Alternate means of diagnosing hard drives??

[Tom Lyczko]

We recently installed a new HP 642 Smart Array RAID controller in our ML350 G3 after HP Diagnostics of the previous controller was claiming all our drives were bad, yet the server was functioning perfectly normally. We had the most current firmware and SMH installed.

After installing the new 642 RAID controller, the new controller now claims that ONE of our three drives is bad, via HP Diagnostics. The HP SMH doesn't say anything.

Is there any kind of alternative means of examining the server's hard drives so we could try to figure out if the 3rd drive really is bad or not??

Would it be worthwhile to remove and reseat the third hard drive and see what happens??

Thank you, Tom

[KarloChacon]

hi

yes of course try reseating the HDDs.

also you can check HDDs using Array diagnostic utility as well as HP Insight Diagnostic Online.

Both are part of the HP system Tools if you installed the Proliant support pack

regards

[Tom Lyczko]

Hello, thank you for telling me about ADU.

HP Insight Diagnostics is what I used that tells me one of the drives is bad.

I created an ADU report, and I don't see anything obvious that jumps out at me, is there anything specific I should look for??

Nothing in the created report says anything about problems, it says no problems detected, even.

For the drive that HP Insight Diagnostics claims is bad, ADU's report says Last Failure Reason: 0x00 (Drive has not failed).

The above is for drives 0 and 2, drive 2 is the one that HP Diagnostics claims is bad.

For physical drive 1, ADU says Last Failure Reason: 0x14 (Drive removed from hot plug) and this makes sense because we did change out this drive last fall.

Is there anything else specific I should look for??

Thank you, Tom

[KarloChacon]

hi

let me ask for this

"After installing the new 642 RAID controller, the new controller now claims that ONE of our three drives is bad"

where is the new controller showing that one HDD is bad?

upload the ADU report I can take a look

regrads

[Tom Lyczko]

The new controller is presently installed in our HP ML350 G3 server. This is the controller which claims Drive 2 is bad.

The previously installed controller was saying that ALL the drives were bad. I didn't know about ADU until today.

I don't know if the zip file will upload...I'll try again if it doesn't...

Thank you, Tom

[KarloChacon]

well

Sorry to ask you again what do you mean is claiming any other error on POST, usually in case there is an error ADU report would report it

and ADU report looks fine

regards

[Tom Lyczko]

Hello Karlo,

Thank you for looking at the ADU report and stating that it looks okay.

Yesterday afternoon I took the offending drive out, then put it back in. Today HP Insight still says it is bad. I attached an image.

I've twice watched this server boot up and not seen any unusual or abnormal POST / boot-time errors appear.

We do have another ML350 server, purchased at the same time as this one, with the same kind of drives. We are thinking of swapping each server's Drive 2, then observing what happens in HP Insight Diagnostics.

Our second option is to buy another similar drive and put it into the same server. Would this make Windows Server think there is added more hard drive storage that it does not know what to do with, or would the RAID controller just spread the data over four (4) drives instead of three (3)?? This is important to us because the consultants who set up these servers in 2004 set them up with dynamic drives, a practice that has fallen into disfavor due to its unwieldiness.

Thank you, Tom

[e4services]

I am joining in this late, but I can address some of the questions. From the last one, I would take a new drive, after removing the so called 'Failed' drive out waiting some time to have it reported as truely "failed' and then inserting the new drive in its place. If it still is showing a predictive failure then something else is messed up. The Slot, The cabling, something.

As for testing the drives. Truely to test the drive, you must connect is to the basic SCSI controller and run diags from the manufacturer. Seagate SeaTools, Fujitsu's SDiag or Maxtors SCSIMax (though HP didn't buy many from them), these are the tools we use to check all our new and RMA drives. We conect the drives to the basic SCSI controller, make sure they report to the controller on Boot, boot from a Windows disk and run a complete low level test of the drive. These tools can check the disks completely and report what if any problems they many have from the manufacturers view.

[Tom Lyczko]

I found out from talking to a support tech at CDWG that swapping drives is not a good idea, so we won't do that.

Thanks a lot for suggesting manufacturer diagnostics, do you have any suggestions how to figure out who made it, would this be shown on the drive itself???

I'm not sure I could connect the drive to the SCSI controller itself, it has a tape drive in it.

Meanwhile I will look for those tools and see if we could use them...

Thank you, Tom

[KarloChacon]

Tom

HP insight diagnostic is a very reliable tool you can replace HDD.

Checking your ADU you have RAID 5 with 3 HDDs

so you take the bad HDD out and replace with a good one

Regards