Skip to content

Instantly share code, notes, and snippets.

View mreider's full-sized avatar

Matt Reider mreider

View GitHub Profile
#!/bin/bash
set -e
function welcome {
printf "\nThis script is necessary in Ops Man early alpha.\n"
printf "First let's check some things...\n"
check_things
ask_and_do $need_stemcell "Wrong Stemcell. Fix it?" rename_stemcell
ask_and_do $need_bosh_pem "Need BOSH keypair in AWS (non-destructive). Ok?" create_keypair
---
name: redis
director_uuid: 6b29bd9c-dea0-48d2-bfcd-969722e1e4dc
releases:
- name: redis
version: latest
resource_pools:
- name: vms
@mreider
mreider / redis.yml
Created May 27, 2015 15:33
Redis Community BOSH Release Product Tile
name: redis
metadata_version: '1.4'
product_version: 9.1.0.0
rank: 10
import string
def analyze(file):
count = {}
remove = dict.fromkeys(map(ord, string.punctuation))
f = open(file ,'r')
f = f.read().lower()
f = f.translate(remove)
wordList = f.split()
for word in wordList:
TL;DR: We are not going to ship the "split release" of Ops Manager. We are going to ship Ops Manager for Azure, GCP, and Photon separately, and work on UX for BOSH so we can support BOSH as an official Pivotal product.
As you may know, we had this plan to split BOSH out of Ops Manager to ship on Azure, GCP, and Photon.
A couple of months ago I started arguing that this was unnecessary. This has resulted in a change of plan.
My opinion was that Pivotal should productize BOSH, add some usability features, provide BOSH support via Zendesk, and invest in documentation and branding for BOSH. At the same time, I felt it was important to shift Ops Manager into something of a read-only dashboard for BOSH deployments.
My views stem from constant customer feedback about how Ops Manager misses the mark. Ops Manager is hard to CI / CD over BOSH. It was never built with a pipeline in mind. Moreover, Ops Manager is and will always be constrained when compared to using BOSH. It will never catch up with BOSH's flexibili