Smart array 410i lockup code 0x12.

[Malcolm Watts]

I have a DL380 G6 with a smart array 410i with 512Mb cache and battery backup, this server for some reason crashes on a "hardware malfunction, system halted", this happened after installing a Adaptec 29320lpe, the card was removed and only the O/S was reinstalled, sp'd and sus updated, the same thing happens.
A driver was obtained from HP as the system can be started in safe mode.
we went through the process of removing all updates and rebooting with standard SP2 installed, same result.
any one with any solutions?

[TTr]

Try removing the cache memory from the 410i and see if it helps. If not you can not replace the 410i, you have to get a new system board.

[Malcolm Watts]

Ih removed the cache memory and its battery backup to make sure there is nothing still being held.
Am now doing a system erase and full reload to see if the same thing happens.

[marcus1234]

Hmmmm. malcom

i have seen dead pci cards cause this
first reseat all cables then proceed


this scisi controller is on board

you should see it in bios on boot try f9 bios do you see it in bios boot options ,

if not is sounds suspect

you can disconnect all drives and pci cards and see is it seen now if so add cards back one at a time then drives in that order

try clear nvram its the maintenance switch 6 it is on the cover anyway the one you remove
start up put switch back

remove hdd backplane disconnect allcables if any update firmware wih cd 8.6 and then take optional firmware
for controller

http://h20000.www2.hp.com/bizsupport/TechSupport/SoftwareIndex.jsp?lang=en&cc=us&prodNameId=3884083&prodTypeId=15351&prodSeriesId=3884082&swLang=13&taskId=135&swEnvOID=1005#29214

it is listed as such below
Firmware CD Supplemental Update / Online ROM Flash Component for Linux - Smart Array P212, P410, P410i, P411, and P712m (multi-part download) 2.50
18 Sep 200

NB when doing updates ensure battery is removed at all times just that it is something i noticed that helps

IF none of the above works replace board as it is next logical step....

This is how we thank in the forum

http://forums11.itrc.hp.com/service/forums/helptips.do?#33

good luck :)

[Malcolm Watts]

After having the system board replaced and reinstalling the O/S, and applying the windows updates, the same thing happens again.
It has been narrowed down to a windows update KB936357, this installs a microcode for Intel processors, as our DL380 G6 has a Xeon X5550 it causes the above problem.
After going in to the O/S in safe mode, we can run the spuninst from the directory $NtUninstallKB936357$\spuninstall, but how to modify the microcode on the processor itself?
any ideas before the systems are returned for some G5's that work.

[marcus1234]

have you done the above re firmwares ??

As i have sorted many of these...

[Malcolm Watts]

After much splatting of the server, putting the PSP firmware to lower then 8.6 maintenance cd and applying patch supplied from HP (http://tinyurl.com/ykmjhtg) the system was able to be reinstalled.

The issue was found to be MS patch KB936357, this does a microcode install for Intel chips, thus if you splat and reload a box the issue is still there until you sort out the microcode, hence the patch from HP.

So if you have/or will be installing DL380 G6's, make sure you have the KB936357 patch not authorized from WSUS.

As we had this previously authorized, it installed with the rest of the patches when the server was built.

Thanks to all the responded with suggestions.

[Malcolm Watts]

Do not install MS patch KB936357 on DL380 G6's, you will get the system halt error.

Alas the server ran for a number of hours before doing the same thing, so it looks like another system board replacement is required.

[benda74]

Funny situation you describe - and we have a server that has just started doing the same thing.

We have booted into Safe Mode and removed the MS patch and also changed the firmware on the Array controller but no change.

Seeing we can get into Safe Mode we are going to look at the start up devices / services in case that is causing the issue.

[Malcolm Watts]

We eventually found that is was Kaspersky Anti Virus ver6 that was causing the problem.
uninstall kav6 and install the new enterprise
kav, i think it is version 8.
this sorted out our problems not the ms patch for the Intel processor micro code update as was previously thought.
Hope this helps.

[benda74]

Hi Malcolm,

Interesting about KAV - I'll check that out today.

Cheers,

Ben

[benda74]

Hi Malcolm,

Just to confirm your comments - we were able to access Last Known Good option and we picked up a corrupted KAV6.

As you mentioned will have to look at upgrading or changing AV products!

Thanks for prompt responses.

Ben

[HUNTER_9]

We just got new HP ProLiant DL380 G6 Server.

After install the windows server 2003 standard edition every thing was ok,...but after updating the windows and install the Kaspersky Business Space Security the server gives blue screen error "hardware malfunction, system halted" ,..and this happens after log in into windows server by 3 to 5 mins maximum and after the message appear it takes 2 to 4 mins then the server make a restart.

when i uninstall the kaserpsky anti virus,.. the error did not appear again,.. but after reinstall it appears again.

our server Shipped Prior to November 16, 2009,... and i follow the following article:

http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?lang=en&cc=us&objectID=c01955503

and i updated the firmware but still the same problem,..

then i checked the HP Proliant Integrated Management Log Viewer and i found the following error :

Uncorrectable PCI Express Error (Embedded device, Bus 0, Device 7, Function 0, Error status 0x00004000)

i did contact HP support here and they send me HPS Reports Enhanced version 8.3.00r2 program by email and i reply to them with the output report in .cab file and the HP support reply and confirmed the PCI Express Error with the following article:

http://forums13.itrc.hp.com/service/forums/questionanswer.do?admit=109447627+1267544665423+28353475&threadId=1395518

and they told me that they will send an engineer to replace the system board and i am waiting for that to happen in the next 3 days because they had to ordered the system board.

Now i want to know,.. is this problem appears because of the Kaspersky Antivirus only,.. or because the PCI Express Error only,.. or because both of them?

i need to know if i am gonna face the same error even after replace the system board of the server or what?

Thanks In Advance,

[marcus1234]

hmm Malcom what is you hp case referance number ,
are you using 2008 R2

did you start the install with a smart start cd and let it guide you to install the os ?

also before kaspersky have you installed the psp pack 8.30 this helps some issue.

also ensure the 512 meg cache is on very tight ,reseat it

points appreciated :)

[Mark Ostroski]

I have a DL380 G6 server that was failing with array controller lockup code 0x12. HP replaced the system board and the problem persisted. We swapped every other internal part with another server except the CPUs, to no avail. Finally I experimented with removing one CPU at a time. The problem it turns out was a bad CPU in socket 1.

[HUNTER_9]

Dear All,.. the problem has been solved by kaspersky support.

the solution is to disable "No execute memory" (No execute memory is disabled) option in BIOS settings.

-System options
--Processor options
---No Execute Memory Protection
----Disable

and here you are what is written about this option in the HP ROM-Based Setup Utility
User Guide.

No-Execute Memory Protection (G6 servers)
No-Execute Memory Protection enables the hardware portion of a feature that protects systems against
malicious code and viruses. When used in combination with an OS that supports this feature, certain
memory is marked as not for executable code. Viruses that attempt to insert and execute code from nonexecutable
memory locations are intercepted and an exception is raised.

Thanks & Best Regards.

[Malcolm Watts]

servers now used for VM ESX servers.