA/vagrant-swift-all-in-one

NVIDIA/vagrant-swift-all-in-one

Description: Vagrant Swift All In One

Language: Ruby

License: Apache-2.0

Stars: 80

Forks: 64

Open issues: 4

Created: 2013-03-07T00:41:51Z

Pushed: 2026-06-25T23:42:10Z

Default branch: master

Fork: no

Archived: no

README: vagrant-swift-all-in-one ========================

A virtualization toolchain for building an OpenStack Swift-All-In-One.

This project relies on vagrant, you'll need to install vagrant and setup one of the supported virtualization providers:

| Provider | Host OS | Arch | | ---------------------------------- | ------------- | ----- | | [VirtualBox](README-vbox.md) | Linux | amd64 | | [libvirt](README-libvirt.md) | Linux | amd64 | | [VMware](README-vmware.md) | Mac | arm | | [VMware](README-vmware-linux.md) | Linux | amd64 | | [Parallels](README-parallels.md) | Mac | arm |

localrc-template ================

Most providers will recommend you set some environ vars via your localrc

1. cp localrc-template localrc 1. vi localrc

Additionally chef provisioning exposes some optional configuration for the vm, see localrc-template.

1. source localrc 1. vagrant provision 1. vagrant up 1. vagrant ssh 1. rebuildswift

running-tests =============

You should be able to run most tests without too much fuss once SSH'ed into the VM.

1. .unittests 1. .functests 1. .probetests 1. vtox -e pep8 1. vtox -e py38 1. vtox # run all gate checks

s3cmd =====

You know you want to play with s3api, we got you covered.

vagrant ssh
s3cmd mb s3://s3test
s3cmd ls

configure statsd_exporter/prometheus metrics ============================================

You should be able to optionally configure statsd_exporter/prometheus metrics for the Swift stack on the VM.

cp localrc-template localrc
sed -i 's/^\(export STATSD_EXPORTER=\)\([^ ]*\) /\1true/g' localrc
source localrc
vagrant provision

These will expose /metrics endpoints on ports 9100-9105 which you can check directly, and configure prometheus to scrape these endpoints every 10s and retain data for up to a day; you can then create ad-hoc graphs at

• http://saio:9090/graph

ninja-dev-tricks ================

You should add the configured IP from your localrc to your /etc/hosts or use the default:

sudo bash -c 'echo "192.168.8.80 saio" >> /etc/hosts'

Then you can easily share snippets that talk to network services running in your Swift-All-In-One from your host!

curl -s http://saio:8080/info | python -m json.tool

A few scripts are available to make your dev life easier.

1. vagrant up --provision will bring up your VM in working order (useful when your VM is halted) 1. source localrc; vagrant provision on your host to push the new Chef bits in place (useful if you change localrc) 1. rebuildswift to reapply everything like it would be at the end of Chef time (useful to revert local config changes) 1. resetswift will wipe the drives and leave any local config changes in place (useful just to clean out Swift data) 1. reinstallswift will make sure all of the bin scripts are installed correctly and restart the main swift processes (useful if you change branches) 1. autodoc [swift|swiftclient] will build the sphinx docs and watch files for changes, and upload them to a public container on your vm so you can review them as you edit 1. vtox will hack the local tox.ini and setup.py so you can run tox tests successfully on the swift repo in the /vagrant directory 1. reec will rebuild/reinstall all the liberasure/pyeclib[/isa-l] bits! 1. venv py37 will make sure your tox virtualenv is ready and let you py3