Use this script in SQL Monitor to determine the general load on the operating system you can get a count of the processes that are in a runnable state. This value will go up and down as various systems run on the operating system. It’s only a measure of load on the system and won’t indicate the cause of problems, but will show you pressure on the system.
It is easy to get database monitoring wrong. There are several common-sense rules that can make all the difference between a monitoring system that works for you and helps to avoid database problems, and one that just creates a distraction. Adam Machanic spells out the rules, based on his considerable experience with database monitoring.
SQL Monitor comes with plenty of metrics, but there are always going to be things that you want to measure that can't be done 'out of the box'. Taking database mirroring as an example, Grant shows that custom metrics can be used to monitor and alert whatever aspect of SQL Server or database that you need.
dbWarden is a comprehensive monitoring and alerting solution for SQL Server 2005 or newer. It features an emailed Health Report and includes email and text notifications with customizable metrics for alerts such as Blocking, Long Running Queries and SQL Jobs, CPU %, Log file and TempDB growth.
Why would you ever need to automatically monitor the SQL Servers in your care? What is the business value of doing so? What are the important features that a DBA should look for in a performance-monitoring tool? Rodney Landrum gives answers based on long experience.