Intel Matrix Storage Manager FAIL - Very annoying RAID5 rebuild problem

[idata]

I use a quad RAID of 1TB drives in RAID 0 and RAID 5 configuration on an ICH9R chipset (on Windows 7 x64). All 4 drives are fine.

Each time Intel Matrix Storage Manager detects something weird or Windows causes a HDD-related error, the RAID5 starts to rebuild.

After reboot, the RAID 0 remains at a normal state (although the error was on this one), but the RAID 5 starts to rebuild although it was not even in use/accessed at the time of the error occured.

I have Intel Matrix Storage Manager 8.9 installed.

This needs to be fixed. It takes more then 30 hours to rebuild a 2TB RAID 5 array and it all happens for nothing and continously. Of course array performance is dezastrous while the array rebuilds (and that's all the time !).

The process cannot be canceled.

Please fix this !

Allow the rebuild to be canceled. Make an option to re-mark the array as OK or stop marking it as 'rebuild' for nothing (it shows the status of rebuild instead of normal at system startup).

Is there any current way or making IMSM stop re-re-re-rebuilding my array ?

Thank you.

[idata]

Lots of bad things are going on with 8.9 try 8.8 or 9.5 (not out yet be works good).

The error you get on RAID 0 that degrades RAID 5 has been know to happen and is sometimes fixed with a BIOS update that should only cause a verifying and repair of the array such like a improper shutdown.

Of course if you have setup a Matrix with RAID 0 and 5 across all the drives then thats the same as problems in 8.9

[idata]

Yes, both setups are across all drives.

I installed the unofficial 9,5 a few minutes ago and it seems to rebuild a lot faster then the 8.x.

I'll test if any more fake errors occur after the rebuild completes.

Unfortunately, there's no option to cancel a rebuild in 9.5 either.

[idata]

Confirmative. I have the same problem too. Only RAID 0 and Raid 1 across all disks. The system hangs for a long while, then reports an error on RAID 0 and starts rebuilding the RAID 1 after marking the failed volume as normal. My first thought was that something wrong with the disks. I switched to disks from another vendor and for a long time everythinf was fine. I was almost sure that this was a real disk problem then. And yesterday it happened again. So, it is definitely not disk related, either Intel RAID controller or some other hardware or software component causes the problem.

I doubt that the Matrix Storage Manager is responsible. It is just a passive shell, as far as I understand. More likely it is something related to the RAID controller firmware or correspondent windows driver.

[idata]

I haven't tested if enableing cache on the RAID 0 (although useless from a performance p.o.v.) reduces to rate of the occurance.

[idata]

You mean cache on write? I have this option on.

I just installed Intel Rapid storage 9.5. We'll see how it will behave.

[idata]

Yes, write cache.

If it's on for you and there's no difference...then it means it doesn't help.