SQL Server Crashes and Critical Failures

Visit Tutorial Page ( Report)
Software crashes. WE can't stop that. What WE can do is prepare ourselves through education and practice to be able to act quickly and resolve these critical issues. In the early days of SQL Server many support calls were server down issues that ended up being bugs in SQL Server. While this still happens, today many of those calls aren't the fault of SQL Server. For example. On Microsoft's case history of database corruption 99% of ALL database corruption can be traced back to the IO subsystem. Either way, when SQL Server crashes it can be a traumatic experience for a DBA. If a production box crashes during core hours WE need to know what steps to take to resolve the problem as quickly and as cleanly as possible. This course will take US beyond where to look. WE will cover RECOVERY INTERNALS and the fundamentals of how to analyze, diagnose and resolve critical errors. WE will cover the SQL Server Error Logs in great detail. One of the first places as SQL Server DBAs WE look to diagnose critical failures is the SQL Server Error logs. WE will cover server startup failures and analyze SQL Server stack dumps. Additionally, we will analyze several commonly occurring exception signatures. WE will create a troubleshooting task list as use it as a template for resolving these crashes and critical errors. While this course has no prerequisites a strong background in SQL Server is highly recommended.
  • Introduction
    • Course Introduction

      As DBAs we don't spend a lot of time practicing how to react to crashes. That's because SQL Server rarely crashes. That makes this course so important. Since this is something we don't do often we need to be even more cognizant of how to handle these critical failures.

    • Instructor Introduction

      A little about me.

    • Download Course Content Here
    • Tools We Need For The Course

      In this lecture we are going to create shortcuts to the tools we need to take the course. We don't want to hunt these down every time so let's pin them to our task bar.

    • The Three Categories of Critical Failures

      There are three broad categories of crashes. Let's learn what they are in this lecture.

    • Summary

      Let's cover what we learned in this section.

  • SQL Server Startup Parameters
    • The Startup Parameters SQL Server Needs

      SQL Server has 3 parameters used upon engine startup. 1) The location of the master data file 2) The location of the master log file. 3) The location of the errorlog. Let's take a look at these startup options. We will also create a shortcut to the command line because we will use it most of the time to troubleshot startup failures.

    • Starting A Named Instance and Putting That Instance in Single User Mode

      There are two kinds of instances. The default instance... which is the name of the server, laptop... etc. Then there are named instances... A named instance is identified by the network name of the computer plus the instance name that you specify during installation.

    • Troubleshooting The Failed Startup With Account Locked Out

      The most common reason the SQL Server engine will fail to start is permission errors associated with the user login you've designated to start that service. Let's learn to troubleshoot this and correct it quickly.

    • Troubleshooting Failed Startup When Port Already In Use

      Often times DBAs like to change the default port of the instance for security purposes. However, they often forget to see if that port is already in use. In this lesson we are going to learn how to check for used ports and change the port number for our instance.

    • Section 2 Summary

      Let's go over what we learned in this section.

  • Error Log Dissection
    • Starting With A Clean Error Log

      We need to see what a clean errorlog looks like so we can spot the anomalies when they happen

    • What's in The ErrorLog - Part 1

      Int this first lecture we will begin combing through the SQL Server errorlog line by line.

    • What's in The ErrorLog - Part 2

      In this lecture we will continue to comb through the SQL Server errorlog line by line.

    • What's in The ErrorLog - Part 3

      In this lecture we will continue to comb through the SQL Server errorlog line by line.

    • What's in The ErrorLog - Part 4

      In this lecture we will wrap up dissecting our errorlog.

    • Summary

      In this lecture we will summarize what we've covered in the SQL Server errorlogs.

  • SQL Server Recovery Troubleshooting
    • Restoring The Resource DB From Another Instance

      The resource db is a hidden system database. It's a true database but we back it up and restore it differently. Additionally, there are a few tricks we can employ to recover it in an outage.

    • Restoring The Master Database When Instance is Online

      There is a straightforward way to restore the master database but it's contingent on two things. Is the instance online? Do you have a good backup for the master database? Let's learn how to restore the master database the easy way.

    • Recover Corrupt Model Database From Another Instance

      This isn't the traditional method we will use to recover the model if it becomes corrupted. However, the more tools we have at our disposal the better prepared we will be during a crisis.

    • Recovering A Lost SA Password

      In this lesson we are going to hack our way into SQL Server using our administrator rights at the OS level. It's a tip that comes in handy on the rare occasion you need it.

    • Repairing Tempdb Corruption

      Repairing tempdb corruption is often straightforward. In this lecture we will walk through how to recover tempdb in a crash.

    • Rebuilding the Master Database

      Rebuilding the master database means starting over. This isn't a fun task but we will walk through the entire process step by step.

    • Summary

      Let's wrap up what we've learned.

  • Conclusion
    • Conclusion and Thank You
Write Your Review