You could have a controlled failover if
you??™re installing new software or patches, but in a perfect world, a true failure should
never happen. This isn??™t a perfect world, though.
If you??™ve done things properly and no hardware failure ever occurs, you will have
recouped no benefit (other than the ability to sleep at night) from the significant amount
of time you spent planning for the failure event. However, dealing with a failover without
planning accordingly (both in the failover configuration and in how to react) is sometimes
more chaotic than losing a server altogether.
Planning for Failover
Windows clustering isn??™t a simplistic tool to just move all running services from one
server to another. It provides a detailed configuration of each cluster resource, and you
may configure each of those resources differently. Usually I encounter cluster environments
in which all the defaults are used, but that is often not in the client??™s best interest.
As important as it is to properly configure the cluster resources, it??™s also important to
prepare yourself for the event of a failover.
Pages:
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364