Strange SCSI Processor seen in HP Lpr Cluster

[Benny Koh]

Hi,
I have 2 HP LPRs, each with a NetRAID 3si card, connected to a RS/12 (cluster-enabled) as a cluster. Both servers are running Windows 2000 Advanced Server SP4.
The odd thing is one server keeps detecting an "AMI Megaprocessor SCSI Processor Device" every time it reboots, while the other doesn't. I cannot find a suitable driver for it in my Nagivator CD (M04.05). Neither could I find it on the HP website.
What is it? Is there some way I can get rid of this device? Or a driver can be found for it? Any help is appreciated. Thanks!

[Martin Breidenbach]

Could this be the management processor of the RS/12 ? There's a INF file for the RS/12:

http://h20004.www2.hp.com/soar_rnotes/bsdmatrix/matrix62500en_US.html

[Benny Koh]

I have downloaded the inf device file for D5989A from the website you have suggested. But it doesn't work.
This server used to have 2 SCSI boards with firmware version 1.06. I had changed out both boards to ones with version 1.05 and the unexpected shutdown problem went away for around 3 weeks then it came back again. So the firmware is not the problem.
I have attached a picture of the netraid assistant for this server. The offending device is at ID 6.

[Martin Breidenbach]

The RS/12 has an internal management processor device assigned as SCSI ID 5. AFAIK that INF file is for that device.

The two 3si controllers should use ID 6 and ID 7.

So I'm wondering why I don't see a device at ID 7 here ? Is the other node switched of or is one 3si set to ID 5 ? In that case there's SCSI ID conflict.

[Benny Koh]

Here is the Netraid assistant layout for the other server (CLUSTER12) in the cluster. The AMI Megaprocessor is seen here and is at ID 7.
Just to clarify, this server (CLUSTER11) that has the Megaprocessor thingy keeps having unexpected shutdowns once every 2-3 days or so. I am trying to find out why this is happening. These shutdowns only happen when the other server (CLUSTER12) is up and running. When CLUSTER12 is shut off, CLUSTER11 is okay.

[Martin Breidenbach]

I've installed several clusters using 3si controllers and RS/12 storage units. But I don't remember whether the 3si of the 'other' server is visible in NetRaid assistant or not.

SCSI ID 5 is the SAF-TE management processor. SCSI ID 6 in one server and ID 7 in the other server are the 3si controllers.

Which firmware do the 3si controllers have ? W.01.30 ?

(BTW these NetRAID controllers are AMI MegaRAID controllers with HP firmware)

[Benny Koh]

So are you saying that it is normal for the 3Si controllers be ID6 in one server and ID7 in the other? No SCSI conflicts here?
Just curious, what will happen in a SCSI conflict?

Both 3si controllers are using W.01.30.

[Martin Breidenbach]

Yes - both are on the same SCSI bus so the SCSI address of one 3SI *MUST* be changed. If both would use ID 7 there would be a SCSI ID conflict and it wouldn't work.

There are HP documents that describe how to configure these clusters:

HP NetServer LPr with HP Rack Storage/12
Windows 2000 Advanced Server Cluster Configuration Guide
http://h200005.www2.hp.com/bc/docs/support/SupportManual/lpn11438/lpn11438.pdf

HP Netserver Rack Storage/12 Cluster Install Guide Supplement
http://h200005.www2.hp.com/bc/docs/support/SupportManual/lpn11576/lpn11576.pdf

[kris rombauts · Accepted solution]

Hello Benny,

here is the .inf file (dummy device) you need to load to get rid of the message at each reboot.

HTH

regards, Kris

[Benny Koh]

Kris, you have hit the nail on the head!
This file is exactly what I wuz looking for. Now I know that this strange thing is actually the NETraid 3si cluster processor. However, why is this processor seen only on one server but not the other?

[Sachin_16]

one cause could be that one server has the SCSI processor driver, while other does not.
or
(guessing) one server has netraid cluster enabled, and second does not.