May 16, 2009 | SQL Server

Connect Digest : 2009-05-16 : please vote!

Here are the items from this week that I feel deserve some attention.

====================

Hotfixes / service packs should leave authentication mode alone

I may have talked about this issue here before, I'm not sure, but I thought I was the only one who saw this happen.  This week someone else came across the issue, where they had a SQL Server 2005 SP2 instance in mixed mode authentication, applied SP3, and the server changed to Windows Authentication only.  This is really not cool and needs to be fixed!  So far I'm the only person who has voted (even the submitter hasn't voted yet).

#450316 : Named instance changed to Windows-Only authentication on SP3 CU1 hotfix
http://web.archive.org/web/*/https://connect.microsoft.com/SQL/feedback/ViewFeedback.aspx?FeedbackID=450316

====================

Fixing documentation, and really fixing it, in both versions

They keep closing this item as fixed, but I don't think it is.  (Louis Davidson pointed this out to me earlier this week.)  First of all, they did not fix the 2005 documentation, and I think it should be corrected in the next revision. And the fix to the 2008 documentation has a typo, which should also be corrected.

#264723 : Doc : dm_os_performance_counters should show cntr_types
http://web.archive.org/web/*/https://connect.microsoft.com/SQL/feedback/ViewFeedback.aspx?FeedbackID=264723

====================

Having samples in the documentation that work

I have pointed out several Books Online topics that have non-functional sample code, and several more that were just bad ideas (like the CREATE DATABASE topic which only used dynamic SQL in the examples).  This week someone found a geography sample that yields a .NET Framework error:

#450386 : Non-Functional Example Code: http://msdn.microsoft.com/en-us/library/bb933811.aspx
http://web.archive.org/web/*/https://connect.microsoft.com/SQL/feedback/ViewFeedback.aspx?FeedbackID=450386

====================

Allow us to specify collation for user-defined types

Just this morning (well, I guess this afternoon for him), fellow MVP Erland Sommarskog filed this issue that suggests we should be able to specify collation for user-defined types.  Of course, as he states, real domains would be a preferable enhancement.

#451830 : Should be possible to specify collation when define a type/domain
http://web.archive.org/web/*/https://connect.microsoft.com/SQL/feedback/ViewFeedback.aspx?FeedbackID=451830

====================

1 comment on this post

    • Alexander Kuznetsov - May 17, 2009, 2:02 AM

      Voted. Thank you Aaron!

Comments are closed.