Adaptec RAID 2405 degraded RAID 10 array
On Mar 12, 11:23*am, none wrote:
Hi, since a few days I have some problems with a RAID 10 array on a
adaptec RAID 2405 card. I've been using this setup for about 10 months
without any problems, but 4 days ago when I turned on my computer I go
a message saying that my RAID array was degraded. In windows using
adaptec storage manager I saw that the array was been rebuilded. It
took 2 or 3 hours and everything was back to normal, the hard drive
that had a exclamation point in storage manager looked fine and I
don't seem to have lost any data. At this time I thought that it may
have been an isolated event maybe due to the fact that my computer
crashed (hang) a few minutes before. But 2 days after I got the same
message (no crash or anything unusual), but this time another drive
had the question mark on it in storage manager. I waited for the
rebuild and everything was back to normal.
Here's some informations I took from storage manager:
First time:
group 0 segment 1 *controller 1, connector 0, device 1 *Rebuilding...
Second time:
group 1 segment 1 *controller 1, connector 0, device 2 *Rebuilding...
Right now I don't have access to the card BIOS so I can't give the
BIOS version and settings (I think if my memory is correct that I have
enabled write cache, which I know could cause some problems). I will
send this information in another post soon.
I have many questions:
- Is this behaviour normal? Can an array loose it's synch sometimes
(after a crash or a power failure) and need to be rebuild?
- Since all drives are working correctly after rebuild and that 2
different drives had the problem, am I right to think that the problem
is not due to faulty drives?
- Can it be caused by a power supply power problem? A few days before
I added a secondary graphic card, a MATROX G450 quad, but I'm pretty
sure I have enough power since my power supply can deliver 850W (see
specs bellow). right now I have removed the MATROX card just in case
- I'm wandering, are group 0 segment 1 and group 1 segment 1 2 HDs
containing the same information? If it's the case I find it strange
that at some point one HD was being rebuild from the other, and that 2
days after this hard drive was rebuild from the first faulty HD. How
do group and segment works?
- Can the card, knowing there was a problem, rebuilt the first half of
my array and 2 days later continue to fix the other half?
- Is there any danger to loose everything when something like that
happens?
- The MATROX card was sitting just on top of the adaptec card with his
heat sink facing the adaptec card, can heat be a problem?
- Someone else ever experienced a behavior like this with his RAID
array?
For the moment I removed the MATROX card and did a complete backup of
the system just in case. I will contact adaptec soon.
Here's my system specs
OS: * * * * * *Windows XP 32bits SP2
motherboard: * ASUS PT6 Deluxe V2
CPU: * * * * * Core i7 940
RAM: * * * * * Corsair XMS3 Dominator 6GB DDR3
Power supply: *Antec CP-850 (850 watts)
Graphic card1: ASUS ENGTS250 1GD3
Graphic card2: Matrox G450-Quad
Capture card: *Diamond TV Wonder 650 Theater
HDs: * * * * * 4 x Western Digital RE3 32MB 1TB 7200 SATA
RAID card: * * Adaptec RAID 2405
Misc: * * * * *DVD burner/BR burner/some fans and fans controller
I'm pretty sure that a 850 watt power supply has enough power to drive
a system like this, or do you think it could cause some problems?
Anyway all monitors on the MATROX were deactivated, the TV wonder was
not in use, and I was not playing any games on the ASUS when the
second problem arised
But I was playing a game (not the latest GPU demanding game) on it
when the computer crashed a fee minutes before the first problem.
Any useful informations, tips or anything really will be greatly
appreciated
Thanks in advance
Here's the BIOS versions and controller configuration
motherboard: ASUS P6T Deluxe V2 BIOS rev. 0302
controller: Adaptec 2405 V5 2.0 build 16343
RAID array config:
mode: RAID 10
stripe size: 512KB
Controller config: (controller 0)
drives writes chache: enable all
runtime bios: enable
auto failover: enable
array background consistency check: disabled
array based BBS support: disabled
SATA native command queuing: enable
default background task priority: high
backplane mode: auto
selectable performance mode: dynamic
Thanks
|