SQL Saturday #434 Holland - Slides
On Saturday I presented my session Monitoring SQL Server, Key performance metrics and how to interpret them at SQL Saturday #434 in Utrecht, Holland.
I have already uploaded the deck to the SQL Saturday website where you can download them from the Schedule page, I am also posting them here.
One thing that I thought I would pull out is information around the MSDB table
dbo.suspect_tables, this is where SQL Server stores information on corrupt pages that it encounters. Monitoring and managing this table is a good idea as it can give you insight into the presence of corruption in addition to your normal
DBCC CHECKDB jobs, defense in depth if you will. It is important to bear the following things in mind with this table;
- There is a 1000 row limit in the table
- One row for each page that generated an 823 or 824 (limited by table row limit)
- When a page is repaired the record is updated but not removed (you need a maintenance task)
Finally, if you attended my session (or any of the other great sessions) we would appreciate your feedback for the sessions you attended, and the event organizers would appreciate your event feedback too. Without this feedback we won't know what worked and what did not, thanks.
John (@SQLDiplomat) is the Product Manager at SentryOne, looking after SQL Sentry and the core monitoring suite. John is also a Microsoft Data Platform MVP with more than a decade of experience with SQL Server and the Microsoft Data Platform. John is an experienced DBA, Developer, and former Microsoft Premier Field Engineer. Having worked with SQL Server for the last decade he has gained a broad understanding of how you can use, and misuse, SQL Server. With the latest PASS Board Election Results, John will be the EMEA representative effective January 1st, 2018.