ProLiant DL380 G6 randomly rebooting for a few days

[xwaresarl]

Hi all,

We have 2 ProLiant DL380 G6 hosting an Hyper-V2 cluster under 2008R2 for many months and were happy with them, but suddently, for a few days, one of the nodes reboots at random which is very annoying as the cluster cannot work correctly.

The Servers have both identical processors, Ram quantity, OS and updates and firmware :

- P62  08/16/2010; backup system ROM: 03/30/2010

Both have advanced ILO, and none had a single problem since the day the cluster was brought up, in February..

IML log says on the defective node :

Critical

OS

12/05/2011 20:26

12/05/2011 20:26

1

Abnormal Program Termination (BugCheck, STOP: 0x0000000A (0x0000000000000004, 0x0000000000000002, 0x0000000000000001, 0xFFFFF8000188413D))

Critical

OS

12/04/2011 08:03

12/04/2011 08:03

1

Abnormal Program Termination (BugCheck, STOP: 0x0000000A (0x0000000000000004, 0x0000000000000002, 0x0000000000000001, 0xFFFFF8000189513D))

Critical

OS

12/03/2011 18:02

12/03/2011 18:02

1

Abnormal Program Termination (BugCheck, STOP: 0x0000000A (0x0000000000000004, 0x0000000000000002, 0x0000000000000001, 0xFFFFF800018D513D))

Critical

OS

12/01/2011 14:26

12/01/2011 14:26

1

Abnormal Program Termination (BugCheck, STOP: 0x0000000A (0x0000000000000004, 0x0000000000000002, 0x0000000000000001, 0xFFFFF800018D413D))

Critical

OS

12/01/2011 09:56

12/01/2011 09:56

1

Abnormal Program Termination (BugCheck, STOP: 0x0000000A (0x0000000000000004, 0x0000000000000002, 0x0000000000000001, 0xFFFFF8000189F13D))

Critical

OS

12/01/2011 02:19

12/01/2011 02:19

1

Abnormal Program Termination (BugCheck, STOP: 0x0000000A (0x0000000000000004, 0x0000000000000002, 0x0000000000000001, 0xFFFFF8000189413D))

Critical

System Error

12/01/2011 02:19

12/01/2011 02:19

1

An Unrecoverable System Error has occurred (Error code 0x00000000, 0x00000000)

Critical

ASR

11/28/2011 20:39

11/28/2011 20:39

1

ASR Detected by System ROM

After rebboting, the system gives us these infos :

Signature du problème :
  Nom d’événement de problème:    BlueScreen
  Version du système:    6.1.7600.2.0.0.274.10
  Identificateur de paramètres régionaux:    1036

Informations supplémentaires sur le problème :
  BCCode:    a
  BCP1:    0000000000000004
  BCP2:    0000000000000002
  BCP3:    0000000000000001
  BCP4:    FFFFF8000188413D
  OS Version:    6_1_7600
  Service Pack:    0_0
  Product:    274_3

(Sorry it's in French..)

What could cause an Unrecoverable error with such an error code ( 0x00000000, 0x00000000 ) that seems to be the origin of all our troubles ?

So we're quite anxious to find a remedy or a fix about this problem.. Any help would be greatly appreciated before we call the support hotline...

Thanks in advance for any hint.

Cheers,

Phil.

[xwaresarl]

If it can be a clue, the insights diagnostics says the error is : Unknown - class : 20 - code :3

Cheers,

Phil.

[Suman_1978]

Hi Phil,

Take a look at this Advisory:

http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=c03065184

Advisory has referred to "Unknown NMI Class 20, Code 3"

Also Note: this problem was observed on ProLiant G7-series servers configured with AMD Opteron 6100 Series Processors running VMware ESXi 4.1, but may also occur on other servers with Intel processors installed or while running other Hypervisor operating systems.

I would suggest first check and update the BIOS.

Good Luck!

[xwaresarl]

Thanks,

I'll download the latest proliant SPP and apply it and come back if it does not solve the issue.

[Topdown]

This may help:

http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=c01955503

[Goutham_Sabala]

perform CPLD update from here 

http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=c01955503