On IT Operations and Infrastructures

self documented agile infrastructure

Gildas LE NADAN — Mon, 02 Mar 2009 14:13:00 +0100

In my latest position, as an IT Operations Manager I was confronted to the classic problems of a non-mature Operations: We were understaffed, in a fire-fighting mode, there was poor documentation (either missing or not up-to-date, often misleading), almost no backup, and the team members had almost no overlap in their skillsets and were demotivated.

I couldn't afford to lose a single person of my team as the knowledge lost would be dire for the company, and to make things even more complicated, our CEO wanted us to be able to deploy our home made software to remote client sites.

On the good side, one of my team member had an excellent knowledge of the home made software, another was a good perl developer, there was a good knowledge of Suse, rpm packaging and they already had a set up a subversion repository and a basic puppet setup.

To consolidate the knowledge and move away from manual operations, it was decided to use svn, puppet, Suse and pxe to build a self-documented agile infrastructure where anyone would be able to deploy new services.

The basic blocks

The applications was packaged using rpm and the latest valid version stored on a file server, but all the configuration files (including those needed to build the packages) were stored in subversion.

This way, it was possible to keep track of the changes (who, why) while at the same time having a way to retrieve the latest valid version using a simple 'svn co'. The svn commits were sent to all team members, so it kept everyone informed of what was going on.

The recipes

The services and server setup were described in puppet and stored in subversion. The services were described in a generic manner using templates as configuration files so you could instantiate a new service by deploying the needed rpms and creating "on the fly" the configuration files adapted to that specific instance. The important idea was that no manual operation was needed to deploy a new service thus allowing it to be perfectly reproductible.

Thanks to this solution, one could easily deploy a new instance of a service on either a physical or virtual machine. As we were in a j2ee world with a multi-tiered application, you could either stack several services on a machine (for development or testing for instance) or one service per machine, depending on your needs.

The nice side effect is that puppet is the live documentation of your systems as it defines and enforces the active configurations! Since the puppet files are also stored in svn, it is possible to see all the changes for a file through time with the associated comments.

The drawback of the system is that extreme care must be taken not to manually tamper with the configuration of the servers: everything MUST go through puppet, and the comments must be kept relevant.

The deployment system

The machines could be either physical or virtual machines, and pxe combined with kickstart is used to deploy a basic setup consisting of a basic Suse + puppet. Of course the kickstart files are stored in svn. Once the server is deployed, puppet can then populate the server with a set of services/configuration.

The backup server

Since a service/server could be easily reinstalled using this solution, there was no need to backup them which is a big time and tape saver.

This way you can concentrate on saving your application data, that is your production dataset as well as the files on the file server and the subversion repository.

In our setup, it was decided to sync the subversion repository and the files stored on the fileserver between 2 sites. Also, thanks to the use of subversion, everyone in the team had the files on their own machine.

Disaster recovery

During the implementation, cross-dependencies between the subversion, installation, puppet, file and backup servers were considered in order to allow a complete restoration of the infrastructure, provided that we had access to the backup tapes and could reinstall the backup server manually using a Suse install media.

It was decided that the subversion, file, build and installation services would be installed on a single machine. From there, you could reinstall the puppet server via a very limited set of operations that were documented with care (basically, installing the packages and checking out the svn repository).

Once this is done, and provided all your infrastructure is described using puppet recipes, you can easily repopulate your servers in a case of disaster recovery, but it could also be used to install everything on a remote site, provided you have a machine were you can bootstrap your infrastructure.

On the Shortcomings Of Systems and Networks Engineers Training

Gildas LE NADAN — Fri, 16 Jan 2009 20:29:00 +0100

As far I know, there is no course to become a Systems and Networks Engineer, aside from courses to learn (and gain certification in) a given vendor's product. In fact, back in my university years, I remember that my teachers seemed to assume that there was no interest in this kind of thing as learning the options and caveats of a particular product was all you needed. In their eyes, algorithmic and development approaches (RAD and OO at the time) were where the real focus lay.

In my case, the situation might have been worsened by the traditional friction in France between university (were the "real, pure, academic" research is done) and the Ecoles d'Ingénieur (where you learn about engineering and sometimes conduct "applied research"), but I'm not so sure the situation would have been so different in an engineering school or another country (I'll be interested in your feedback there to prove me wrong!).

So, how does one becomes a Systems and Networks Engineer? Well, it's easy, you learn by yourself, usually starting with a small set of machines and mainly by a trial-and-error approach. If you're lucky enough, you might benefit from someone else's experience and coaching. But still, it remains mostly an ad-hoc approach.

