Question AMD RaidXpert array went critical and then failed -can I remount an offline drive without losing data?

[Nigeman]

Hi,
I have an old HP microserver running win 10. I was mostly through transferring data off it onto a new machine when I noticed the array had gone critical... it has 4 2tb disks as RAID 1 named Z and U. Looking at the logs attached I see Task 20 disk errors on Port 2 ( disk 2 of Z logical drive), then suddenly Port 1 is listed as 'unplugged' and then replugged. That drive now shows as a single disk. I still had Z drive functioning on one drive (Port2). but I decided to do a CHKDSK /r to try to repair any disk faults on restarting the machine. Unfortunately that resulted in disk 2 now showing as 'offline'... it seems /r also 'fixes' logical disk errors and in this case it decided that this disk wasn't in an array and should be offline, unless /r always dismounts a disk?

I am not sure if i have lost data and I now don't have access to check if I have got all the data off the Z array. Is there any way to remount this 'offline' disk without losing data please - so that I can rebuild the array? Do the error messages mean the data was recovered and so I may not have lost data?
I can use a caddy and to see if there is additional data on either drive as neither is bitlocked, and I do have a 5TB 2.5" that I can compare with freefilesync if I need to.

The detailed data is:
after CHKDSK:
drive on port 1 2TB single disk
Drive on Port 2 2TB Offline
Before CHKDSK
drive on port 1 2TB single disk
Drive on Port 2 2TB assigned to LD1-02
2023/11/22 14:15:37 disk (port Nmber1, Target ID1) Plugged in
2023/11/22 14:15:21 logical drive "ZDrive" goes critical
2023/11/22 14:15:21 disk (port Nmber1, Target ID1) unplugged
2023/11/22 14:14:25 Task 20 timeout on Disk (Port number 1, Target ID1) at LBA 0x0b147eb28 (length 0x8)
2023/11/22 13:28:26 Task 20 timeout on Disk (Port number 1, Target ID1) at LBA 0x072e641f8 (length 0x8)
20223/11/20 22:30:50 disk error at LBA 0x0e4607b98 (Port number 2, Target ID1) fixed
3 other disk errors at different addresses fixed
2023/11/20 22:30:50 Task 20 disk error on Disk (Port number 2, Target ID1) at LBA 0x0e4607b98 (length 0x100) with Status 20; error register:0

Many Thanks,
Nigel

 

[fzabkar]

CHKDSK is a bad idea when run against drives that have physical problems. MS cares more about the consistency of the file system than your data, and it will often sacrifice your files to this end.

Start by getting a SMART report with a tool such as CrystalDiskInfo or GSmartControl.

I would clone your drive(s) using a tool that understands how to work with bad heads/media (eg HDDSuperClone/OpenSuperClone), and then run data recovery software against the clone.

 

[Nigeman]

Thanks Fzabkar. So if I cloned this drive would I be able to re-mount this into the array? and how please?

 

[fzabkar]

If you have allowed CHKDSK to make changes to one drive in a mirrored RAID, then this will have broken the RAID. You need to recover the data, reconstruct a new RAID, and then restore your data.