Daily Series: A SQL Server DBA Myth a day

A SQL Server DBA Myth a day In questa serie di posts di Paul Randal vengono sfatati alcuni “miti” diffusi tra i DBA SQL Server.

(1/30) in-flight transactions continue after a failover (2/30) DBCC CHECKDB causes blocking (3/30) instant file initialization can be controlled from within SQL Server (4/30) DDL triggers are INSTEAD OF triggers (5/30) AWE must be enabled on 64-bit servers (6/30) three null bitmap myths (7/30) multiple mirrors and log shipping load delays (8/30) unicorns, rainbows, and online index operations (9/30) data file shrink does not affect performance (10/30) database mirroring detects failures immediately (11/30) database mirroring failover is instantaneous (12/30) tempdb should always have one data file per processor core (13/30) you cannot run DMVs when in the 80 compat mode (T-SQL Tuesday #005) (14/30) clearing the log zeroes out log records (15/30) checkpoint only writes pages from committed transactions (16/30) corruptions and repairs (17/30) page checksums (18/30) FILESTREAM storage, garbage collection, and more (19/30) TRUNCATE TABLE is non-logged (20/30) restarting a log backup chain requires a full database backup (21/30) corruption can be fixed by restarting SQL Server (22/30) resource governor allows IO governing (23/30) lock escalation (24/30) twenty six restore myths (25/30) fill factor (26/30) nested transactions are real (27/30) use BACKUP WITH CHECKSUM to replace DBCC CHECKDB (28/30) BULK_LOGGED recovery model (29/30) fixing heap fragmentation (30/30) backup myths

Per i più pigri Paul ha condensato la serie in questo pdf.