0

I've got a file server still running on Ubuntu 7.10 on a ASUS PK5 Pro, with a RAID 1 volume set from BIOS with the mobo raid hw controller (SATA 1 + SATA2).

When one of the 2 hard disks seems to fail, then there's no advice and you can guess it only by seeing the server slowing down a lot.

Now I'm facing this situation, and since in the BIOS RAID shell there's no way to check the hds, then the way would be to delete the RAID volume in BIOS (loosing all the data on disk), then checking the 2 hsd separately in linux and then to create RAID again in BIOS and finally to copy all the data again.

Is there a way to check the hds from linux without deleting the RAID volume before of doing it?

  • I'm going to guess you don't actually have hardware RAID, and a check of dmsetup table will show you're actually using software RAID. – derobert Sep 14 '16 at 9:36
0

Only if the RAID software decides to expose something to the system to allow you to interface with the controller. I don't know the specific RAID implementation of that mobo so couldn't tell you. I also haven't seen a motherboard-based raid that does this yet.

I would recommend backing up your RAID data and destroying the RAID and replacing it with Linux's mdraid in order to achieve what you want. mdraid will allow you to check individual disk health and raid array health while your machine is active.

Mobo "HW" RAID it typically not hardware RAID at all so you get none of the performance benefit of an actual hardware RAID and all of the downside of having to find a replacement mobo to recover your data if your mobo dies.

See this Serverfault thread for a bit more detail and discussion of pros/cons: https://serverfault.com/questions/214/raid-software-vs-hardware

|improve this answer|||||

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.