Sometimes RAID volumes fail but the good thing of RAID 1 is that the information is also saved in the other HDD.
The easiest way to rebuild the degraded RAID is to set the disks to Non-RAID from CTRL-I. Once you do this, the information will be saved in the working HDD and you can boot to that drive and in that way recreate the RAID using the Intel® Rapid Storage Technology software.
Your information is useful to anyone who loses a mirror RAID1 disk -- which is why one makes a mirror set in the first place I guess. I did this while recently swapping in larger drives.
But I'd be grateful if you could comment on the "spin-up" question.
a) On a soft reboot, there's no problem with the volume.
b) However, on a hard-reboot, where the disks are spun down, there is a problem. At least this is what Ottavio figured out (see reference) -- and so far seems to be the case. My initial reboot resulted in an IRST boot screen which listed the RAID1 volume as "degraded" AND only one of two disks present!
1. Is the disk spinup during a cold boot a possible cause of RAID1 set degradation (i.e. one disk is incorrectly seen as "not available")?
2. Could a delayed boot (e.g. during POST) fix this problem?
Of course if this is in fact the case, then likely the Intel firmware should be updated to add a little startup delay.
This problem is hard to describe, and easy to describe incorrectly. I believe Ottavio could be on to something. I'd be grateful if you could look into the original question -- I think there are more people with this issue that is immediately apparent.
This is not a problem and should not affect the volume performance or life; however, we have received feedback from other customers about spin ups and all this information has been sent to engineering for revision.
Thanks Kevin. This is very interesting news. What kind of ETA can we expect?
UPDATE LATER RE BELOW: I see from Intel doc that the ROM version of the driver and the Windows 7 deployed version of the driver do not need to be the same. I still have the question though -- I'm assuming the DELL ROM version of the Intel RAID software will never and can never be changed? Thanks!
HERE IS MY ORIGINAL SUPPLEMENTARY QUESTION, IN CASE ANY OTHER READERS ARE WANTING TO KNOW WHAT THIS IS ABOUT.
Also my IRST software is listed (per recent update) as follows:
- INTEL® Desktop/Workstation/Server Express Chipset SATA RAID Controller 220.127.116.112
- RAID Option ROM Version 18.104.22.1681
The first item is the in-place Windows 7 Console and IRST driver.
The second item comes from the IRST help screen System Report. (Apparently this version number will also show up on the RAID BIOS splash screen at the beginning, where you can do <CTRL-I> and go into BIOS-level RAID management.)
What is the difference between the two please?
* Is the second item (ROM) what Dell has installed in the firmware?
* Can it (ROM) be and/or does it need to be updated?
* How is No. 2 (ROM) related to No. 1 (Driver)?
The Raid option ROM is also known as the RAID BIOS where you can create or delete Volumes without accessing the Operating System. The Intel® Rapid Storage Technology is the software that allows the user to interact and monitor with the RAID Volume within the Operating System.
So, these are different numbers but at the end are related.
1 of 1 people found this helpful
Important Update --> For anyone with this issue (and especially with a Dell Optiplex 755), the new IRST 13 series drivers are the problem it seems. I didn't read the fine print that the Q35 Express Chipset was only supported up until the most recent Version 11 driver. I uninstalled the Version 13 software, rebooted, installed the older driver that was compatible with the chipset, rebooted, and everything is now running better. On both reboots, there was no RAID set member loss. So, I kind of expecting that this is the answer.
Two weeks later confirmed running perfectly on multiple 755's.
It is good to hear it is working fine now.