I have a script that begins every night at 0100 to restore a full backup from one machine, across the network to another machine. At about 0140 each night until roughly 0215, sometimes later, I get several messages in the SQL Server Error log stating that a significant part of SQL Server process memory has been paged out. This problem was so severe that it caused a critical app using SQL Server to crash. Why would running a restore cause this? I can reproduce this. All I have to do is run the restore script and the error starts popping up about 30 minutes before the script finishes.
RESTORE DATABASE MyDB
FROM DISK = N'\\MachineName\sqlbackups\MyDBFull.bak'
WITH FILE = 1,
MOVE N'MyDB_DATA' TO N'D:\NewMachine\MyDB.MDF',
MOVE N'MyDB_LOG' TO N'E:\NewMachine\MyDB.LDF',
REPLACE
GO