Very quick as I did not find any good solutions to this on Google. This is actually an interim post as I ran across this while configuring Sensu in my lab. A full post on that, along with configuring it with my Puppet set up is coming. RabbitMQ is in EPEL (slightly old, but not drastically) so install that first, then install from yum. yum -y install https://mirrors.ircam.fr/pub/fedora/epel/7/x86_64/e/epel-release-7-2.noarch.rpm yum -y install rabbitmq-server Well that was easy, so just start it with

Hyper-converged seems to be all the rage at the moment. VMware's announcement of the EVO:RAIL has naturally got lots of tongues wagging. They are jumping into a market already well populated. I was looking at the pricing/features and though it all looks a little expensive. Nutanix for example, will sell you the following: NX-1000: $80k+ (4 little nodes) NX-3000: $144k+ (4 big nodes) NX-6000: $120k+ (2 massive nodes) I’d imagine that, once you include licensing, EVO:RAIL systems will come in similar.

Despite my penchant for tools written in Ruby (Puppet, Gitlab, Jekyll/Octopress etc) I do not actually like Ruby. I am more of a Python guy. I also like Vim, so whenever I use a GUI IDE I end up with something littered with :w and ZZ. Despite my pythonic leanings, I also need something that can handle Ruby and Puppet’s DSL. To which end, this is a bit of a mixture.

Fedora do not bundle Microsoft’s core Truetype fonts for licensing reasons. Normallly I do not care, personally I prefer Liberation fonts anyway. However, today I needed Verdana. Traditionally, the way to install these on RPM based distributions has been: Grab the RPM spec file Build an RPM from the spec file Install RPM using the rpm command. All well and good, however there are a couple of problems. Using RPM directly is frowned upon Nowadays, Yum does various bits of house keeping in addition to RPM, so this can lead to the rpm and yum databases getting their knickers in a twist.

With a superb buzzword laden title like that, then I reckon massive traffic boost is inevitable. Puppet is my favourite Configuration Management tool. This is not a post to try and persuade anyone not to use Ansible, Chef or any other. What I want to do is show I build Puppet based infrastuctures in such away that it meets all the basic tenets of DevOps/Agile/buzzword-of-the-month. What to we need: CentOS 6 - RHEL/CentOS is pretty much the defacto enterprise distro.

This used to be quite complex, but now is astoundingly simple. Now there is a new project call realmd. It is in recent version of Debian (Jessie and Sid) and Ubuntu (since 13.04). For Red Hat types, it is RHEL7 and Fedora (since 18). If you're on Debian/Ubuntu, install with: apt-get install realmd For RHEL/Fedora: sudo yum install realmd Now you can go ahead and join the domain: sudo realm join --user=<admin-user> example.

A break from computing today and into the world of nutrition. Cyclists love to talk about nutrition as the nature of our sport makes it a major consideration. I do not really know of any other sport where your fuel gives out before the rest of your body. This means that cycling nutrition is big business, and expensive. A typical box of energy bars will cost about 1 euro a bar from a big box pusher, more from your LBS.

The current project (as of this week) has me moving away from Openstack for a while. For the next couple of months I will be immersing myself in monitor, metrics and logging. Naturally, this being a shiney new environment, this involves a significant amount of VMware time. I have inherited an Icinga install running on Ubuntu Server, so I will be needing to run CLI commands to create checks. Simply runnning the installer does not work, as the vmware-cli package is a mixture of 32 and 64 bit commands.

For the last few weeks I have been consulting on a private cloud project for a local company. Unsurprisingly this has been based around the typical Openstack setup. Nova - KVM Neutron - Openvswitch Cinder - LVM Glance - local files My architecture is nothing out of the ordinary. A pair of hosts each with 2 networks that look something like this: All this is configured using Red Hat RDO. I had done all this under both Grizzly and, using RDO, it was 30 minutes to set up.

It’s been a while since I last posted anything, but it is time to. I’ve been playing around a lot with various tools for gathering information about my environment recently. One of the most important tools for storing that information is decent logging. Syslog is proven and solid, but a little creaky. For storing everything it is fine, but getting anything out is not so great. Logstash is an awesome tool written by Jordan Sissel that is used to “collect logs, parse them, and store them for later use (like, for searching)”.