SUMMARY: Problems with DEC DSP 3210 2.05 Gb drive on a SS-2 -> SOLVED

From: Marcel Bernards (bernards@ecn.nl)
Date: Wed Nov 30 1994 - 05:31:10 CST


----------
X-Sun-Data-Type: text
X-Sun-Data-Description: text
X-Sun-Data-Name: text
X-Sun-Charset: us-ascii
X-Sun-Content-Lines: 58

I got this one solved on the SS-2 and on a SLC with the same
symptoms.

My original question:

----- Begin Included Message -----
Dear Sun mgr's

I have a SS-2 configured like this:

SS-2 with Weitek accelerator CPU
Artecon SBUS expansion box

I hooked up a DEC DSP 3210 2.05 Gb drive on the Artecon SCSI port esp1
and this drive seems to act flakey on this controller.
I removed another Seagate drive from that bus and let the DEC appear alone on the
SCSI bus.

these are the messages I get:
Nov 21 03:15:55 lotus vmunix: esp1: bad sequence step (0x6) in selection
Nov 21 03:15:55 lotus vmunix: sd4: SCSI transport failed: reason 'reset': retrying command
Nov 21 03:16:15 lotus vmunix: esp1: bad sequence step (0x6) in selection
Nov 21 03:16:15 lotus vmunix: sd4: SCSI transport failed: reason 'reset': retrying command
...

Are there known caveats using these DEC drives on a Sun ?

----- End Included Message -----

Thanx to all who replied:
szh@zcon.com (Syed Zaeem Hosain)
bobr@houston.wireline.SLB.COM ( Bob Reardon )
David M. Di Gioia <David.Digioia@gain.com>
David Burwell <dburwell@telecom.telecom.com>
rloh@asl.com.hk ("Raymond Loh")

David Di Gioia and Raymond Loh were right.
There exist a SunOS/Solaris patch for this problem
I installed it and all trouble dissapeared at once
So it was not a cable problem nor the SBUS extender at all.

>patch 101245-01 on Solaris 1.1 and 1.1c.
>patch 101229-02 on Solaris 2.1
>patch 101228-02 on Solaris 2.2

The system patch I got from ugle.unit.no was much newer:
101245-06.README
101245-06.tar.Z

I'll include the 101245-06.README

Thanx again,

Marcel Bernards, UNIX & Net sysadm Netherlands Energy Research Foundation ECN
(and SURFnet IC/ICP), Phone: (+31 /0)2246 4579 Fax: (+31 /0)2246 1864
E-Mail: Bernards@ECN.NL, SnailMail: P.O. Box 1, 1755 ZG Petten
SCREAMNet : AAAAAARGHH!HUH?? : Disclaimer: "The AntiChrist is the Computer !"
----------
X-Sun-Data-Type: readme-file
X-Sun-Data-Name: 101245-06.README
X-Sun-Charset: us-ascii
X-Sun-Content-Lines: 90

Patch-ID# 101245-06
Keywords: sequence step fas-101 scsi esp xbox 600MP Storage Technology 4280
Synopsis: SunOS 4.1.3: Merging SES/B related and "illegal bit" fixes
Date: Aug/31/94

Solaris Release: 1.1, 1.1C

SunOS Release: 4.1.3, 4.1.3C

Unbundled Product:

Unbundled Release:

Relevant Architectures: sun4m, sun4c

BugId's fixed with this patch: 1140764 1128862 1096405 1136580 1096405 1136580 1161712 1097817

Changes incorporated in this version: 1096405

Patches accumulated and obsoleted by this patch: 100906-01

Patches which conflict with this patch:

Patches required with this patch:

Obsoleted by:

Files included with this patch: esp.o , espvar.h , espreg.h

Problem Description:

Bug 1096405:

When running 4.1.3U1 on an SS10,FSBE with 1.3GB SCSI or 669MB PBox disk and
running the /usr/lib/acct/diskusg on the disk connected to FSBE, the
kernel reports the following error:
 
        "esp2: ILLEGAL bit set etc.etc"

This bug had an incomplete fix in the -02 release of this patch and with
this rev of patch, both the FAS236 and FAS101 chips have the delay
parameters correctly set.

(From 101245-05)

Bug 1097817: With a Storage Technology 4780 connected, esp interface returns
             PKT_RESID NEGATIVE, Restored pointers

(From 101245-04)

Bug 1161712: Machine is crashing with watchdog resets intermittently

(From 101245-03)

Bug 1136580: Heavily loaded SS600MP with DSBE/S on SES/B panics

(From 101245-02)

Bug 1096405:

When running 4.1.3 on an SS2,FSBE with 1.3GB SCSI or 669MB PBox disk and
running the /usr/lib/acct/diskusg on the disk connected to FSBE, the
kernel reports the following error:

        "esp2: ILLEGAL bit set etc.etc"

(From 101245-01):

Bugs 1140764/1128862:

The FAS-101 is falsely reporting a "bad sequence step 7" instead of the normal
sequence step of 4. The causes the SCSI peripheral to fail on that I/O with the
potential of not being able to boot the system or install it

Patch Installation Instructions:

             mv /sys/`arch -k`/OBJ/esp.o /sys/`arch -k`/OBJ/esp.o.FCS
        mv /sys/scsi/adapters/espvar.h /sys/scsi/adapters/espvar.h.FCS
        mv /sys/scsi/adapters/espreg.h /sys/scsi/adapters/espreg.h.FCS

            cp `arch -k`/esp.o /sys/`arch -k`/OBJ/esp.o
        cp `arch -k`/espvar.h /sys/scsi/adapters/espvar.h
        cp `arch -k`/espreg.h /sys/scsi/adapters/espreg.h

 Please Refer to the System and Networking Administration Manual
             section on building and installing a custom kernel.
 
IMPORTANT NOTE: This patch works for both normal and SES/B scsi configurations.



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:09:16 CDT