Recently in development Category

Some top picks from the last few days:A good overview and comparison of various Virtualization technologies from the ever interesting Kris Buytaert.The High Scalability blog pulls together a list of places that are discussing the state of the art in scaling in the cloud. Not entirely un-related is High Scalability’s overview of the Second Life Architecture. 12Gbit/sec in 2007 and growing huh?The Bend in the Weather blog covers how to setup your networking in VirtualBox.Enjoy! Comments or email always welcome.

What technology platforms do the big boys run (Digg, ebay, Amazon and YouTube)?What architectures do they use?What tips and lessons can they share?twit88.com has the details.

We now have a public (read only) git repository.

Currently containing the vast total of our Munin plugin template and a work-in-progress plugin for timing how long urls take to download.

git clone http://sysmonblog.co.uk/git/ [your dir name]

Access via http is read only. We’d be delighted to accept patches either by email or via comments below.Those new to git might be interested in our git by example tutorial.

We’ve recently been learning and using git, the version control or distributed source code management tool, similar to Mercurial or Bazaar.

There’s some really good documentation out there already, however we wanted a simple step-by-step tutorial that guided you through learning git without cluttering the process by explaining a whole bunch of theoretical jargon or concepts first.

Git by Example

One of the annoyances we find with using Munin is writing plugins. It’s not hard or complicated per se (or any worse than with any of the other monitoring tools). However it often seems confusing and fiddly to our little brains. We’ll admit we aren’t Munin Experts, nor want to be.

The implementation details we do know tend to leak out of our heads over time as, although we use the reports frequently, we typically don’t fiddle with the guts of Munin very often. Which is perhaps a good thing.

Yet from time to time we need to tinker and it is mostly with the plugins.

Kris Buytaert and Tom De Cooman have posted the slides to their OLS 2008 talk:  “Systems Monitoring Shootout”.

For anyone evaluating monitoring solutions, it is a great read. They’ve obviously done a lot of research and if nothing else highlight a huge bunch of issues and topics you would want to consider when selecting a product.

Interesting article from Edd Dumbill :

… the cost of infrastructure has dropped … other costs remain high … System administrators … demand high wages. Commercial software license fees spiral out of control … power is already troubling large companies …Help is at hand … If they don’t yet make massive resource management trivial, they at least make it possible.

In short, if you want to be ahead of the curve, you want to be ahead of the curve with these technologies:

• Distributed revision control systems
• Virtualization
• Configuration management

… because they let you do more with less. Pleasingly, we’re already making heavy use of the first two and had started scratching our heads about the last one.

There is a tendency sometimes to think that system monitoring tools like munin, zenoss, zabbix or nagios are mainly for the sysadmin. We think the people who can benefit from monitoring is considerably bigger.