Why is it always on a Friday? (Part 1)

The weekend is tomorrow; you’ve been head-down all week, nose to the grindstone making sure that your people and your apps are doing well , everything is running smoothly and your end-users are experiencing great service. Now is the time to go home, make sure your wife is OK and to “monitor” your family. Suddenly, on Friday afternoon you get the call – something is wrong… and it’s not one of the “easy ones” (they never are on Friday): you’re calling a war room. REALLY!?!?

It’s called the War Room for good reason – that’s just what it is. You’ve been attacked, and you need to fire back with everything you’ve got. The problem can be one of 100 different issues – latencies are cropping up in your transactions, end users are starting to get timed out, and are growing more and more frustrated. The pressure is on to solve the problem immediately. What’s driving you crazy is that all the screens are showing green! Nothing is indicating a problem anywhere along the pipe, so there’s no one to blame. DBA’s, Network, System Admin, app owners – they’re all happy with their dashboards and more than happy to show you. And that’s great for them, but with end-user transactions timing out, what’s to stop your customer from going straight to the competition?

In fact, end user experience monitoring would show that your users are just plain fed up. You and your teams need to figure out what, exactly, is going wrong.

IT can be a needy bunch. They work hard, and they’re stuck behind the scenes where they get very little accolade (thank God I at least have this blog to complain!). If things go well, end users never even think of them. In fact, the only time they get any attention at all is when things start to fall apart – and that’s just the way it should be.

So like children, your IT team is suddenly all over you. Each person is trying to prove that on their end, everything is fine. It’s green across the board. But you know that somewhere in between those green lights, there is a pile-up.

What are you going to do in this situation?

How do you save the operation from days of latency and lost customers?