This is an internal documentation. There is a good chance you’re looking for something else. See Disclaimer.

Storage Usage as CSV

This document describes an Ansible playbook that generates a CSV file containing detailed information about the storage used by a customer/installation.

About the Generated CSV

The information is provided as CSV and contains these columns:

Column	Description
Customer	Customer to which the installation belongs.
Installation	Installation name.
DB Size	Size of the database as reported by Postgres. This corresponds to the actual storage used on disk and may include dead rows, indexes other internally used data.
History DB Size	Size of the history DB as reported by Postgres.
S3 Size	Size of the S3 bucket as reported by Cloudscale. All installations of a customer share a bucket. Thus, this size is only listed on one, the first, installation of a customer..
Solr Size	Size of the Solr search index.
Total	Space used by an installation. S3 storage is included for the first installation of a customer only.
Total Customer	Total storage used by a customer.

All number are provided in GiB, rounded down.

Generating the CSV

Hint

This documentation assumes you’ve setup Ansible already. See Setup Ansible.

Generate CSV:

cd ${ANSIBLE_REPO}/tocco
ansible-playbook playbooks/storage_usage.yml -e output=data.csv -f 50

This stores the data in playbooks/data.csv.

You can generate a report for a limited number of customers using -l like this: -l customer_agogis,customer_bk.

Hint

-f 50 is a precaution to limit the number of simultaneous requests. 50 is likely fine but in case you hit a ratelimit, lower it.