Hi All,

I have a ML350 G5 server with an embedded E200 controller. I had configured a raid 5 with 4 x 147 HP SAS drives.

I do not know the exact scenario (Windows crash, then disk crash, etc...), but I have changed the drive in bay one. I believe I did too many reboots yesterday disconnecting and reconnecting the faulty hard drive. I had the choice to press F2 to activated the logical drive, which I did, hoping that Windows would boot. It didn't.

I'm actually installing a new sbs 08 on a simple sata drive. I was hoping I could insert back (i have removed them for now) the three initial disk and that I could recuperate the data on it. Is it possible ? How ? I'm new in the recovery processes... i read here that you use Studio-R software to create a new image. Is this what I should do ?

Thank you for your help.

Well, yes as a general plan, but the prerequisites for a successful secure recovery is to get a diagnostic on each drive to determine if there is something wrong with them. Then create file images or clone drives to healthy media using equipment that is appropriate based on the diagnostic found for each drive. After that, take the images or clones and reassemble RAID in R-Studio. Hope you have labeled the drives in order and try to remember the RAID stripe size.

Suggest professional help if the data is important as I believe it will be a little more challenging for an inexperienced user to DIY when one of the drives has issues and you have reinitialized the array.

_________________
Hard Disk Drive (HDD), Solid State Drive (SSD, SATA, NVMe, etc), USB Flash Drive and RAID Data Recovery Specialist in Massachusetts

Thanks labtech for your answer.

With the HP Array configuration utility telling me that the 3 disks are okay, can I directly make an image of my 3 disks ? What software should I use for this ? How can I make an image if Windows doesn't see the three disks because they are part of the array?

How can I determine the stripe size ? I did not see this information in the HP Array configuration utility... I would imagine it's 32K or 64K ... what's the most standard ? Can I try one and if it doesn't work, go to the next size ?

What's the most important to you? Recovering the data successfully and in a timely manner OR trying do this on your own with no experience, with a high risk of possibly causing more damage, thus jeopardizing complete data loss and taking a lot of time as you will need a lot of info to do this correctly.

_________________
Hard Disk Drive (HDD), Solid State Drive (SSD, SATA, NVMe, etc), USB Flash Drive and RAID Data Recovery Specialist in Massachusetts

Well sure the data is important, but the most most important is saved. I was hoping that I could at least make an image (what software?) of my three disks and then figure out a way to read them ...

This is HP server which uses proprietary algorithm for parity distribution. Also there is 100% an issue with one member which can leave sync problems in the volume. You say you changed the disk, but made many reboots with disk both attached and removed? Did you make a rebuild, did this complete?

If data is important there are too many factors here that can compromise its recoverability, especially since you have relatively little experience with both cloning of the disks and the rebuild.

You should certainly consult a professional firm who at the very least can assess and give you an accurate idea of both the problem and what can be done to save the information.

You need to connect each drive to a separate sas controller in a separate PC one at a time and make sure you ID/label each drive SAS0 SAS1. Download FTK imager from access data which you can create a raw image (same as dd) in Windows of the two good disks.

If the images get created ok with no errors then once you have the images then try r-studio but r-studio can't figure out the parameters they should be in the SAS raid card configuration which you have access to in the server. With a lot of (luck) you may be able to re-build the raidset and recover the data. Don't bother contacting r-studio support they only have general/generic canned responses they cant help you figure out the raid parameters. If you have only pulled the one drive each time and not the any of the other 2 you "may" get lucky it just depends on how the re-build process started to reconstitute the 3rd drive and the potential level of damage on the other 2 drives if any. Long road you have yet to travel good luck with your recovery

R-Studio will require manual creation of block/XOR ordering. Not simple and definitely not recommended for someone with limited experience.

Also you would require to image each disk from as additional SAS HBA as opposed to the current RAID controller which I assume would not allow access to individual disks but only to the striped volume as determined by the controller.

Additional image files / donor destination disks will be required then a designated machine will be needed to work from.

The integrity of the failed disk and/or its replacement should be questioned as this can make a huge difference to the overall rebuild.

Block size should be identified but specific parity rotation should be considered. Disk order needs to be identified and any valid offsets found, usually found following analysis of partition and File System information.

This is not a simple case for any home user.

If professional DR is not an option and you have the resources to attempt to secure the disk images then by all means try and I wish you the best of luck