Of course, you quickly learn to avoid tinkering with the production platform on a Friday evening, and given enough experience you can even begin to "guesstimate" - to a greater or lesser degree of accuracy - the impact of such-and-such a modification, then hopefully the number of systems you manage will increase until eventually you find out the hard way that complexity doesn't grow linearly with the number of systems.

I would even claim that given the chance to work with different environments and large scale platforms (highly available, highly loaded web platforms; HPC clusters; heterogeneous banking environments), one might infer common rules of thumb and even have the hubris to try to find a meaning in the chaos.

The fact, however, is that I believe this ad-hoc approach to learning the job and the lack of (field proven) best-practice references to be The Source Of All Evil.

First of all, from this learning process comes an approach comprising unproven beliefs, mythology or carved-in-stone rules ("one needs twice the amount of ram as swap space"). It also makes it difficult to assess someone's ability as a Systems and Networks Engineer if not by considering her technical knowledge/certifications or previous experience in a similar position.

Secondly, the good practice of "not changing what works" forged by the trial-and-error approach, tends to encourage cruft accumulation and creates a certain reluctance to change anything at all. As a result risk-mitigation approaches such as continuous integration and minor steps are replaced by "big-bang" style changes with increased risks of failures.

All in all, I believe that it has created a situation whereby IT Operations is working against the (in my eyes desirable) goal of becoming agile and business-oriented - a true competition differentiator and not just a "cost center" working in firefighting mode.

The "cost center" aspect has motivated the few approaches trying to address the lack of maturity in IT Operations: ITIL, Cobit and so on. To the best of my knowledge, they are all process-oriented and mostly address the problem from a financial perspective (ROI, risk management).

While I believe there are interesting ideas in all of them, and that cost is an important factor in the need - solution equation, I am not too convinced by the "process" approach which limits risk but adds weight and inertia to the organisation and kills pleasure and innovation. I confess I might be too influenced by the ideas of the Agile Manifesto here, but I can't stop myself thinking that neither Google nor Facebook used ITIL to get where they are.

I also find them too complicated to be real enablers and believe that even though they warn against it, they incite dogmatism where pragmatism should rule. Because of this, I think they fight against the exact goals they are trying to achieve.

So how can we get out of this mess?

We would definitely benefit from an increase in interest from the academic world towards IT Operations and Infrastructure realities. Consider Google's study on Hard Drives failures. Before its publication different people had wildly differing beliefs about disk failures based on factors such as: their own experience with a statistically-insignificant sample size of drives; manufacturer advertising (propaganda); luck. With a large scale, scientific study to turn to, people gained a much better understanding of the subject matter.

Naturally, courses about availability, scalability, large scale systems and networks design and management would be welcome in Universities.

But successful companies such as Google or Amazon couldn't have emerged without good IT engineering practices and a sound infrastructure (after all Amazon even sells its services now via EC2 and S3!), so, it is certainly possible today to build an IT infrastructure that makes a difference.

Then we definitely have the responsibility to learn from those leaders and spread that information around if we want IT Operations and Infrastructures to mature and serve the business and our own users (kudos here to websites such as High Scalability or Storage Mojo for their excellent work).

Undoubtedly most of the technologies those companies use to manage their infrastructures are purpose-built in-house developments that won't be published, so we as a community need to build the tools we need in the same way developers have started open-source re-implementations of well known building blocks such as MapReduce for instance hadoop.

Tools such as Luke Kanies' Puppet configuration management, rapid deployments tools such as openqrm or easily adaptable and scalable monitoring systems such as hobbit (now renamed Xymon) should be endemic to our infrastructures, yet they are sadly too often an exception.

Yet Another Blog?

Gildas LE NADAN — Tue, 13 Jan 2009 14:53:00 +0100

Hello there!

In this introduction post I will try to explain why on earth I've started Yet Another Blog.

For years now I've exchanged ideas about IT Infrastructure and Operations with my colleagues and friends, be they IT Ops guys or dev dudes (or even from a completely different background). I've learned a lot from those discussions and I believe my work has matured as a result.

Lately though, this flow of communication has dried up for several reasons and I've grown frustrated about it, hence the idea of this blog. Hopefully it will allow for fruitful interaction with people I know and indeed others that I don't know. People with whom I am impatient to share ideas and experience!

So, welcome aboard!

Gildas