Failed RAID0 Array - Need Help Reconstructing RAID
Posted on 2006-05-02
I have a client whose RAID0 array consisting of 4 SATA 250GB drives went down. While waiting for me to arrive on-site, the client got 'click happy' and had unknowingly removed the array in the BIOS add-on control program.
I have done a little bit of research on reconstructing failed/removed RAID arrays and had come across z-a-recovery.com (zero assumption recovery software), which is a nice product for $99 that claims to be able to take care of these sorts of things. The software installed and performed RAID analysis (took about 2 days) and showed a disk map (with no bad sectors) - and finally discovered the RAID0 array. The idea behind the software is it analyses the drives & creates a virtual raid array in order to allow users to take important data off the failed ARRAY drives. On the next screen, "the Volume analysis" screen, the software just stalled at 90% with no CPU or MB/sec HD cycles...
I got in touch with ZAR's developer and had followed his instructions to run the software in debugging mode again. The second time, during the initial analysis the program said that one of the drives had bad sectors and wouldn't pass this initial phase - got stuck at 10%... again, got in touch with the developer, sent him a link to the 800MB debug log file. He refunded me the $ for the software & said I should try out the competition.
Since then, the client needed his business to continue to run using that specific workstation. We connected external RAID0 SCSI 1.6TB storage (it's a high definition video shop, needs lots of data throughput) & I took the RAID card along with the 4 SATA drives home to continue the recovery effort.
Took one of my workstations and realized that the RAID controller needs a 64bit PCI. Since I don't have a system with the 64bit slot, my idea is to connect the failed hard drive to my non-raid PCI SATA controller in order to recover it.
My question is, can this be done at all? Would you recommend a product that will scan a failed drive and restore raw data from its last state to an image file, which can then be used to reimage a new, working hard drive? I would then take this new hard drive, reconnect it to the original RAID to attempt to restore the RAID ARRAY again using the ZAR software.
Is this a common practice?
There is no redundancy in a RAID0 configuration - being that I don't have a 4 port RAID controller, will I be able to create an image of the hard drive with the bad blocks (skipping over the bad blocks if possible)?