leifmadsen

ruby_block "Get starting values" do
  block do
    record = File.open("#{Chef::Config[:file_cache_path]}/#{db}-status.txt") {|f| f.readline}
    node.set['mysql']['file_value'] = record.split("\t").first
    node.set['mysql']['pos_value'] = record.split("\t")[1]
    Chef::Log.info("Current values for file_value and pos_value are: #{node['mysql']['file_value']} and #{node['mysql']['pos_v#
  end
  action :create
end

leifmadsen

# vim: set ft=ansible
- name: Deploy on OpenStack
  hosts: localhost
  gather_facts: false
  vars_files:
    - ~/.ansible/vars/elk_vars.yml

  tasks:
    - name: Deploy an instance
      os_server:

leifmadsen

Deploy SA Telemetry

A brief bit of information on how I've deployed OpenShift Ansible from the release-3.9 branch with GlusterFS storage, service brokers including Ansible Service Broker.

Permissions

oc login -u system:admin
oc create namespace sa

oc project sa

leifmadsen

Telemetry Post-Install Actions

Some actions that might be done post-install of telemetry to setup the system.

add user to htpasswd

htpasswd -bn lmadsen Sup3r$3creT
<copy_hash>
sudo su -

cat >> /etc/origin/master/htpasswd <

leifmadsen

Resizing CentOS Atomic /dev/atomicos/root disk

I found out that the default Atomic root disk is only 3GB! That causes all sorts of problems on the master when you try to spin up more than a handful of things (you get disk pressure errors, and things end up not spinning up properly).

NOTE

You'll need to make sure you have libguestfs-xfs package installed or virt-resize won't be able to expand the XFS partition.

leifmadsen

Deployment Commands

VM spinup

ir virsh --host-address localhost --host-key ~/.ssh/id_rsa --topology-nodes "undercloud:1,controller:1" --topology-network bridged_undercloud --disk-pool /home/images/infrared --image-url https://cloud.centos.org/centos/7/images/CentOS-7-x86_64-GenericCloud.qcow2

Files

undercloud.conf

leifmadsen

Telemetry POC Deployment

DEPRECATED

Documentation is now available at https://github.com/redhat-nfvpe/telemetry-framework/tree/master/docs

• Pre-Requisites and Container Building
• Telemetry Platform Server Side Installation
• Telemetry Platform Client Side Installation

leifmadsen

DEPRECATED

Documentation is now available at https://github.com/redhat-nfvpe/telemetry-framework/tree/master/docs

Pre-requisites Overview

Prior to deploying the telemetry platform, there is some pre-requisite builds that need to happen. Items that need building include:

• barometer-collectd container (with patches)

leifmadsen

DEPRECATED

Documentation is now available at https://github.com/redhat-nfvpe/telemetry-framework/tree/master/docs

Telemetry Platform Deployment

Instructions in this gist are for the deployment of the telemetry platform on top of OpenShift Origin v3.9. Deployment of the platform is done in two (2) steps. The first step is bootstrap of the virtual hosts and installation of the virtual machines (VM) on those virtual hosts (virthost).

leifmadsen

DEPRECATED

Documentation is now available at https://github.com/redhat-nfvpe/telemetry-framework/tree/master/docs

Installation of the client will result in two containers running, along with two folders installed on the host system which provides the configuration to the Docker containers.

Containers being installed and executing will be barometer and qdr for the Barometer collectd and the local QPID Dispatch Router.