Obviously, though, anyone other than the on-staff DBAs and
myself would not have been aware of the tail-log. The full backup was in the process of
restoring, so we lost 17 minutes??™ worth of data. Or course, it may not have been possible
to back up the tail-log, but I verified later that it wasn??™t even attempted.
The second mistake was simply not following the directions.We??™d written everything
out in T-SQL statements, and the restore process was started by means of Enterprise
Manager. The default restore option was to recover the database at the end of a restore.
CHAPTER 11 n DISASTER RECOVERY PLANNING 279
That meant that when the full restore finished, the database was live and no further files
could be restored. In the documentation, we had listed RESTORE DATABASE...WITH NO
RECOVERY. If that command had been executed, I would have been able to finish the
process. Instead, I was stuck with a database that was available, but missing about 18
hours??™ worth of critical data. I had to start over.
The third mistake was making undocumented changes to the backup scheme. This
was a subtle issue.
Pages:
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517