Learning how to detect, diagnose and resolve performance problems in SQL Server is tough. It often takes years to learn how to use the tools and techniques that help you detect when a problem is occurring, diagnose the root-cause of the problem, and then resolve the problem.
In a recent webinar, Kevin Kline, Principal Program Manager at SentryOne, provided demonstrations of the tools and techniques which make difficult troubleshooting scenarios much faster and easier. In this webinar (which received rave reviews from attendees), Kevin delves into techniques and tools and provides demos to help you understand his Troubleshooting Checklist.
This checklist was designed with the thinking that if you spend some time on each step in the beginning, then you’ll save tons of time on the troubleshooting process.
Step One: (The best shortcut): Know what has changed! This requires some prep work in advance. Validate in a systematic way that something has changed.
Step Two: Is the issue inside or outside of SQL Server? Spend time to answer this, so you don’t waste time looking in the wrong area.
Step Three: Is the issue caused by a SQL Server error?
Step Four: Are there excessive wait stats?
Step Five: Correlate wait stats against other metrics.
Step Six: Follow up – Make sure to set up SQL Server so that it notifies you, not your customer, if the problem arises again in the future. Be sure you are the first to know, then you can fix problems before anyone is alerted, which increases your credibility with customers as they encounter a problem free environment.
In this webinar, Kevin covers:
XEvents, Profiler/Traces, and PerfMon
Using Dynamic Management Views (DMVs)
Advanced Diagnostics Using Wait Stats
Reading SQL Server Execution Plan
For many, starting the day with detecting an issue, then diagnosing and resolution can take up the whole day. Kevin’s tips will save you time spent on troubleshooting. Watch this webinar to learn all the skills every DBA needs to know how to keep their SQL Server in tip-top condition.