Two weeks ago, I asked about a SCSI Problem after upgrading to Solaris
2.4 where a warning message (LUN not ready) was displayed during boot
time of a SPARCserver 20.
I received two answers:
From: albert@esther.rad.tju.edu (Dr. Micheal Albert)
[He described a similar problem he had with SunOS 4.1_3_U1 after
attaching a digital scanner to the scsi bus. After he added two disks
to the same bus later, the error messages went away.]
From: Kevin.Sheehan@uniq.com.au (Kevin Sheehan {Consulting Poster Child})
[He assumed that it was a problem with the particular drive (and
not with cabling or termination. He also suggested to turn off tagged
queueing to see if it is a problem with that.]
Thanks for your answers, especially for the hint to switch tagged
queueing off. I tried this and the problem disappeared: the message is
no longer displayed during boot. I now run the server without tagged
queueing because I think one of the drives has a problem with
it. Currently the server has one SCSI controller only, so I simply
added
set scsi_options & ~0x80
to /etc/system. After adding another SCSI controller to the server, I
will reinvestigate the problem and ensure to disable tagged queueing
only on the controller requiring it (using /kernel/drv/esp.conf)
[see the Solaris 2 FAQ (http://www.fwi.uva.nl/pub/solaris/solaris2.html)
for further information on turing tagged queueing off].
Original question:
>I just upgraded a SPARCserver 20 from Solaris 2.3 to Solaris 2.4. Now
>I'm seeing the following SCSI error sometime, especially during the
>fsck calls during boot:
>
>Sep 24 19:28:46 petri.isa.de unix: WARNING: /iommu@f,e0000000/sbus@f,e0001000/espdma@f,400000/esp@f,800000/sd@0,0 (sd0):
>Sep 24 19:28:46 petri.isa.de unix: Error for command 'read' Error Level: Retryable
>Sep 24 19:28:46 petri.isa.de unix: Requested Block 16, Error Block: 263461
>Sep 24 19:28:46 petri.isa.de unix: Sense Key: Not Ready
>Sep 24 19:28:46 petri.isa.de unix: Vendor 'IBM':
>Sep 24 19:28:47 petri.isa.de unix: ASC = 0x4 (LUN not ready), ASCQ = 0x1, FRU = 0x0
>
>Together with the software update, I added a new internal Sun 535MB
>disk (SCSI id 1) which was originally built into a SPARC 5 (this disk
>should work in a SPARC 20, too, shouldn't it?). The other disks are as
>follows ("Silberdisk 4" is an IBM OEM 0664M1H 2GB disk):
>
>ERR-> 0. c0t0d0 <Silberdisk 4 cyl 2878 alt 2 hd 15 sec 91>
> /iommu@f,e0000000/sbus@f,e0001000/espdma@f,400000/esp@f,800000/sd@0,0
>NEW-> 1. c0t1d0 <SUN0535 cyl 1866 alt 2 hd 7 sec 80>
> /iommu@f,e0000000/sbus@f,e0001000/espdma@f,400000/esp@f,800000/sd@1,0
> 2. c0t2d0 <Silberdisk 4 cyl 2878 alt 2 hd 15 sec 91>
> /iommu@f,e0000000/sbus@f,e0001000/espdma@f,400000/esp@f,800000/sd@2,0
> 3. c0t3d0 <SUN1.05 cyl 2036 alt 2 hd 14 sec 72>
> /iommu@f,e0000000/sbus@f,e0001000/espdma@f,400000/esp@f,800000/sd@3,0
>
>What does the above SCSI error mean? Is it critical? Is it a problem
>with the SCSI bus (length, termination)? Or with the disk
>configurations? Since we run Solaris 2.4 on several other machines
>without any problems, I doubt it is a software problem related to the
>upgrade, but who knows? (All Solaris 2.4 mandatory and the Recommended
>Patches from the SunSolve 2.6.3 CD are installed).
-- Andreas Luik E-Mail: luik@isa.de (postmaster@isa.de) WWW: http://www.isa.de/~luik
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:10:34 CDT