RAID HDD faulty possibly causing AD probs

[TraineeGeek]

Hello Everyone,

I have done some research before posting and I'll be spending the evening reading the RAID manual. In the meantime...

Today one of the HDDs in our IBM Lenovo ThinkServer (TD200x -[382241G]- 06) reported faulty .The RAID hardware is a IBM ServeRAID-MR10i . On the desktop we got this message:-

Controller ID: 0 SAS topology error: Device not addressable.

I open the MegaRAID software and one of the disks is reported as:-

(Foreign) Unconfigured, Bad

 

Also today we've had other servers and workstations losing their trust with the domain and sometime taking a few attempts to get them back on the domain.

Is it just a case of replacing the HDD? If so any tips?

Is the faulty RAID causing the AD problems?

Will AD be corrupt?

This is the warning in AD:-

Log Name: Directory Service

Source: Microsoft-Windows-ActiveDirectory_DomainService

Date: 20/07/2015 10:27:05

Event ID: 1539

Task Category: Service Control

Level: Warning

Keywords: Classic

User: ANONYMOUS LOGON

Computer: ####.##########.###.##

Description:

Active Directory Domain Services could not disable the software-based disk write cache on the following hard disk.

Hard disk:

c:

Data might be lost during system failures.

Event Xml:

1539

0

3

12

0

0x8080000000000000

9887

Directory Service

WCDC01.westcliff.blackpool.sch.uk

c:

 

Your help is very much appreciated

TG

Edited July 20, 2015 by TraineeGeek

[Michael]

You're running RAID5 over four physical SAS hard disks (500GB each), so you just need to replace the faulty disk using port 0, then the utility should allow you to re-build the array.

 

I think the other issues you're reporting are just a co-incidence, unless one other hard disk is also failing, but not yet reported. Best advice would be to probably buy two new SAS disks, so you have another spare.

 

I should note if you do have another failure, then you would have to restore your server from a backup which can be a lengthy process.

[Michael]

The error messages you've posted relate to the disk write cache - you've a higher chance of losing data seeing as you've lost one drive out the array. So long as your server's connected to a UPS, the risk is minimal, however if it isn't, the risk of corruption is higher once re-building the array. In other words, if you're re-building the array and you lose mains electricity, (without a UPS) it could be extremely problematic.

 

I presume Active Directory still opens OK and all users/groups appear normal?

[TraineeGeek]

Yeah AD Users & Computers opens fine. Thanks for your advise.

[Michael]

Yeah AD Users & Computers opens fine. Thanks for your advise.

 

You're welcome - I would make replacing that hard disk a priority, with next day delivery if I were you