All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Justin T. Gibbs" <gibbs@scsiguy.com>
To: Michael Madore <mmadore@aslab.com>, linux-kernel@vger.kernel.org
Subject: Re: Adaptec 79xx > 1GB I/O errors
Date: Tue, 14 Jan 2003 17:18:01 -0700	[thread overview]
Message-ID: <741380000.1042589881@aslan.btc.adaptec.com> (raw)
In-Reply-To: <3E24A5EF.2060903@aslab.com>

> I have been getting the following I/O errors while stress testing a
> system with the latest Adaptec 79xx driver (1.3.0BETA2):
> 
> Jan 14 09:29:13 asl200 kernel: SCSI disk error : host 0 channel 0 id 0 lun 0 return code = 8000002
> Jan 14 09:29:13 asl200 kernel: Info fld=0x86c552, Deferred sd08:02: sense key Hardware Error
> Jan 14 09:29:13 asl200 kernel: Additional sense indicates Internal target failure
> Jan 14 09:29:13 asl200 kernel:  I/O error: dev 08:02, sector 487312
> 
> Jan 14 10:36:37 asl200 kernel: (scsi0:A:0:0): Locking max tag count at 64
> 
> This is with kernel 2.4.19 + 2.4.19rc5aa1.  I have tested with several
> different Ultra 320 drives, with the same result.

Different U320 drives of the same make/model, or different makes and models?
The driver cannot "fake" a device returning an internal target failure
error.  I have only seen this on certain U320 Seagate drives and my
understanding is that it is a drive firmware problem exposed under high
transaction loads.

The locking max tag count diagnostic is normal.  It means that the
driver has determined the maximum queue depth of your disk.

> If I remove memory from the machine so that it only has 1GB, then
> everything is solid as a rock.

I bet there is less I/O going to the drives too.

> If I plug a zero channel raid controller into the same system (dpt_i2o),
> then I don't get any I/O error regardless of the amount of RAM.

The RAID controller will eat these messages.  Remeber that it only
*emulates* a SCSI controller.  It doesn't always act like one.

> Any thoughts?

Beat up on your driver vendor?

--
Justin


      reply	other threads:[~2003-01-15  0:09 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-01-15  0:06 Adaptec 79xx > 1GB I/O errors Michael Madore
2003-01-15  0:18 ` Justin T. Gibbs [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=741380000.1042589881@aslan.btc.adaptec.com \
    --to=gibbs@scsiguy.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mmadore@aslab.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.