HP Proliant ML350 G5 shutdown by itself

[Daniel Marinov]

Hello,

I am managing HP Proliant ML350 G5, running a Windows 2008 Standard as a member server on a SBS 2008 network. The issue is that the server shuts down by itself on random basis. In the event log, there are no events logged, also in the application log.

HP system management home does not show any items. But in HP insight diagnostic I found the following error:

test_components\storage.dll ID

Malformed NVRAM detected.
Device: HP Smart Array Controller, Slot 0
Property name: World Wide ID



4

F205


The diagnostic of the controller does not show any issues.

The server is not under warranty, so how can I diagnose the issue and solve it.

[Johan Guldmyr]

Hello,

this looks like it indicates an NVRAM issue on the smart array controller.

The RAM on the controller may be possible to replace yourself. What you could try would be to just reseat the DIMM.

What controller do you have installed?

It might also be a good idea to run an ADU test (HP array diagnostic utility) on it and see if that also concurs with the errors.

Also, as you're seeing ram-errors on the raid controller - do run a file system check and check that your data is still OK.

[Daniel Marinov]

Hello,

The controller on the server is: Smart Array E200i.

For this server, I found that software:

http://h20000.www2.hp.com/bizsupport/TechSupport/SoftwareDescription.jsp?lang=en&cc=us&prodTypeId=15351&prodSeriesId=1121586&prodNameId=3279711&swEnvOID=4024&swLang=8&mode=2&taskId=135&swItem=MTX-c3e7acb4f845425d9be5dea323

Is it the correct one for diagnostic ?

I opened Array Configuration Utility, and it show that the controller is ok at the moment.

[Johan Guldmyr]

It may be that the RAM module is not a serviceable part in your model of the E200i.

Yes that is the software.

[Daniel Marinov]

I attached the report from the diagnostic utility as I am unable to see anything wrong in it :)

[Johan Guldmyr]

Hi, yeah I can't see any problem in regards to the cache in there either.

Maybe there is a problem with the diagnostics.

The ADU finds 128MB cache (96MB available for caching).

[Daniel Marinov]

I also have no idea. 

It has the latest Support Pack. I noticed  that there is a new release for firmware upgrade, may be it will help.

[krusej23]

Did you ever figure this out because I am having the exact same problem?

[beana02]

Anyone resolve this? Also having the same problem!

[Johan Guldmyr]

And neither of you have the server under warranty?
Did you try upgrading firmware, proliant support pack, drivers, etc?

[beana02]

Hi, 

No the warranty and support packages have run out. 

All are upto date. 

[ashwin pai]

did you guys get anywhere ? we are experienceing similar issues

[florian_d]

Hello,

I hope somebody have solved the problem, I have the same issue with a windows server 2003 standard x64.

I have tried to change all the hardware, and I finally put hard disks on an other ML350 G5 and I have the same issue.

I can reproduice the issue if I launch the HP insight diagnostic when the server is booted on  windows.

If I try to boot with a usb key on hp utilities, the insight diagnostics finished and don't show me any error except .

Malformed NVRAM detected.
Device: HP Smart Array Controller, Slot 0
Property name: World Wide ID

I also have a very slow network (never more than 15% of my 1Gb teaming)

any idea?

[Goutham_Sabala]

Malform NVRam detected can be ignored , ML350 G5 servers have known issue with the power supplies, If you have two power supplies remove both of them and use a spare ( new or working psu ) from another server and check if that resolves the issue. Also when the server shutsdown(standyby mode ) open the chassis to find any amber leds on the board and let me know

[florian_d]

Hello goutham.

Thank you for your reply.

I use the insight diagnostic and the server restart each time when he arrives to 53%.

As I said, I have the same problem with a spare server:

-I have plugged the disks on an other ml350, booted and tried to launch insight diagnostic on windows, the new server have also done a hard reboot (suddenly black screen and restart on his own).

The logs indicate nothing.

[Daniel Marinov]

Hello,

I forgot to post the solution. I replaced the power supply and the issue was fixed. However the HP system management home page and all other diagnostics were showing that it is working fine, but it seems that it was the issue. A lot of months since I replaced it, and no more restarts.

[florian_d]

Hello Daniel,

Thank you for the update.

My server have 2 redundant power supply and the test with the other server was NOK.

This looks like a software problem.

I will keep you in touch if I found the solution.

Florian

[florian_d]

Hello,

I have solved the problem:

I have deleted and reinstall network card drivers and configure the teaming again ( it was in NFT and i put it in 802.3ad and configuring the router).

Everything is ok now.

Thank you for your help.

Regards.

FLorian

[Goutham_Sabala]

Hi Florian

I'm guessing you resolved servers slow network performance by reinstalling the network card drivers. I dont see Malformed NVRAM detected has anything to do with this and can be safely ignored.

Thank you

[surgeon1976]

I have problem with unexpected shutdown to standby without possibility to boot server through iLO, which i'v described here:

http://h30499.www3.hp.com/t5/ProLiant-Servers-Netservers/Unexpected-server-shutdown/m-p/5882151