You can sometimes use recovery fault isolation to assess the extent of corruption by forcing recovery to run and examining the number of pages marked suspect and the objects to which they belong.
For example, if users report problems in a particular database, set the recovery isolation mode to “page,” and force recovery by restarting Adaptive Server. When recovery completes, use sp_listsuspect_db or sp_listsuspect_page to determine how many pages are suspect and which database objects are affected.
If the entire database is marked suspect and you receive this message:
Reached suspect threshold ’%d’ for database ’%.*s’. Increase suspect threshold using sp_setsuspect_threshold.
use sp_setsuspect_threshold to raise the suspect escalation threshold and force recovery to run again. Each time you get this message, you can raise the threshold and run recovery until the database comes online. If you do not get this message, the corruption is not isolated to specific pages, in which case this strategy for determining the number of suspect pages does not work.
