Sounds OK.

When you say "service was refreshed" do you mean restarted?

I'd check the boot.ini on both nodes and reboot each node in turn.

I'm guessing that the server has "forgotten" about the PAE...

See this about changing collation.

http://support.microsoft.com/default.aspx?scid=kb;en-us;325335

I have a procedure as well (in word format) I wrote for changing the collation of an entire database (including computed columns...).

private me if you...

What service pack for SQL? What hotfixes? What OS and SP?

SQL cannot use min memroy because it cannot see it. Why can it not see it?

"Total server memory" (if you read the explanation in perfmon) is how much memory SQL server is using. Not how much the server has fitted. It does not include sqlservr.exe...

If you set max and min memory and the working set size, you will probably make the disk issues worse by causing paging (for agents etc  as well as IIS...

http://support.microsoft.com/default.aspx?scid=kb;EN-US;316749 explains that SQL will use all but 384MB of it's memory between 2 and 3GB.

I guess here that AWE and PAE isn't being recognised, and that SQL is only...

http://support.microsoft.com/default.aspx?scid=kb;EN-US;316749 explains that SQL will use all but 384MB of it's memory between 2 and 3GB. I don't know why it isn't mentioedn elsewhere

Less than 2GB it is dynamic. More...

Are you sure about 3605? Send trace output to the error log?

-T1205 and -T1204 are the usual ones for extra deadlock info.

If your performance is better after, then your data is fragmented on disk and logically. Look at DBCC SHOWCONTIG. Hence your server gets better afterwards.

Deadlocks will be in the SQL...

Do you mean after the reindex or during?

And is it DBREINDEX or INDEXDEFRAG he is using?

If it frees up resources AFTER and DBREINDEX then it's because the table is contiguous...

I agree. This is not a set based solution.

Because there are mutiple path a and sums to get 50, then ther will have to be lot of iterative logice

But: If...

I don't think he is restoring daily, hence the problem with truncate log.

Also, why would your theory not bollix up the first copy and restore of the database, but only...

1. Start the first log-backup and -restore after the db-restore is finished

He only restores once... and teh log falls over after a day

And explain to me why truncate log does...

I know this is obvious, but you are either restoring logs in the wrong order or you have a "broken" LSN chain. he LSN sequence can be affected by an action...

Ha Ha! 14 mins! Central European Time!