NetNews Usenet Archive 1992 #20/NN_1992_20.iso/spool/comp/sys/sun/admin/6217

Internet Message Format | 1992-09-09 | 1.6 KB

Xref: sparky comp.sys.sun.admin:6217 comp.sys.sun.hardware:4298
Newsgroups: comp.sys.sun.admin,comp.sys.sun.hardware
Path: sparky!uunet!usc!wupost!cs.utexas.edu!torn!cunews!nrcnet0!emr1!craven
From: [e-mail] (Jim Craven)
Subject: format vs. fsck on a SCSI disk
Message-ID: <[e-mail]>
Followup-To: comp.sys.sun.admin
Sender: [e-mail]
Nntp-Posting-Host: cocos.cg.emr.ca
Reply-To: [e-mail]
Organization: Dept. of Energy, Mines & Resources, Canada
Date: Wed, 9 Sep 1992 13:06:33 GMT
Lines: 30

We have been getting some errors on a SUN 1.3 GB SCSI disk when doing
an fsck or newfs.  These errors are, I believe, at kernel level and
appear in the form:

sd2c: Error for command 'read'
    : Block 1072, Absolute Block 1105
    : Vendor 'SEAGATE' error code 0x12

If I use any of the format/analyse subprograms such as read, test or
compare on the blocks mentioned above, or in fact the whole disk, I do
not get the the above error message nor do they flag any bad blocks.

Why should format not have a problem where fsck and newfs do?

Also, the Sun sys admin guide mentions when specifying a scsi disk
block for format, one should use bfi addressing (byte-from-index).  Can
someone explain bfi?

We are running 4.1.2 on a Sparc 2.  There is also a presto card, but the
problem persists without the presto.

Thanks in advance,

Jim
---
Love is like the ozone layer,            |Jim Craven ([e-mail])|
   You don't miss it until it's gone.    |Geological Survey of Canada  |
        ---Frank Drebbin, Police Squad   |1 Observatory Cres., Ottawa  |
-----------------------------------------|K1A 0Y3 [phone]-------|