That means you need to have a failover plan
that??™s ready and that you??™ve tested. As you??™ll see shortly, a failover plan can actually be
quite complex, depending on your environment and requirements.
Failback
With log shipping, the only thing requiring more manual time than a failover event is a
failback event??”that is, resetting the topology back to its original configuration. You literally
have to re-create the topology. There is no simple switch that allows you to revert
back to the primary server. Remember, during a failover, the standby server is now
actively running transactions. Unless absolutely no data changes occurred since the
standby server was made the primary server, you??™ll have to create a backup of the new
primary server, move it to the standby server (the old primary server), and reconfigure
the entire thing. If absolutely no changes occurred during the failover situation, failback
might be a bit easier to manage by using a local backup on the original primary server.
Experience has taught me that having no changed data is a rare or nonexistent scenario.
Pages:
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300