When Planning Disaster Recovery, Don’t Forget the Small Stuff

As solid and reliable as fishtanks and rifles are, they share a common weakness. The tiniest crack in either will eventually destroy the entire product. Why did I choose “fishtanks and rifles”? Lori and I’s hobbies include those two items, but there are other things the same applies to. My oldest son said his tom-tom had a crack at the bottom, and I made the same observation to him… Best to get that fixed now, since the point of a drum set is to beat on it and make vibrations, I can guarantee what will eventually happen if nothing is done about that crack.

A small crack in the barrel of a rifle is likely to become dangerous rather quickly. When you think of danger from rifles, you don’t generally think of danger to the operator, but in this case it is. Or those nearby. This is more of an issue for someone like myself who collects older rifles than for the guy that’s buying brand new rifles with tungsten barrels and chrome lined chambers, but it is a truth of all rifles, the pressure used to expel a bullet is far too much for a crack to remain a crack. The same is true of fishtanks. It does not matter how many tens of thousands of dollars you’ve invested into your aquarium setup, glass (or plexiglass) with gallons and gallons of water pushing on it will eventually give way around a crack – there’s just too much pressure to contain the problem without repairs. And repairs generally include emptying the tank, which is a hazard-prone operation for all of the beings you’re keeping in it.

 

When planning your disaster recovery, you need to identify the places that are going to be under pressure in the event of an emergency and make certain they are solid enough to handle the increased pressure that downtime and recovery from downtime will create. There are several things that spring to mind, some of them you no doubt have covered, some, if the buzz out there in the enterprise world is to be trusted, you do not.

While we could go through the whole monster list, let’s just stick to one that lots of organizations seem to miss or put as a low priority – the backup WAN connection. Some of you don’t have one, and that really is planning for outages, as long as your organization is good with that, you’re probably fine. Most organizations are only good with it until they have to pay the piper, then there is a scramble to make sure that never happens again. Eventually, be it from a router misconfiguration or a bad piece of hardware, or sun spots (sorry, been reading BoFH), your primary vendor will go down. Having a backup is insurance against that day.

But having a backup WAN connection is also a lot of work, and that’s where most of you have found yourself hovering. Merely putting in a backup WAN connection is a start, but you have to configure failover, test the configuration, etc. And then you need to revisit it on occasion.

The one thing that we’re going to see more of in the future is sizing issues. When the volume of data that you are pumping through your primary WAN connection is on the rise, you have to make certain that your backup connection can keep up with requirements should it be required. There are of course a variety of ways to go about this, with getting a larger connection, making sure your WAN Optimization products switch over when the connection does, and defining which services can be cut should there be an outage on your primary connection being the top three.

My new IBM Hardware, as mentioned

on the InfoSmack podcast

Not selling you anything, just pointing out that you need to keep your eye on all of the pieces. When your primary WAN connection goes down because someone dug through a wire, all that money you paid to set up a backup WAN connection is either going to pay off or not. Since you’re paying for it, I  thought I’d give you a friendly reminder to review your architecture and make certain you have everything set to get the benefit of the connection when it is needed.