One drive of my raid5 failed. I have two new ones and an old one and the old failed.
Looks like a drive error. I replaced the drive and it was automatically added to the
drive array and the rebuild starts [I have expected, that I have to tell this action
explicitely]. Astoundigly, the new disk appears at another port on the controller.
I had to remove all, so I was not sure, that I have really used the same connector
on the MoBo [but think so]. This disk was originally seen at port 3 and is now port 1.
But ISM accepted this as a valid drive for the array.
After the rebuild finished [~18h], the new drive failed again. I was wondering, that
the new drive should have been died that fast [new electronics has the trend to
die within the first 100h], but just used the action "clear the error [or such]" and
then started the rebuild again. The next array rebuild failed again, but this case with
another disk!!!! This other drives is not a year old and I cannot imagine, that it
should have failed nearly at the same time like the really old one!
Any ideas, what I've made wrong?
Should I re-arrange the drive connection to the MoBo?
I must remove the disks and test them into another computer. The Intel Matrix
Storage Manager has really not the minimum of diagnostics in it and I feel really
bad with it now. Even the usual tools have SMART diagnostics, which - in case
of the array controller - is naturally hidden for other tools. Until now, Intel is too
lazy to build in access to the SMART data of the drives.
After I read all the threads about the ICH10R, I think, there is a worm in it.
Are there any payable alternates [using a separat controller] for RAID
controllers which work well under Windows?
Thanks anyway and
Using: "Windows Server 2008 R2 [x64]", all patches, MoBo: ASUS P6TSE,
all disks are WD, ISM 126.96.36.1993
[Note:Another bad intel behavior:It does not show the full version,
you must find the right program and uses explorer properties dialog ....]
“Some users have reported that they experienced random drive failures after installing Intel® Matrix Storage Manager version 8.9.”
Latest is this and has a name change now called Rapid Storage Technology
I see this too randomly.. particularly with usage/access occuring heavily on those drives such as during a backup and i try to access another application using data on that array. I'm using v8.9 too since when i tried about 3 different versions of 9.x i had issues with my win 7 x64 system and sound card crackling and i seem to recall even that version was having rebuild all the time issues with the arrays. seemed like interrupt/multitask or possibly x64 problem.
Im using 6 1.5TB seagate 7200.11's (latest firmware they really ony patched this once my drives came preloaded). I have two arrays, Raid 0 and Raid 1 (on 2 drives), and two other arrays raid 10, and raid 5 on the other 4 drives. The raid 10 and raid 5 (ports 3-6) usually are what have the port 3 fail. I shutdown, disconnect that drive. reboot (it then says degraded). I shutdown again, reconnect, reboot. it says rebuilding. This last time it failed during rebuild, so i'm trying it again now...
I'm using almost latest version of gigabyte GA-p55a-ud6 motherboard Bios. They have a beta out there now to fix SSD, but i feelt hat is probably unrelated to ICH10 (suspect its for C300 6/gbs and the micron chip) which i'm also using for boot drive. I had this problem though even when i had standard 3/GBs drive for my boot drive. Also in my system but i feel is unrelated is an LSI 6/gbs raid controller since this issue has happened all along even without that.
I'm probably going to try the 10.1 drivers once i get a clean rebuild. and hope the sound crackling doesnt happen with that version like 9.
I'd be curious if anyone else has tried the december 10.x drivers and is running well on win7x64?
I am very sorry, to post this so late! I have - so it looks - made an error
posting my results and the post is just not here!!
After I recognized, that Intel released a newer version, I gave it a try.
Just everything is ok now and the error disappeared!!
NB:I am using the new 10.x version on "Windows Server 2008 R2" [x64]
i just replaced the entire RAID after a drive failure last month. that was with an older version of the Storage Manager, and the drives were all about 4 years old, so it was nothing out of the ordinary.
however, upon upgrading the drives i updated the Storage Manager to 8.9 and got hit with a "failed disk" message this morning. i'm glad i saw this posting though - installing a different version to fix this is a welcome thought!
question: do i still need to rebuild the volume that 8.9 indicates was failed? when do i install the different version? basically i need to know how can i get back to work quickly without risking problems?
no worries at all. it's all working now!
first i marked the failed drive as ok. then i took the safe approach and waited it out to rebuild the RAID (it took nearly 2 days with only 400GB of data!). after this was complete, i installed version 8 and everything is happy, including me!
Happy, that this worked for you!
A little benefit:From now on, your rebuilds will be MUCH faster.
If I remember this really right, it was about 4 hours [have 400 MB disks
too]. The first time this happens, I thought, the next bug in the
software ... ;-) But is not!!