HP-UX with Secure Path anf HSG 60 raid CTRL can not see lun great 4

[Boris Sedash]

Hello.

I have HP-UX with Secure path and HSG 60
raid controller.

The HSG60 raid have total 8 luns
Lun0, Lun1, Lun2, Lun3,Lun 10, Lun 11, Lun 12, Lun 13

The HP see only first 4 luns as show in figure:
/0/1/1.15.0 disk HP 18.2GMAN3184MC
0/0/2/1.15.0 disk HP 18.2GMAN3184MC
255/255/0/0.0 disk HSG60
255/255/0/0.1 disk HSG60
255/255/0/0.2 disk HSG60
255/255/0/0.3 disk HSG60

Did any one know how I can cause to HP recognize the 4 adational luns 11-13 ?
Thanks.

[Stephen Kebbell · Accepted solution]

Hi,

could you post (as a Text-File attachment) the output of a SHOW THIS command on your HSG60?
Looks like you're running in SCSI-2 mode, which only allows you to see LUNs from 0 to 7. Changing to SCSI-3 mode is not so simple! You'll also have to remove LUN0 and define a different unit number with the same storage set (should be possible without data loss)
The host connections must also have the OS type set to HP_VSA, which is only available with ACS 8.7. A SHOW THIS command will get most of this information.
What version of Securepath are you running?

Regards,
Stephen

[Boris Sedash]

Hello.
Thank you for answer.

There is show this output

Controller:
HSG60 ZG03901307 Software V85L-0, Hardware E12
NODE_ID = xxxx-xxxx-xxxx-xxxx
ALLOCATION_CLASS = 1
SCSI_VERSION = SCSI-3
Configured for MULTIBUS_FAILOVER with ZG03xxxxxx
In dual-redundant configuration
Device Port SCSI address 7
Time: NOT SET
Command Console LUN is lun 0 (IDENTIFIER = 88)
Host PORT_1:
Reported PORT_ID = xxxx-xxxx-xxxx-xxxx
PORT_1_TOPOLOGY = FABRIC (fabric up)
Address = xxxxxx
Host PORT_2:
Reported PORT_ID = xxxx-xxxx-xxxx-xxxx
PORT_2_TOPOLOGY = FABRIC (fabric up)
Address = xxxxxx
NOREMOTE_COPY
Cache:
128 megabyte write cache, version 0022
Cache is GOOD
No unflushed data in cache
CACHE_FLUSH_TIMER = DEFAULT (10 seconds)
Mirrored Cache:
128 megabyte write cache, version 0022
Cache is GOOD
No unflushed data in cache

As you can see I use the SCSI-3 protocol
the connetion type also set to VMS
Can you axplain the changes with Lun0

[Uwe Zessin]

As you are already running SCSI-3 it is unlikely that you have a unit "D0", but you can check it anyway:

> show D0

You should change the connection type to HP instead of VMS.

[Boris Sedash]

Thanks.

I don`t have D0 unit it start from
D1-D4 and next D10-13.
It is possible that problem in "emty"
lun`s between D4 and D13 ?

[Uwe Zessin]

No, sorry. The problem is that the OS_TYPE=HP only allows 8 LUNs. As your controllers are running in multibus_failover mode, you usually have 4 paths to a single unit. If you can live with two paths, I suggest you raise the UNIT_OFFSET to 9 on both connections for controller ports 2. Then you will access D1..D3 through Port_1 and D10..D13 through Port_2.