|
|
|
Ten Centuries
      
Group: General Forum Members
Last Login: Thursday, May 09, 2013 8:07 AM
Points: 1,221,
Visits: 2,614
|
|
|
|
|
|
SSC-Dedicated
           
Group: General Forum Members
Last Login: Today @ 8:21 PM
Points: 32,893,
Visits: 26,765
|
|
|
|
|
|
SSC-Enthusiastic
      
Group: General Forum Members
Last Login: Wednesday, November 28, 2012 4:08 AM
Points: 160,
Visits: 1,006
|
|
| Ditto Jeff's comments, Ken.
|
|
|
|
|
Old Hand
      
Group: General Forum Members
Last Login: Monday, May 13, 2013 10:22 PM
Points: 358,
Visits: 393
|
|
| Excellent checklist! May I add a caveat to #1: The latest patch level supported by the front end app's picky vendor (if applicable).
|
|
|
|
|
SSC Rookie
      
Group: General Forum Members
Last Login: Yesterday @ 10:51 AM
Points: 36,
Visits: 224
|
|
Great article Ken. Much appreciated.
Can I ask your views on the following - I recently had auditors in and one of their recommendations was to increase the maximum number of error log files to a value of 25000 or higher via a registry hack. In the vulnerability report they give the following reason for this:
To prevent the loss of auditing data, it is recommended that you set this value high enough that the error logs will not be overwritten when restarting the database. Also note that there is a stored procedure, sp_cycle_errorlog, that closes an errorlog and creates a new file. An attacker could attempt to cover their tracks by overwriting files using this stored procedure. It is recommended that you set the value high enough that an attacker could not cycle the logs enough times in a reasonable amount of time to overwrite the error log containing an attack.
Many thanks
Steven
|
|
|
|
|
SSC Eights!
      
Group: General Forum Members
Last Login: Yesterday @ 12:37 PM
Points: 941,
Visits: 1,041
|
|
Steve,
For auduting purpose I would recommend saving up the Default Trace log; depending on your work load, you can save that daily, or multiple times/day. And that trace log has alot more information the SQL Server error logs do.
- Mohit.
---
Mohit K. Gupta, MCITP: Database Administrator (2005), My Blog, Twitter: @SQLCAN. Microsoft FTE - SQL Server PFE
* Some time its the search that counts, not the finding... * I didn't think so, but if I was wrong, I was wrong. I'd rather do something, and make a mistake than be frightened and be doing nothing. 
How to ask for help .. Read Best Practices here.
|
|
|
|
|
SSC Eights!
      
Group: General Forum Members
Last Login: Yesterday @ 12:37 PM
Points: 941,
Visits: 1,041
|
|
Oops forgot to ask questions regarding the article :P.
Point #2, Enable Successful and Failed logs ...
I understand having that on for the first little while in a new system to make sure you know who is hitting the system. But by enabling both Successfull and Failed will it not fill up the error log alot quicker. Like for example on some of my serers there are over 3000 successful connections/day. I track successful for a while when setting up new server to make sure things are funcitonal; then I switch to Failed log ons only.
Recommendations are both or just failed? Comments?
Point #7, Create a maintenance database ...
I just starting doing that recently on SQL Server 2000 environment because we needed to track access to the user/database login. And be able to report it out quickly. I am also using this for storing some Stored procedure for selective reindexing? What else can be in this database?
Thanks for the Article, really good pointers :D.
I will be reading the articles you posted today they look really good :D.
---
Mohit K. Gupta, MCITP: Database Administrator (2005), My Blog, Twitter: @SQLCAN. Microsoft FTE - SQL Server PFE
* Some time its the search that counts, not the finding... * I didn't think so, but if I was wrong, I was wrong. I'd rather do something, and make a mistake than be frightened and be doing nothing. 
How to ask for help .. Read Best Practices here.
|
|
|
|
|
Ten Centuries
      
Group: General Forum Members
Last Login: Thursday, May 09, 2013 8:07 AM
Points: 1,221,
Visits: 2,614
|
|
Failed logins are really the most important. It just depends on how busy your system is. It is all preference on what you would like to monitor.
I use the maintenace database to hold procedures and tables for... reindexing, capturing database size for trending, capturing critical events from the error log, etc.
Jonathan Kehayias did a real good presentation called Automating Common DBA Tasks that has a lot of other good ideas. It can be found under the live meeting recordings here.
http://www.sqlpass.org/Community/SIGs/DatabaseAdministrationSIG/tabid/80/Default.aspx
Ken Simmons http://twitter.com/KenSimmons
|
|
|
|
|
Ten Centuries
      
Group: General Forum Members
Last Login: Thursday, May 09, 2013 8:07 AM
Points: 1,221,
Visits: 2,614
|
|
Jeff Moden (11/25/2008) Very nicely done, Ken. Lot's o' work went into that one. Great job.
Thanks Jeff. It really wasn't that much work. I just wanted to share some of the things I do (or at least think about doing) when setting up a server.
I would love to hear what some other people do as well. I am always looking for new ideas to help improve performance and maintain system stability.
Ken Simmons http://twitter.com/KenSimmons
|
|
|
|
|
SSCarpal Tunnel
       
Group: General Forum Members
Last Login: 2 days ago @ 2:50 AM
Points: 4,785,
Visits: 1,334
|
|
Well written. Excellent article...
|
|
|
|