Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
| Author |
Topic |
|
eyechart
Master Smack Fu Yak Hacker
3575 Posts |
 Posted - 2007-08-30 : 04:08:23
|
| Is there any reason to not have read_committed_snapshot on for all SQL 2005 databases? Why did Microsoft make this an option instead of a default?-ec |
|
|
Kristen
Test
22859 Posts |
Posted - 2007-08-30 : 04:53:24
|
| Lots of stuff like that is off by default when it first comes out. Torn Page detection OFF in SQL 7, ON by default in SQL 2000. All the new corruption-detection stuff is off by default in 2005 isn't it?Kristen |
 |
|
|
rmiao
Master Smack Fu Yak Hacker
7266 Posts |
Posted - 2007-08-30 : 21:19:43
|
| I think it's because read_committed works in most case. |
|
|
|
eyechart
Master Smack Fu Yak Hacker
3575 Posts |
Posted - 2007-08-30 : 21:38:42
|
| actually, read_committed totally sucks. Reads block writes in read_committed.Anyway, my original question is do people think there is any drawback to enabling this on all SQL2K5 databases? It is already enabled by default for Master and Msdb, why not all databases in 2005?-ec |
|
|
|
rmiao
Master Smack Fu Yak Hacker
7266 Posts |
Posted - 2007-08-30 : 22:18:44
|
| Then have to watch tempdb usage. |
|
|
|
|
|
|
read_committed_snapshot