Editorials

Troubleshooting Approach – Optimize Your Chances for Success

Learning SQL Server, SharePoint, Business Intelligence…
Learn More or Register: Virtual Workshop: SQL Server Performance – key concepts, learning and how-to information for working with SQL Server

– Register: Spring 2010 SSWUG.ORG Virtual Conference – SQL Server, Business Intelligence, SharePoint – 75 sessions, 20 speakers…

– Register: DBA School – (60% sold-out already) 15 ppl max – Apr 19, 20, 21 – In-Person class that focuses on the things you really need to know – and shows/teaches how to apply it.

Troubleshooting Approach – Optimize Your Chances for Success
Kent wrote in about the troubleshooting post from yesterday. I mentioned "one thing at a time" and "don’t forget to undo things that don’t work!" as key things to keep in mind. Kent had a couple of additional points to consider:

"I completely agree that a disciplined and systematic approach is key in order to have any hope of understanding the root issue. The important practical upshot here hits not on the exotic and mysterious glitches, but the simple stuff. You might scramble to “fix” something – and manage to get it back on its feet – while all along some seemingly innocent change made by someone else is the whole problem. Now your workaround is in production, probably giving somebody else a headache, and when the environment changes back the problem resurfaces and you go into another tailspin. Had you taken the time to learn the real situation you could have applied a solution that would prevent the same issue in the future. (Sounds like communication is the root issue in my example!)

I will offer a modification, or an extension, to the “change one thing” rule. We must use a flexible definition of “one thing” to include sets of things that interoperate. Without the ability to work with variables in various combinations, it’s impossible to address mixed-mode cases. And in my experience, lots of misbehavior ends up being the result of interaction between multiple factors. Often the problem has been dormant, or below the threshold where anyone cares, until something exacerbates it into a complete failure. Either irritant alone wouldn’t cause so much damage, but their mutual presence is disastrous.

While in multivariate troubleshooting mode, you still have to be systematic. All the more so! Change “one set” of parameters, and then change it back before trying the next set (which might overlap with the previous one). It gets very complex, so taking notes and planning your moves is really important."

Featured Script
fn_BracketedWords
Text processing. User-defined function for extracting bracketed words from a string. Returns word plus bracket start and end … (read more)

Webcast available on-demand now:
High Availability Through SQL Server

With failover clustering, data mirroring, peer-to-peer replication, and log shipping available there is an option or two for every scenario. While each technology has a place in the world, they aren’t interchangeable and provide the greatest benefit when applied to the proper solution. A walkthrough of each of these technologies and where they fit in your environment.
Presented by: Jason Strate

> Watch Now