HELP ME ! Driver detected a controller error on \device\scsi\cpqkgpsa1(EventId 11)

[M. Gjaltema]

In the event viewer I've an event :

The driver detected a controller error on \device\scsi\cpqkgpsa1 (Event Id 11).

After this the machine loses connection to the disk. HP says "maybee it is a lun drop, you need to check all your settings". I think the settings are all OK. Other servers attached to the same ma8000/hsg80's are running fine.

This machine is a Windows 2003 machine, booting from the SAN.

Can you body help me?

[Marino Meloni_1]

I suggest you to connect to the switch and to check the statistic for the port where the server is connected, a communication problem (due to switch, GBic, cable,HBA )can cause this errror

[M. Gjaltema]

Thanks for the response.

I've checked all the connections, but they seems to be OK. I've attached a supportShow of the switches to this reply.

I have the feeling this more software/driver related.

I've seen different types of drivers on the HP site, can this be an issue?

http://h20000.www2.hp.com/bizsupport/TechSupport/DriverDownload.jsp?locale=en_US&pnameOID=315735&taskId=135&prodTypeId=12169&prodSeriesId=315733&submit.y=7&submit.x=4〈=en&cc=us&swEnvOID=1005

There is a "normal" driver and a "Boot from the SAN" driver. Does anyone knows the difference between those two?

[Anders Trolle]

I have had a similar problem. I solved it by updating the firmware on the HBA's and updating the driver.

[M. Gjaltema]

Anders,

Where you also running Windows 2003 in a Boot from the SAN configuration?

Thank you,

Marcel

[dentistcorn]

Give me the stats on your HBA:

Platform kit used or

- Driver version
- BIOS
- Firmware
- Registry settings for your HBA.

Do not use platfrom Kit higher than B for the CPQKGPSA HBA.

[M. Gjaltema]

Hello gvalor,

First of all, have installed "hp StorageWorks Windows NT/2000/Server 2003 Kit V8.7B for Enterprise Modular Array".

The versions ....

Driver version : 5-4.82a16
BIOS version : 1.63a1
Firmware version : 3.91a1

Registry settings ...

NodeTimeOut=10;
LinkTimeOut=40;
HlinkTimeOut=5;
Enabledpc=1;
queuetarget=1;
queuedepth=15;
Topology=1;
ElsRetryCount=6;
SimulateDevice=1;
ResetTPRLO=2;
EmulexOption=0xBA00;
ErrRetryMax=0

Thank you for your help/time!

[Adam White_4]

I have seen these messages in the past under Windows 2000, caused by either

1) bad GBIC in the switch port or HBA (or a damaged fibre cable)
2) non-DILX'd physical disk in the array
3) incorrect TPRLO value in the registry (I still haven't worked out what the best practice is for these)
and/or
4) HP Storage Agent service (apparently causes this to occur and turning it off resolves the issue)

HTH

[Bob Itguy]

The Microsoft KB also gives a couple of suggestions in terms of making sure multiple hosts aren't mounting the same volume and possibly moving any pagefile on the SAN to the local system.