Trying to Configure 9730 on HP-UX 11.0 getting SCSI errors

[nandrew]

Tape Library connected via HP High Voltage Differential SCSI A4800A card terminated at library. OS unable to see drives, following SCSI errors at boot time.

SCSI: Unexpected Disconnect -- lbolt: 10977, dev: cb050002, io_id: 5000010
SCSI: Unexpected Disconnect -- lbolt: 10978, dev: cb050002, io_id: 5000010
SCSI: Unexpected Disconnect -- lbolt: 10979, dev: cb050002, io_id: 5000010
SCSI: Unexpected Disconnect -- lbolt: 10980, dev: cb050002, io_id: 5000010
SCSI: Unexpected Disconnect -- lbolt: 10981, dev: cb050002, io_id: 5000010
SCSI: Unexpected Disconnect -- lbolt: 10982, dev: cb050002, io_id: 5000010
SCSI: Target detected parity error -- lbolt: 11067, dev: cb054002
scb->cdb: 12 00 00 00 80 00
SCSI: Target detected parity error -- lbolt: 11068, dev: cb054002
scb->cdb: 12 00 00 00 80 00
SCSI: Target detected parity error -- lbolt: 11069, dev: cb054002
scb->cdb: 12 00 00 00 80 00
SCSI: Target detected parity error -- lbolt: 11070, dev: cb054002
scb->cdb: 12 00 00 00 80 00
SCSI: Target detected parity error -- lbolt: 11071, dev: cb054002
scb->cdb: 12 00 00 00 80 00
SCSI: Target detected parity error -- lbolt: 11072, dev: cb054002
scb->cdb: 12 00 00 00 80 00
SCSI: Unexpected Disconnect -- lbolt: 16559, dev: cb050002, io_id: 500001f
SCSI: Unexpected Disconnect -- lbolt: 16560, dev: cb050002, io_id: 500001f
SCSI: Unexpected Disconnect -- lbolt: 16561, dev: cb050002, io_id: 500001f
SCSI: Unexpected Disconnect -- lbolt: 16562, dev: cb050002, io_id: 500001f
SCSI: Unexpected Disconnect -- lbolt: 16563, dev: cb050002, io_id: 500001f
SCSI: Unexpected Disconnect -- lbolt: 16564, dev: cb050002, io_id: 500001f
SCSI: Target detected parity error -- lbolt: 16656, dev: cb054002
scb->cdb: 12 00 00 00 80 00
SCSI: Target detected parity error -- lbolt: 16657, dev: cb054002
scb->cdb: 12 00 00 00 80 00
SCSI: Target detected parity error -- lbolt: 16658, dev: cb054002
scb->cdb: 12 00 00 00 80 00
SCSI: Target detected parity error -- lbolt: 16659, dev: cb054002
scb->cdb: 12 00 00 00 80 00
SCSI: Target detected parity error -- lbolt: 16660, dev: cb054002
scb->cdb: 12 00 00 00 80 00
SCSI: Target detected parity error -- lbolt: 16661, dev: cb054002
scb->cdb: 12 00 00 00 80 00

I Have no documentation for Library.

Thanx - nandrew...

 

[comtec17]

Have you configured the SCSI ID of the Robot and Drives on the front of the Library?

Are you using an HVD Terminator?

How are the drives daisy chained?

 

[nandrew]

I have configured SCSI ID's on the front of the library and using HVD terminator. Not sure about daisy chaining, Server connected to library, library connected to drive 0 then to 1 and 2. Drive 3 disconnected due to hardware error. I have used SCS ID's 0-4, 0 is drive 0 and 4 is library.

thanx nandrew

 

[comtec17]

From previous installs I have done with 9730's I always gave the robot ID 0 and Drives 1 - 4 ID's 1 - 4. At the end of the last drive on the chain, do you have it terminated?

 

[nandrew]

Tried setting SCSI ID's as above - no joy. Have got 9730 working by disconnecting drive 2, this leave 2 drives in silo. Have located a product manual, this mentions TERMPWR jumpers, suspect this could be the problem but currently unable to test this out.

Thanx for your assistance on this - nandrew

 

[Schotte]

The configuration is absolutely awfull if I understood it correctly. You can have one drive and a library on one scsi bus if you do not disable hardware compression in the application for the drive. You can have two drives on the same bus if you accept the knock-on effects of bus resets on drives that are running correctly, provided a) the customer accepts the effects of bus resets b) hardware compression is disabled for both drives.
You can have a third drive as long as it is only used when one of the other drives fall over.......not together.
If the drives are streaming drives (most modern drives are) , they must be able to stream......
assuming a F&W HVD drives and library, a single DLT7000 will need 10MB/Sec if compression only reaches 2:1 (more if better!). The F&W bus can only deliver 10MB/Sec minus handshaking, you have about 14.5MB/Sec of bandwidth, that leaves a max (still assuming 2:1 only) of around 4MB/Sec for the other drive. This just wears out the coating on the tape, dirties the heads and causes the cleaning tapes to be used up quickly. After about 12 months of this the cassettes will start to collapse, if the drives haven´t already..........