• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 419
  • Last Modified:

how to tell what caused the raid failure?

I have a hp proliant server. I would like to know in case of raid failure, how do I find out what cause it? From what I know there are several possibilities that could cause raid failure:

- 2 hard drives failed
- raid controller failed
- raid configuration corruption

how to identify in each of the above situation? and is there any other possibilities?
0
okamon
Asked:
okamon
  • 5
  • 5
  • 3
  • +2
3 Solutions
 
DavidCommented:
The raid controller's event logs are a good start.  Are you using HP's smart array? If so download their software to look at the logs>
0
 
okamonAuthor Commented:
yes, it's hp smart array. How to access the log if i cannot even boot into windows?
0
 
andyalderSaggar makers bottom knockerCommented:
Boot SmartStart CD (or Intelligent Provisioning in BIOS) and run ADU/ACU and then upload adureport.txt as an attachment to the thread and we will diagnose, or learn to read the report yourself.
0
VIDEO: THE CONCERTO CLOUD FOR HEALTHCARE

Modern healthcare requires a modern cloud. View this brief video to understand how the Concerto Cloud for Healthcare can help your organization.

 
okamonAuthor Commented:
thanks. besides using smart start cd, is there a way to tell in bios, or in the raid controller setting?
0
 
noxchoCommented:
If the RAID Controller has problems or it is bad then you will simply not have a chance to log into RAID Configuration utility. That should be a sign for you that problem could be with RAID controller and not the RAID set or drive.
0
 
andyalderSaggar makers bottom knockerCommented:
ORCA/BIOS isn't very powerful, it really has to be the ACU/ADU but with Gen8 that's built into the boot ROM anyway.
0
 
okamonAuthor Commented:
thanks. >>If the RAID Controller has problems or it is bad then you will simply not have a chance to log into RAID Configuration utility

how about if the raid configuration corrupted? can I still log into utility? if yes, how to tell it's corrupted?
0
 
DavidCommented:
If the RAID configuration is corrupt, (which happens), then problem solved.  Root cause was controller, not HDD failure.
0
 
andyalderSaggar makers bottom knockerCommented:
If the controller is bad you can simply put the disks on another one of the same or later generation since all HP Smart Array controllers use the same metadata. It's pretty hard to corrupt the configuration since it's stored on every disk although taking them out and shuffling them can confuse it. The ADU is only a point in time snapshot though so if you have dual disk failure it won't tell which one failed first. For that you would need a previous report or you may get enough info out of the integrated maintenance log stored on the motherboard and again read by SmartStart diagnostics.
0
 
Gerald ConnollyCommented:
It's pretty hard to corrupt the configuration since it's stored on every disk although taking them out and shuffling them can confuse it.

When i worked in HP Presales we always used to tell people that you could take them out and juggle them, cuz you could!   :-)
0
 
DavidCommented:
It is hard, but it happens, especially in multiple failure scenarios.  If somebody wants reliability bordering on infallible, then shift the decimal point over to the left in terms of cost ;)
0
 
okamonAuthor Commented:
dlethe, I am not sure what you mean here -> If the RAID configuration is corrupt, (which happens), then problem solved.

so how do I tell if it's corrupt? is it just like what noxcho mentioned that I wouldn't even be able to log on to the utility?
0
 
DavidCommented:
Signs of it being corrupt are basically no configuration found messages; grossly incorrect settings, like it saying you have a 37-disk RAID3, or error messages saying the configuration is corrupt or invalid.
0
 
noxchoCommented:
The the sign will be no bootable device in case system runs from this RAID. If RAID is data drive then it will not be accessible from Windows or throughing errors.
0
 
okamonAuthor Commented:
so how do I repair it? will I lose all my data?
0
 
DavidCommented:
No way to give you a definitive answer in all cases w/o having somebody run some analysis.  I doubt anybody will do that for free as it requires talent and software that you don't have, which means sending drives off and paying for a forensic storage expert to look at it.

It is more money than you want to pay. Trust me.  

If you want something that will catch most problems with a high degree of confidence, then look at controller event logs as that is good enough for most people.
0

Featured Post

How to Use the Help Bell

Need to boost the visibility of your question for solutions? Use the Experts Exchange Help Bell to confirm priority levels and contact subject-matter experts for question attention.  Check out this how-to article for more information.

  • 5
  • 5
  • 3
  • +2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now