Aside from the hypervisor (ESXi) and vCenter, one of VMware’s longer serving Enterprise products is Site Recovery Manager (SRM). Originally launched in June 2008 as a version 1.0 product, SRM promised to make the act of failing a virtual infrastructure over to a Disaster Recovery (DR) site simpler. To sum its functionality up in the simplest way possible SRM was “automation for DR”.
DR has always been a tricky topic from an IT perspective. A bit like insurance for your car, you hope that you never need to use it. Sometimes it’s also possible to delude yourself into thinking that it’ll never happen to you.
By failing to prepare, you are preparing to fail.
The quote above, or versions of it, have been attributed to many different people over the years. I’m going with Benjamin Franklin. The point though is that DR planning is something that it’s better to in advance than when you really need it. SRM then is a tool that can help simplify and automate that planning.
That’s not to say that it’s the only tool, the only way or the panacea of DR. VMware have worked hard through product updates and versions 4.0 / 4.1 to make SRM fit as many scenarios as seamlessly as possible but ultimately there will be limitations and any SRM based DR solution must be properly designed and supported. Version 5.0 adds exciting new features like automated failback, VM dependencies and vSphere Replication and makes SRM very compelling indeed.
With all that said though, my experience with it since it was first released has been fairly light. Many of the environments that I have worked in have had alternate methods of implementing DR. Not all of the customers that I’ve worked with could have used SRM effectively at the time. Many of the others did not want it. Most of the time I have had to content myself with setting it up in my own lab. Much like the point of the product itself I have effectively been preparing for a time when I might need it in anger.
One man who has been through it all though is Mike Laverick, author of the book that this review is all about – Administering VMware SIte Recovery Manager 5.0. Mike quite literally wrote the book on SRM. Back in 2008 he created and self-published a book corresponding to version 1.0. In 2010 he struck again with another self-published book for version 4.0 (I think you can still get it through his site). The version 5.0 book however marks a slight departure though and is professionally published as the very first title from newly formed VMware Press.
I received a copy of the book a few weeks ago (disclaimer: directly from VMware Press themselves – thank you – but they’re not paying me to do this review, these are all my own opinions) and I was pleased that it had the look and feel of a book that belonged in a book shop and alongside the many other titles that I have stashed away on my shelf. The typeface and screenshots were all clear and legible, the index and contents lists easy to navigate too. All good so far.
Having read both of Mike’s previous SRM books, I thought that I knew what to expect. I was pleasantly surprised therefore by the flow of this book. Don’t get me wrong, I’m not criticizing his previous efforts at all but this new book seemed to hang together better in my opinion. The opening chapter – dealing with what SRM is, what’s new in version 5.0, DR and storage replication principles – made great reading material whilst I was waiting at a doctor’s surgery for my appointment.
The following five chapters cover the basics of storage replication on some of the more common storage vendor platforms (Dell, EMC, HP and NetApp). It’s not possible to cover all of the storage vendor replication technologies in detail in one book. That would take too much time and waste a lot of space. At the same time though, those big four are going to cover a large percentage of the people using SRM. Plus, and more importantly, the principles involved are at least going to be similar with other storage vendors and they are essential to the operation and use of SRM. It’s a fine line to tread and I can’t imagine that adding more or taking anything away would improve the book.
Installation of SRM is comfortably covered after that and then comes one of the new features – vSphere Replication. Personally I felt that this chapter was a little light for such an important new features. However that is possibly due to VMware’s engineering efforts making it simple to implement and use. It might have been good to include some basic troubleshooting tips for vSphere Replication although you could equally argue that they might be better placed in KB articles or blog posts.
The remainder of the book is dedicated to the configuration of protected and recovery sites and the creation / configuration of SRM’s plans etc. It is this section that many will buy the book for and it’s obvious that Mike knows his stuff. Indeed, in conversation with him last year I know that he had been spending a lot of time with VMware SEs and Engineers learning about SRM 5.0 himself as it was still being developed and tested.
I would happily recommend this book to anyone who uses or is thinking of using SRM, there will probably not be a better resource. I would also recommend this book to anyone looking at DR more generally as many of the concepts and issues related to the topic are touched on or covered in some detail and even if you’re not thinking of using SRM, it’s useful to have another perspective.
A great first effort by VMware Press and another good book from Mike. I’m looking forward to the next volumes from both of them.
Administering VMware Site Recovery Manager 5.0 is published by VMware Press and available from any good bookstore in paperback and electronic formats.
Michael Poore is a Senior Consultant for Virtual Clarity, a small virtualisation / cloud consultancy based in London and San Francisco. Michael works on all aspects of datacenter virtualisation, automation, orchestration and management for various global companies. He started the vSpecialist blog in 2008 and convinced co-author Jeremy Bowman to join in over a beer a while later.