SUMMARY: disk & scsi bus errors

From: bill@ixcim.att.com
Date: Thu Dec 17 1992 - 05:13:39 CST


I'm sorry this summary is so late. Here's the original
question:

| I have attached a Pinnacle Micro REO-6500 optical
| jukebox to an IPC. The problem is that I keep on
| getting the following kinds of errors:
| -------------------------
| Oct 15 22:41:06 ixcim6 vmunix: sd0a: Error for command 'write'
| Oct 15 22:41:06 ixcim6 vmunix: sd0a: Error Level: Retryable
| Oct 15 22:41:06 ixcim6 vmunix: sd0a: Block 10528, Absolute Block: 10528
| Oct 15 22:41:06 ixcim6 vmunix: sd0a: Sense Key: Aborted Command
| Oct 15 22:41:06 ixcim6 vmunix: sd0a: Vendor 'MAXTOR' error code: 0x47
| -------------------------
| Oct 15 12:56:10 ixcim6 vmunix: esp0: Target 3 didn't disconnect after sending COMMAND COMPLETE
| Oct 15 12:56:10 ixcim6 vmunix: sd0: SCSI transport failed: reason 'tran_err': retrying command
| Oct 15 12:56:10 ixcim6 vmunix: jb0: SCSI transport failed: reason 'reset'
| -------------------------
|
| Any ideas as to what may be causing this and how I might
| fix it would be greatly appreciated.

The answer probably won't help anyone unless they have one of
these jukeboxes. Firmware level 7.114 has problems unloading
a disk. I received new firmware which fixed the problem and
all the errors stopped.

Here are the answers I did receive, in case they might help
someone else:

----- Begin Included Message -----

From: canuck@masc38.rice.edu (Mike Pearlman)
Content-Length: 334
Content-Type: text
To: bill@ixcim.att.com
Subject: Re: disk & scsi bus errors
X-Lines: 7

My guess based on the scenario is that you have a MAXTOR 8760 disk drive
since we have similar problems with them. If this guess is correct
        1) get the latest Sun specific version of the firmware from Maxtor
        2) use as short a cable as you can get
        3) use an active terminator versus a passive one

michael pearlman <canuck@rice.edu>

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

----- Begin Included Message -----

From: ups!upstage!glenn@fourx.Aus.Sun.COM (Glenn Satchell)
To: ups!fourx!ixcim.att.com!bill@fourx.Aus.Sun.COM
Subject: Re: disk & scsi bus errors
Content-Type: text
Content-Length: 2069
X-Lines: 48

Hi Bill,

Looks like a bad cable, bad connection, or bad scsi termination to me.
Even if there is other stuff on this scsi bus which is working fine you
can still have a bad cable or connection. I find that the small scsi
connectors on the suns are notorious for bad connections. Make sure
that it is not hanging down where it plugs in as this can cause a bad
connection on the top row of pins. If necessary prop it up a bit.

regards,

--
Glenn Satchell          ups!glenn@fourx.Aus.Sun.COM  |
Uniq Professional Services Pty Ltd  ACN 056 279 335  |  "The answer is no,
PO Box 70, Paddington, NSW 2021, (Sydney) Australia  |  and I'll negotiate
Phone: +61-2-360-7434           Fax: +61-2-331-2572  |  from there."
       "Sun Accredited System Consultants"           |

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

----- Begin Included Message -----

From: adam%bwnmr4@harvard.harvard.edu (Adam Shostack) Subject: Re: disk & scsi bus errors To: bill@ixcim.att.com Content-Type: text Content-Length: 1878 X-Lines: 43

Caveat Emptor: We had one of these buggers, and returned it to Pinnacle for three REO-650's, which don't work well either. The unit got flakier and flakier; it seems the device drivers are not well written.

Your actual best bet may be to get a few of the 650 drives like we did, if they will fit what you need. We found that we were spending upwards of 15 hours a week making the jukebox work, rebooting the machine it was attatched to, etc. Its easier to spend our time mounting and unmounting platters. (Of course, we do that via sudo now...)

I'd examine closely the justification for using a jukebox, the unit seemed to be about as robust as a snowball in hell.

Adam

-- Adam Shostack adam@bwnmr4.harvard.edu Systems Manager 617-732-7692 Surgical Planning Lab, Dept of Radiology Fax 732-7963 Brigham and Womens Hospital, Boston

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

----- Begin Included Message -----

From: ldavis!woden!gunn@snowbird.Central.Sun.COM (David Gunn) To: snowbird!ixcim.att.com!bill@snowbird.Central.Sun.COM Subject: Re: disk & scsi bus errors Content-Type: text Content-Length: 384 X-Lines: 8

Try swapping out your terminator.

====================================================================== | David Gunn Voice: (801) 375-0177 | | Larson-Davis, Inc. Fax: (801) 375-0182 | | 1681 W. 820 North | Provo, UT 84604 ldavis!gunn@snowbird.central.sun.com | ======================================================================

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

----- Begin Included Message -----

From: dank@teleng.telxon.com (Dan Kelley) Subject: Re: disk & scsi bus errors To: bill@ixcim.att.com X-Mailer: ELM [version 2.3 PL9P] Content-Type: text Content-Length: 2767 X-Lines: 51

Uhmm, I have the same problem with a Seagate WREN IX. My config is a secondary sbus SCSI card on a 670MP and on the bus is a Fujitsu 1.3G external disk, the Seagate WREN IX 2.0G external disk (both in the same external cabinet) and an Exabyte 8mm tape drive. My errors are as follows:

Oct 20 23:08:02 casper vmunix: sd6h: Error for command 'write' Oct 20 23:08:02 casper vmunix: sd6h: Error Level: Retryable Oct 20 23:08:02 casper vmunix: sd6h: Block 531872, Absolute Block: 1541312 Oct 20 23:08:02 casper vmunix: sd6h: Sense Key: Aborted Command Oct 20 23:08:02 casper vmunix: sd6h: Vendor 'SEAGATE' error code: 0x47 Oct 20 23:19:31 casper vmunix: sd6h: Error for command 'write(10)' Oct 20 23:19:31 casper vmunix: sd6h: Error Level: Retryable Oct 20 23:19:31 casper vmunix: sd6h: Block 1317184, Absolute Block: 2326624 Oct 20 23:19:31 casper vmunix: sd6h: Sense Key: Aborted Command

Look familiar? Recently I talked to a friend of mine who suggested looking at the SCSI terminators so just today, I ripped open both the disk cabinet and the tape drive. Seems the Seagate had internal terminator AND the tape drive had internal terminator. I removed both and put an external terminator on (I like to *SEE* the terminator - that way I know it is there :-). So far all seems ok so this could be your problem as well - double SCSI termination. It is worth a look.

Also, I would be interested in any responces you get from the mailing list (even though I am pretty certain this is the problem).

Dan...

-- Dan Kelley () ===== ===== = = = //// = = System Administrator () = === = = ////// == = dank@telxon.com () = ===== ===== = ////// = == Corp. (216)-867-3700 () //// Akron, OH

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



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:54 CDT