How you deal with things when they go wrong is how you pave the way to make things go right. This holds true whether
Requires Free Membership to View
When you register you’ll also receive the latest news, advice and technical tips designed specifically for midmarket IT leaders like yourself. Our award-winning editorial team will give you immediate access to emerging business and technology trends.
Scot Petersen, Editorial Director, SearchCIO-Midmarket
it's a desktop machine or a server that's crashed, but I think that desktops need their own brand of attention.
A disaster on a desktop machine usually isn't as crippling as a server going offline. On the other hand, it's still nothing to treat lightly: What if 100 other users depend on that one user whose desktop went down?
Here are five best practices for dealing with a problem on the desktop.
1. The user comes first
Before you bring all your attention to bear on the computer, think about the user first. If the workstation crashes and the user is working on something time-critical, get the user to another workstation that does work.
Make sure you keep the user informed about how repairs are going, even if you don't know everything yourself. If the user has data on the machine that's in danger of loss, pull that out first before doing anything else.
2. Find out what happened and why
The majority of desktop disasters fall into the three following categories:
- Damaged operating system. A virus or hardware mishap has rendered the OS unbootable or unstable.
- Damaged applications. Critical applications no longer load or behave correctly.
- Data loss. The computer can't read any stored documents.
The reasons for disaster also fall into roughly three categories. They are:
- Hardware failure. The system's hard drive or other components have failed, possibly triggering one of the other symptoms described here.
- Malicious behavior. Viruses, malware or even sabotage can cause things to go awry.
- User error. Everyone makes mistakes, although some of them are more fatal than others.
Also remember that most of what you see may only be symptoms of the real problem. For instance, if an application crashes repeatedly despite being reinstalled, the problem may be a corrupted user profile that's damaging the application's local data, not the application itself. Try creating a new user account on that computer and running the program there (and then deleting the old user profile after migrating all the needed data). This will be far less of a hassle than re-imaging the machine from scratch -- which, yes, might fix the problem now but wouldn't tell you why it happened, or how to avoid it later.
3. Keep bootable rescue media handy
If a system is teetering on the edge of collapse and you need to dive into it to pull out critical data without actually booting it, the best way to do that is with a bootable rescue CD. This lets you quickly get into a system that's been trashed or compromised, copy off or repair what you can and salvage what there is to be salvaged -- without having to rely on the operating system, or the hassle of hitching up another hard drive.
While this can be done with the Windows Recovery Console, it's extremely limited, and often ornery to work with. The Ultimate Boot CD, for instance, is an excellent Linux-based recovery CD; there's a Ultimate Boot CD. You may also be interested in the tip "No separate license required for his Ultimate Boot CD."
4. Keep fresh system images handy
If there's no time to try and disentangle what's wrong, the best deal may be to simply re-image everything: pull out what data you can, pop in a system imaging disk and pick up where you left off. "Swabbing the decks" minimizes variables and saves time, which is why there should always be a source of fresh system images, even when you're not actually prepping any new desktops at that time. Be warned that imaging a system is not a cure-all, since it may only paper over the real problem.
5. Document everything
When you're done mopping up after a disaster, document both the problem and the solution. If the same system keeps failing, for instance, simply wiping the machine and re-imaging it is not a solution anymore; it's a stopgap. Find out what the real problem is -- any system that fails catastrophically more than twice in a row is probably suspect and shouldn't be used again.
If you keep records of past failures, it'll help you be better prepared for future failures -- whether on the same machine or not. Memory can fail you, and if someone comes along to do the same job in the future, they deserve to learn from your experiences.
Serdar Yegulalp is editor of The Windows Power Users Newsletter. Let us know what you think about this tip; email editor@searchcio-midmarket.com.
This was first published in July 2005