Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 6 Mar 1999 01:14:48 -0600
From:      Zach Heilig <zach@uffdaonline.net>
To:        "Kenneth D. Merry" <ken@plutotech.com>, Zach Heilig <zach@uffdaonline.net>
Cc:        scsi@FreeBSD.ORG
Subject:   Re: 'camcontrol rescan' wierdness
Message-ID:  <19990306011448.A1224@znh.org>
In-Reply-To: <199903060523.WAA04079@panzer.plutotech.com>; from Kenneth D. Merry on Fri, Mar 05, 1999 at 10:23:44PM -0700
References:  <19990305191914.A26718@znh.org> <199903060523.WAA04079@panzer.plutotech.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Mar 05, 1999 at 10:23:44PM -0700, Kenneth D. Merry wrote:
> > cd0 at ncr0 bus 0 target 5 lun 0
> > cd0: <MATSHITA CD-R   CW-7502 4.10> Removable CD-ROM SCSI-2 device
> > cd0: 10.000MB/s transfers (10.000MHz, offset 8)
> > cd0: Attempt to query device size failed: NOT READY, Medium not present - tray closed

> Okay, this is interesting.  Did you turn the power back on or something?

Yes.

> It appears that your Quantum disk is getting confused and responding on
> multiple LUNs.

> > and 'camcontrol devlist' now prints out:
> > <  >                               at scbus-1 target -1 lun -1 (xpt0)
> > <SEAGATE ST15230W SUN4.2G 0738>    at scbus0 target 0 lun 0 (pass0,da0)
> > <QUANTUM FIREBALL_TM2110S 300X>    at scbus0 target 1 lun 0 (pass1,da1)
> > <  >                               at scbus0 target 1 lun 1 (pass2)
> > <  >                               at scbus0 target -1 lun -1 ()
> > <MATSHITA CD-R   CW-7502 4.10>     at scbus1 target 5 lun 0 (cd0,pass3)
> > <IOMEGA ZIP 100 J.02>              at scbus1 target 6 lun 0 (pass4,da2)
> > <  >                               at scbus1 target -1 lun -1 ()
> > 
> > [note the bogus pass2 entry]
> 
> Hmm, I suppose that device didn't completely go away.

The CDROM? That was SCSI #5..  The bogusly found da3? probably..  da0 and da1
were accessable [and they didn't seem slow] while the several minute rescan
was happening.

> Can you try putting a quirk entry for the Quantum in cam_xpt.c to disable
> multi-LUN probing?  (CAM_QUIRK_NOLUNS)  That may fix the problem.

It doesn't happen on boot.  It also doesn't give problems if I avoid plugging
the CDROM in [which I no longer have a real need to do].  For example, I can
rescan now (I rebooted it to get rid of the bugus entries), and nothing
happens, other than

$ camcontrol rescan 0
Re-scan of bus 0 was successful
$ camcontrol rescan 1
Re-scan of bus 1 was successful
$ camcontrol rescan 0:1:0
Re-scan of 0:1:0 was successful
$ camcontrol rescan 0:1:1
Re-scan of 0:1:1 was successful

[no kernel messages or pauses].

$ camcontrol devlist
<  >                               at scbus-1 target -1 lun -1 (xpt0)
<SEAGATE ST15230W SUN4.2G 0738>    at scbus0 target 0 lun 0 (pass0,da0)
<QUANTUM FIREBALL_TM2110S 300X>    at scbus0 target 1 lun 0 (pass1,da1)
<  >                               at scbus0 target -1 lun -1 ()
<MATSHITA CD-R   CW-7502 4.10>     at scbus1 target 5 lun 0 (pass2,cd0)
<IOMEGA ZIP 100 J.02>              at scbus1 target 6 lun 0 (pass3,da2)
<  >                               at scbus1 target -1 lun -1 ()

I think I will suspect that NEC CDROM -- it doesn't play well with other
devices [it really slowed down access to the two disks -- it may not have
disconnected while reading CD-audio -- and it would only read at 1x].  The
only reason it was there is I had no other place to attach it, and I have 5 or
6 audio cd's that wedge the cd-writer [this has nothing to do with FreeBSD --
it wedges while reading the TOC].

> The other thing I wonder about is your bus topology.  You say you "removed 
> cd0".  Where was it on the SCSI chain?  How did you remove it?  Was the bus
> still terminated once you removed it?

yes...
        NEC CD  Seagate  Quantum  terminator
ahc0     cd0     da0      da1
  |-------+-------+--------+-------|
   [internal cable]

          CW-7502      zip [termination switch turned on]
ncr0      cd1/cd0      da2
  |---------| |---------|
   [external cables]

> Taking things off an active SCSI bus only works in controlled
> circumstances.  If the device removed was the one terminating the chain,
> you would be leaving the bus unterminated.

I yanked the power plug before removing the scsi connector...  There is a
separate terminator on the last connector.  It may have been a mistake to
avoid rebooting to temporarily add the CD-ROM... It surely didn't gain me
anything this time [I ended up rebooting twice anyway].

-- 
Zach Heilig <zach@uffdaonline.net>


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-scsi" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990306011448.A1224>