Subject: Re: problems with NCR with multiple drives
To: None <current-users@netbsd.org>
From: Michael C. Richardson <mcr@sandelman.ottawa.on.ca>
List: current-users
Date: 08/17/1998 00:22:30
-----BEGIN PGP SIGNED MESSAGE-----


>>>>> "Michael" == Michael Richardson <mcr@sandelman.ottawa.on.ca> writes:
    Michael>   I just added a new drive to my file server. A 4Gb Quantum VIKING (SCA
    Michael> with an adaptor to Ultra+Wide).
    Michael>   My suspicion is that there is some parameter that needs tuning on
    Michael> the NCR driver.... I have 6 devices, counting LUNs and adaptor, 8.
    Michael> The system hangs at the end of the boot sequence with:
    Michael> ncr0: no ccb.
    Michael> sd3: not queued, error 5

  BTW: my NetBSD is as of July 27. I'm sup'ing again now.
  I reinstall my boot blocks to fix things up, and as I had a UVM kernel
hanging around waiting to be installed, I switched to that as well. My
boot blocks had been telling me that /netbsd was of an inappropriate
file type for loading.
  
  While I was mucking around, testing things, I moved my drive at targ 15
("sd2" originally, the FUJITSU) to targ 8, and switched the new drive to targ
9 (QUANTUM)
  This was to make the drive order stay the same regardless of whether the
new drive was attached or not. Well, things work fine, but when I try
to access "sd3", I get:

istari# disklabel sd3
disklabel: /dev/rsd3d: Input/output error

  I noticed on boot that it said:

sd3 at scsibus0 targ 9 lun 0: <QUANTUM, VIKING 4.5 SCA, 880R> SCSI2 0/direct fixed
sd3(ncr0:9:0): WIDE SCSI (16 bit) enabled
sd3(ncr0:9:0): 40.0 MB/s (50 ns, offset 16)
sd3: drive offline

  Errno #5 is, btw, "input/output error"
  My conclusion: there is definitely something wrong with accessing sd3 on
my system. I note that I *was* able to fsck my FUJITSU when it was sd3, but
it died as soon as the system tried to really come up. 
  ... I'm now looking through sys/dev/pci/ncr.c, but I don't understand
why I'd be out of CCBs. This card is:

ncr0 at pci0 dev 19 function 0: ncr 53c875 fast20 wide scsi
ncr0: interrupting at irq 12
ncr0: minsync=12, maxsync=137, maxoffs=16, 128 dwords burst, large dma fifo
ncr0: single-ended, open drain IRQ driver, using on-chip SRAM
ncr0: restart (scsi reset).

  I have their 4.10.00 firmware. It appeers from their web site that there
isn't anything newer, but in any case, that should only affect boot, I
believe.  
  No, the drive doesn't have any jumpers for auto-spin up: it is an SCA
drive. If I go into the symbios config and ask to verify the disk, it works
fine, so *something* must be happening.

   :!mcr!:            |  Network and security consulting/contract programming
   Michael Richardson |         Firewalls, TCP/IP and Unix administration
 Personal: mcr@sandelman.ottawa.on.ca. PGP key available.
 Corporate: sales@sandelman.ottawa.on.ca. 
	ON HUMILITY: To err is human, to moo bovine.





-----BEGIN PGP SIGNATURE-----
Version: 2.6.3ia
Charset: latin1
Comment: Processed by Mailcrypt 3.4, an Emacs/PGP interface

iQB1AwUBNdewA9iXVu0RiA21AQEekAL9GKgCGcQujLyGqHcagbZtF/XKO0IM0o/v
bClz3j8cfPcQW5JUqw7VvMIp2rkTp7SGJlFZECgtuUfZmwW+jM0HJ6XngUaFIo6t
K9AZlO/JFdLr8WxI2ALHPnSv9/vAKu2c
=gUZB
-----END PGP SIGNATURE-----