SCSI problem with Seagate Hawk & SS 10

Article View: comp.sys.sun.hardware

Article #2381

Back to comp.sys.sun.hardware Previous Next

#2381

From: Thomas Tornblom
Date: Wed, 03 Jan 1996 20:28

32 lines
1639 bytes

I have been chasing a problem with a SparcStation 10 that recently got
a new disk, a Seagate Hawk 4.3G.

The machine runs a pretty stock SunOS 4.1.3 kernel.

The symptoms show up as lousy performance and the logs cluttered with
entries like these:
Jan  3 21:12:26 tellus vmunix: esp0: bad sequence step (0x6) in selection
Jan  3 21:12:26 tellus vmunix: sd1:    SCSI transport failed: reason 'reset': retrying command

I just tried to switch off some scsi options, and when I switched
synch SCSI off in the kernel (scsi_options &= ~0x20), the problem
vanished.

This system has two Sun 1.05G drives, one QIC-150, one Exabyte 85xx
and one CDROM besides this disk. I have tested with just one 1.05G
disk (/) and the Hawk, and that made absolutely no difference.

I also hooked the disk to my brand spanking new SparcBook running
Solaris 2.4, and it found no problems with it.

Is this a problem with the drive, or the kernel?

Thomas
-- 
Real life:      Thomas Törnblom           Email:  Thomas.Tornblom@Nexus.SE
Snail mail:     Combitech Nexus AB        Phone:  +46 18 171814
                Box 857                   Fax:    +46 18 696516
                S - 751 08 Uppsala, Sweden

Unix is simple, but it takes a genius to understand its simplicity. -DMR

Technical Details

Message-ID: <THOMAS.TORNBLOM.96Jan3212840@bond.Nexus.SE>

Path: arc1-ca.pugleaf.net!bond.Nexus.SE!usenet

Show Full Headers

Complete Article Headers

Path: arc1-ca.pugleaf.net!bond.Nexus.SE!usenet
From: Thomas.Tornblom@Nexus.SE (Thomas Tornblom)
Newsgroups: comp.sys.sun.hardware,comp.periphs.scsi
Subject: SCSI problem with Seagate Hawk & SS 10
Date: 03 Jan 1996 20:28:40 GMT
Organization: Nexus AB
Distribution: world
Message-ID: <THOMAS.TORNBLOM.96Jan3212840@bond.Nexus.SE>
NNTP-Posting-Host: bond.nexus.se