It's an active-passive configuration, so there is still some minimal downtime when a node goes down as the cluster moves to the other node. (You probably already knew that, but sometimes it bears repeating).
I like it for what it does. It adds a level of complexity to your DBA activities though. For example, configuring the Oracle Agent for Grid Control on a Windows Cluster requires you to create a "VirtualAgent" directory on the shared storage. Not terribly complicated but requires the cluster to be moved to the 2nd node as part of the configuration. (Note Id: 464191.1)
Make sure you use the latest version of Oracle FailSafe downloadable from OTN and apply any patches from Oracle support. Prior to the latest version you couldn't have both 10g and 11g instances running in the cluster (as I found out after I tried to add a new 11g installation).
There's an additional component called Oracle FailSafe manager (I believe it is) that can be installed but is not required. It however is (was?) only supported on Windows 32-bit systems and would crosscheck your cluster configuration and do some administrative tasks by running the commands for you.
Once I went through the process and figured out what that particular tool did, I realized it was really needed so I add my resources to the cluster manually now which I think is more straight forward.
We have 7 oracle databases running on our cluster and when we initiate a failover, it takes approximately 3 minutes to move all the resources over and have them online. An actual surprise failover is a bit quicker since in that case the cluster doesn't have to shut down the databases on the primary node as they are already down.
If you have an Oracle Users Group near you, it would probably be of benefit to tap into them and see if they have any Oracle guys who have FailSafe setup and be willing to do a bit of consulting (looking over your shoulder as you do the setup/installs) on the initial setup/configuration.
Sr. Oracle DBA
Ingram Barge Company
Nashville, TN 37205
"Quality is never an accident; it is always the result of intelligent effort."
-- John Ruskin (English Writer 1819-1900)
CONFIDENTIALITY NOTICE: This e-mail and any attachments are confidential and may also be privileged. If you are not the named recipient, please notify the sender immediately and delete the contents of this message without disclosing the contents to anyone, using them for any purpose, or storing or copying the information on any medium.
From: email@example.com On Behalf Of Reimer, John J.
Sent: Friday, November 11, 2011 10:15 AM
Subject: Oracle Fail Safe
Does anyone have any experience with Oracle Fail Safe? We are looking into this as an option for High Availability of our Oracle databases running on Windows. It is supposed to work with MS clustering for server failover. From what I have found, it is available for standard or enterprise editions of Oracle, and there is no additional cost. Any opinions, good or bad, would be helpful.
Cooper Tire and Rubber Company