If an Oracle host OS has a complete disk failure and the host can be restored (assuming from a recent clone of the host OS disks) from backup, is it possible to then point the host to the existing Oracle datafiles on a SAN, as they were before the host disk was lost, and bring the database back up? This is assuming we are using Oracle ASM and that there were no OS or Oracle configuration changes made on the OS between the imaging of the host drive(s) and the drive failure.
- Anybody can ask a question
- Anybody can answer
- The best answers are voted up and rise to the top
There is nothing 'magical' about Oracle and its recover ability. As long as you have your ASM metadata somewhere available, you can import it on a spare host and use that in a fresh installation to attach your SAN hosted files/LUNs and use that to continue your run on. Important is to know that all your data is in ASM and all your ASM disks are on the SAN. If you do happen to have some files containing application data are left on the Server that ran your databases before it crashed, it might cause an extra challenge.
Knowing that it is technically possible is one thing, the ability to perform such a recovery is an other one. The only way to gain that ability is to train for it. Each OS has some things that make it more or less easy to do. This is for any failure scenario.
Yes, certainly, assuming your controlfile and spfile are in ASM too and therefore survive the failure intact.
This is the kind of thing you need to test in your own environment to give confidence you will be able to pull it off for real when disaster strikes - there really is no substitute for testing your DR procedures to make sure you are not making any invalid assumptions such as: