Skip to content

Latest commit

 

History

History
executable file
·
84 lines (62 loc) · 4.55 KB

README.md

File metadata and controls

executable file
·
84 lines (62 loc) · 4.55 KB
Raw

pcf-usage

The goal is to collect information from a PCF foundation and assess the value by leveraging the 5s framework (Speed, Scalability, Stability, Security and Saving) and align with business outcomes (developer productivity, Operational Efficiency, Day 2 Ops, Infrastructure Savings, Innovation, Cloud first).

We have 2 components:

  • pcfusage.sh - script to capture foundation information into a json file
  • jupyter notebook - to help analyse the data

Running the script

The first step is running an script that uses the CF API to collect data in a json format. The scripts uses the cf curl command and requires you be logged in as admin. The script uses jq to create an output that can be process by a jupyter notebook.

$ ./pcfusage.sh dev

Where 'dev' is the prefix used to identified the foundation we'll be collecting data

In the samples folder of this repo you can find an example output.

Create tackle import

The script can creat a konveyor tackle import file (https://github.com/konveyor/tackle) from the collected data.

Usage: pcfusage <PREFIX> <CMD>[ALL|APPS|SRVS|TACKLE] <STAGE>[dev]
  
  where: CMD=ALL    - all foundation information
         CMD=APPS   - Apps in CSV format - REQUIRED THE OUTPUT OF 'ALL' RUN
         CMD=SRVS   - service bindings - REQUIRED THE OUTPUT OF 'ALL' RUN
         CMD=TACKLE_APP - Apps in Tackle CSV format (https://www.konveyor.io/tackle) - REQUIRED THE OUTPUT OF 'ALL' RUN
         CMD=TACKLE_ORG - Apps in Tackle CSV format, one CF organisation is treated as one application (https://www.konveyor.io/tackle) - REQUIRED THE OUTPUT OF 'ALL' RUN

Examples:
  pcfusage dev - defaults to ALL
  pcfusage dev apps - creates apps csv file
  pcfusage dev srvs - creates a file with the app bindings guids
  pcfusage dev tackle - creates apps Tackle csv file, stage defaults to 'dev'

API usage

Below is a list of current CF API used by the script. In some cases, we handle the paging for API calls that can return a large number of elements.

  • /v2/apps
  • /v2/users
  • /v2/organizations
  • /v2/spaces
  • /v2/service_brokers
  • /v2/service_instances

Using the Jupyter Notebook

You'll need to install Jupyter and open the notebook in this repo. As per the instructions on the Jupyter Notebook website, Anaconda is the easiest way to install Jupyter Notebook, but any working installation with the proper Python libraries should work. To open the Jupyter Notebook, run the following command from the root of this repository:

$ jupyter notebook pcf_foundation.ipynb

This command should launch your web browser with the Jupyter Notebook loaded.

In the very first cell we load the file you captured with the pcfusage.sh script. There is some basic metadata information that is not currently on the json file that you need to update. Below is an example loading the borgescloud_foundation.json file. It's important to add the capture date so we can calculate the number of days since the application was last updated and the information on the DIEGO CELLS we use for calculating infrastructure utilization.

file = "/Users/mborges/Tools/PCF/scripts/borgescloud_foundation.json"
capture_date = datetime.datetime(2018, 6, 26, 0, 0)
diego_cell = {"number_of": 4, "vcpu": 4, "ram_gb": 32, "disk_gb": 32, "operators": 2 }

The diego_cell structure provides metadata for the foundation, hence the operators field we'll use for calculation operation efficiency like ops to apps or ops to containers ratios

In the first cell we create the dataframes that are used in the following cells. Each cell tries to look at the data from the 5s framework.

<<<<<<< HEAD

Exporting notebooks

jupyter can convert notebooks to various formats to make it easier to share. We're exploring PDF, HTML and even reveal.js

mactex

jupyter nbconvert --to PDF --template hidecode pcf_foundation.ipynb

We have a table that is too wide and the PDF generation is cutting it out. Creating PDF in landscape orientation then run pdftext to convert to PDF. I notice the PDF bookmarks when converting directly to PDF were gone.

8e64c05483b3bd9c13a855ff45b82ad4a43d3b56