SAS Predictive Failure process and how to avoid corruption Server 2012 R2

[KCLLC]

Product Name: DL380 G6

Operating System: Microsoft Windows Server 2008 (64-bit)

This may not be the area to post this and this is a solution to a problem so feel free to move this to a more enterprise area of the forums.  I had a HP Proliant DL380 G6 with a predictive hard drive failure.  I replaced the drive via the hot swap option:  CAUTION: DO NOT REPLACE A DRIVE WITH PREDICTIVE FAILURE WHILE SERVER IS OPERATIONAL.  You must shut down the server to replace drives with predicitve hard drive failures.  As a result of the hot swap, it corrupted one of my vhdx files (a 2010 Exchange Server).  The symptoms of the problem was I could no longer copy files from this RAID 5 volume to any location, the copy would consistently fail and as a result I could not backup the data.

SOLUTION:  Power off the server.  Replace the drive with predictive failure with a known good drive (preferrably NEW drive) and then power on the server.  The RAID rebuild should take place automatically.  Then, run a CHKDSK /R on the affected RAID volume to allow Windows to replace the bad blocks.  If all goes well, Windows should repair the drive and with any luck (as in my case) you will be able to recover your data and fire up your VM successfully.

The reason I am posting this is because it took me many hours of work and research to resolve this problem and I am trying to save someone else the trouble.

[Torsten.]

There maybe was another problem with your system, unless it is a RAID0 an online replacemement is supported for most smart array controllers, see for example:

http://h20564.www2.hpe.com/hpsc/doc/public/display?docId=emr_na-c03426341