Use Current Drivers, SPs, Patches
Importance: Critical
Difficulty: Easy
Used to:
-
Increase hardware HA
-
Increase OS HA
-
Increase SQL Server HA
-
Increase application HA
-
Increase human error HA
-
Increase outside forces HA
-
Increase performance HA
As much as we all hate updating our servers, it’s a mandatory task. We just need to accept the inevitable, and at the same time, test, test, and test before you roll any of these changes into production. Use scheduled downtime for these.
Minimize Use of Third-Party Applications
Importance: Important
Difficulty: Easy
Used to:
-
Increase hardware HA
-
Increase OS HA
-
Increase SQL Server HA
-
Increase application HA
-
Increase human error HA
-
Increase performance HA
Includes:
-
Anti-virus
-
Anti-spyware
-
Defraggers
-
Monitoring software
-
Unnecessary services
-
Application software
-
Etc.
Avoid running any third-party software on your SQL Servers, as the more software you have running on it, the greater the chance it will negatively affect SQL Server’s availability.
Cross-Train Backup DBAs
Importance: Critical
Difficulty: Easy/Intermediate
While management might not feel this way, you as the DBA, are the most critical link in SQL Server high availability. Because a single DBA can’t be responsible for everything, all of the time, there must be multiple people who can act as DBA when necessary. Cross-training is critical, and there must be no single point of failure (no DBA with unshared knowledge).
Is That All?
It’s all of this article, but not all you need to know about SQL Server availability. The purpose of this article is to point out that SQL Server availability is critical, that it is made up of many different parts, and that is is not necessarily expensive. Most of the suggestions presented here are easy and inexpensive, but critical to high availability.
High availability does not happen by accident. It is a result of doing lots of things correctly and consistently.
]]>