openstack-instances-monitoring

OpenStack Instances Monitor is a simple project to show how much each instance in your project is consuming on your hypervisors.

Installing the tool

• clone this repository

$ git clone git@github.com:renato-farias/openstack-instances-monitoring.git

• installing all dependecies (OS)

$ cd openstack-instances-monitoring
$ setup/setup.sh

• installing all dependecies (Python Application)

$ cd openstack-instances-monitoring
$ mkvirtualenv monitoring
$ workon monitoring
$ pip install -r setup/requirements.txt

• change the setting and run the app

$ cd openstack-instances-monitoring/app
$ mv application_sample.yaml application.yaml
$ gunicorn -b 0.0.0.0:8000 app:app --log-level info --env APP_ENV='production' --reload -w 4 --timeout 60

• point your browser to http://SERVER:8000/
• In your first access access: http://SERVER:8000/renew - To make the initial cache.
• insert in the root's cron the following entry (It's responsible to collect data from openstack API avoiding to take too time to get some informations about all instances):

1 * * * * curl http://localhost:8000/renew

Installing the collector on the Hypervisors

• Get the script

# wget -c http://SERVER:8000/scripts/monit_instances.sh

• Edit the script pointing the server IP or Name to send data.
• insert in the root's cron the following entry:

* * * * * /root/monit_instances.sh

Customization

It's a tool with a simple endpoit API to collect data. Feel free using your imagination to customize this.

How it looks

I'd like to say thank you to Crapworks. I used his template from Ceph-Dash to turn easy my job =]