Disk Enclosures

Array failure

 
ChrisMorris
Occasional Visitor

Array failure

Hello

We're experiencing an intermittent storage failure. We have two servers of the following spec:

Server: GEN8 DL380p
DAS: StorageWorks D2600 at Port 1E : Box 1 / Smart Array P421 in slot 4
o/s: Windows 2012 R2 Standard 

These were purchased about 6 years ago, each with 7 x 4TB drives in the 12-slot DAS box configured as one array designated as D:\.

Recently we've added 5 x 4TB drives to each DAS box as a new array E:\ for extra storage, retaining the original D:\ unchanged. Most of the time these work as they should, but every couple of weeks, during periods of extremely high io, the new array fails. Only the new array. No other arrays are affected and work which doesn’t involve the new array can continue to completion.

Both servers are prone to this. 

Anecdotally, it's as if the new array has hung. Windows file manager shows the drive but without properties. Applications, including our PRTG diags, fail to report any issue prior to failure. Stopping all services / applications does not resolve the problem and we have to restart the server. 
The issue is exactly the same on both servers. iLO reveals nothing useful. It doesn’t appear to be a specific app in use at the time of failure – both SQL Server native backup and Robocopy appear to have triggered the failure.

One of the servers has recently had a new mainboard fitted under warranty, the other has had a new controller card fitted under warranty. Firmware has been recently brought up to date.

Both servers are due to be replaced with gen10 equivalent in the next couple of months, meantime we're limping along, fixing when the need arises.

Any suggestions?

Many thanks.

4 REPLIES 4
sudhirsingh
HPE Pro

Re: Array failure

Hi,

As i understand, array/logical drive does not fail basically but it goes into unresponsive state ?

Your new array is built on 5 drives ?

Please share drive model also spare part number ?

Did you updated drive firmware as well ?

What is the raid type ?

Not sure how much IOPS it can handle but how much the IOPS thrown on this volume ? any idea.

 

Could you please capture and post ADU report from both these servers to review ?

 

I work for HPE
Accept or Kudo

sudhirsingh
HPE Pro

Re: Array failure

Please revert with required details to assist you further.

I work for HPE
Accept or Kudo

ChrisMorris
Occasional Visitor

Re: Array failure

Hi Sudhirsingh, apologies for delay. Here are the details you requested.

array/logical drive does not fail basically but it goes into unresponsive state ?
- correct
Your new array is built on 5 drives ?
- correct

Please share drive model also spare part number ?
- Model is MB4000FCWDK
- Spare part number is unknown

Did you updated drive firmware as well ?
- yes, to HPDB

What is the raid type ?
- RAID 5

Attempting to obtain ADU reports.

Thank you for your interest.


ChrisMorris
Occasional Visitor

Re: Array failure

Dear @sudhirsingh , I have ADU reports from both servers. I can't find a means of attaching these reports to a forum post, how would you like them delivered?

Cheers

 

ChrisM