Clustering will do everything for you.
Most of us are not so lucky. A failover event usually causes a great deal of activity that
has nothing to do with clustering or SQL Server. For example, you might need to follow
one of these plans of action:
CHAPTER 7 n CLUSTERING 189
??? Notify the help desk: For many users, an error equals a call to the helpdesk. Many
users don??™t try to restart the application; instead, a sort of technopanic consumes
them. It??™s actually good that they want to call the help desk; they want to be sure
they didn??™t do something wrong. Well, if a failover occurs, you??™d best notify your
help desk to expect calls and tell them to reassure callers that nothing is wrong
with the system and they can simply restart it. If you neglect to inform your help
desk, you could be making some enemies.
??? Restart custom applications or services: Not everything can be automated. Certain
applications that run in the cluster environment (or interact with it) may halt if
they experience a failure. I once dealt with a middle-tier application that would
simply hang in memory if it lost the connection to the database.
Pages:
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367