Wednesday, May 28, 2008

Why are we still fighting fires?

"We have spent so much money on monitoring tools and consulting, why are we still fighting fires?" I heard this from a recent prospect. The simple answer is because we live in a complex and high demand world. As IT proffesionals we are trying to do a lot with a little; little training, little vision, little strategy, little focus. Organizational process alone is not going to cut it. If we want to fight fires, then we have to campaign like Smokey the Bear. "Only you can prevent forest fires!" was his motto. What is yours?

How about these? "Only you can prevent outages from unplanned changes!"- By implementing some more rigourous controls around change and release management.
"Only you can prevent unnecessary downtime by distracting priorities!" - By implementing better incident management procudures, you could avoid the "SWOT" call mentality that takes the engineers attention off of restoring the service and diverts it into CYO for mangaement.
"Only you can prevent business impact from IT service failure!" - By properly planning and validating your infrastrcuture through capacity and load testing, infratstucure validation for fail-over and applicaiton profiling, you can ensure your build and release managment processes are catching issues before your users do.

Why are you still fighting fires? Probably because your processes are like disconnected piles of twigs and your culture is overly reactive.

Tuesday, May 27, 2008

The value of visual monitoring systems

It still amazes me how many clients I still sit with at lunch or at a business meeting, where their pager or cell phone goes off and without even looking at the message just clear it. Then they will smile and say "stupid monitoring system". I've started to joke with them that they must have stock in Duracell. Is the monitoring system stupid, or is it the way they are using it?
The reason people ignore the messsages is because of false arlarms. Why are they false to begin with?

The whole ide of of paging people is great to get their attention while they are out and about. With technology advancements in cell phone, why are we not using more of a visual alerting capability. When is RIM (makers of BlackBerry) going to step up to the table and give us some visual icons to alert us IT folks of systems issues. It can tell me I have a text message or voicemail. How about a disk full or critical application failure? A few simple icons that in a glance I could see whether i need to look at the alter details or not.