Intel ESRT RAID 5, Medium Errors, Backups on SBS2008

Experts-

My SBS2008, Intel S5500HCV with Four WD1003FBYX Drives in on-board RAID 5, failed this morning.  It was in a circular BSOD reboot pattern failing on the megasrx.sys file.  I knew this was a RAID driver so ran a consistency check on VD0 (OS volume) and the system now boots and is running.  The RAID Web Console is showing multiple Medium errors on three of the four drives now.  A few questions...

Does the fact that multiple drives showing multiple medium errors mean that the on-board RAID controller has issues, one drive has issues that is multiplying to others, or are those drives actually bad?  I have a hard time thinking multiple drives are bad as these WD1003fbyx drives have been rock solid (I have them in 7 other servers)

If this is bad drives, what is the easiest way to replace all the drives?  If the backup works I can backup the server, replace all the drives, recreate the volumes, and do the SBS2008 restore.  I've done this before with fairly good luck.  Not sure if this is the easiest way to do it though so wanted the experts opinion.

Thanks!

-Joe
jetcosysAsked:
Who is Participating?

[Webinar] Streamline your web hosting managementRegister Today

x
 
noxchoConnect With a Mentor Global Support CoordinatorCommented:
Testing is not destructive. If you say it to fix the errors then it will try to mark the bad sectors as bad and stop the system accessing them - this could function as destructive. But as long as you do not let it fix - it must be ok.
Just find out what is the culprit, the controller or the drives. If you test the drives then you narrow it to its bottleneck.
0
 
noxchoGlobal Support CoordinatorCommented:
You can connect the drives one at once to another machine (slave them) and test with WD HDD test utility. I suspect the problem is in Onboard fake RAID controller. Avoid using them. Best is to invest into PCI RAID Card which will let you configure a real RAID.
0
 
jetcosysAuthor Commented:
Good idea to test with WD's utility.  Do you know if this is destructive?  

While I agree that a add-in RAID card is better, the on-board ESRT on the Intel server boards for me has been rock solid.  I've been using them for 10 years.  I've also used add-in cards and actually have had two of those fail.  :-)  But, I have to agree that this seems like a controller problem.  I think I'll do a full backup (VHD) then update the MB/RAID firmware and do a restore - unless someone can tell me a better way to move from one RAID5 to another more easily.  The good news is last nights full backup worked without issue.

Thx,

-Joe
0
Get your problem seen by more experts

Be seen. Boost your question’s priority for more expert views and faster solutions

 
jetcosysAuthor Commented:
Ok, I'll give that a whirl.  Thanks noxcho.

-Joe
0
 
jetcosysAuthor Commented:
Replaced all the drives, updated the controller firmware, and restored the server.  No errors in any logs after a week of running.  Looks like this is resolved for now...probably never know why 3 of 4 drives had media errors.
0
 
jetcosysAuthor Commented:
Thanks for taking the time to respond Noxcho.
0
 
noxchoGlobal Support CoordinatorCommented:
Glad to be of any help.
Take care
Nox
0
All Courses

From novice to tech pro — start learning today.