From nobody@FreeBSD.ORG  Mon Sep 11 17:46:01 2000
Return-Path: <nobody@FreeBSD.ORG>
Received: by hub.freebsd.org (Postfix, from userid 32767)
	id 3FA3737B423; Mon, 11 Sep 2000 17:46:01 -0700 (PDT)
Message-Id: <20000912004601.3FA3737B423@hub.freebsd.org>
Date: Mon, 11 Sep 2000 17:46:01 -0700 (PDT)
From: dwstevens@conduit-it.com
Sender: nobody@FreeBSD.ORG
To: freebsd-gnats-submit@FreeBSD.org
Subject: BTX HALTED before loader
X-Send-Pr-Version: www-1.0

>Number:         21214
>Category:       misc
>Synopsis:       BTX HALTED before loader
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    rnordier
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Mon Sep 11 17:50:01 PDT 2000
>Closed-Date:    Fri Sep 15 16:35:07 PDT 2000
>Last-Modified:  Fri Sep 15 16:40:01 PDT 2000
>Originator:     David W. Stevens
>Release:        FreeBSD 4.1 - stable
>Organization:
Conduit Internet Technologies
>Environment:
FreeBSD hobbes.conduit-it.com 4.1-STABLE FreeBSD 4.1-STABLE #2: Wed Aug 23 10:50:32 EDT 2000     dwstevens@hobbes.conduit-it.com:/us
r/src/sys/compile/HOBBES  i386
>Description:
I have an external raid controller/array that I'm connecting to 2 Dell Poweredge 2450's with Adaptech 2940 U2W controllers.  Channel 2 of the raid array is mapped to logical drive 0 on id 8, and channel 3 of the raid array (4 channels 0-contoller cache, 1 drive array, 2 host 1, 3 host 2) is mapped to logical drive 1 on id 10.  The first host will boot fine when connected to channel 2, but the second host will fail when connected to channel 3 of the raid array.  I get the following error after hitting F1 FreeBSD (FreeBSD boot manager):

int=00000000 err=00000000 efl=00030246 eip=00001654
eax=00000000 ebx=0000038e ecx=00000000 edx=00000000
esi=00009d3e edi=00001a3e ebp=00000396 esp=0000038a
cs=ce00 ds=0040 es=9d3e fs=0000 gs=0000 ss=9d3e
cs:eip=f7 f1 33 d2 8a 4e f6 f2-f1 3d ff 03 76 03 b8 ff
ss:esp=00 00 3f 00 00 00 00 00-00 00 02 00 df 09 00 ce
BTX HALTED

If i switch the cables (channel 2 to host 2, channel 3 to host 1) which ever host that is connected to channel 3 dumps that error message.  

I've checked cables, termination, adaptec bios settings, etc.

This configuration was tested under winNT 4.0 and Solaris 2.8. 

The kernel is stable, however I did not build world, so BTX is release.


>How-To-Repeat:
reboot, reboot ...

>Fix:


>Release-Note:
>Audit-Trail:
Responsible-Changed-From-To: freebsd-bugs->rnordier 
Responsible-Changed-By: rnordier 
Responsible-Changed-When: Tue Sep 12 16:52:09 PDT 2000 
Responsible-Changed-Why:  
I'll handle this. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=21214 

From: Robert Nordier <rnordier@nordier.com>
To: dwstevens@conduit-it.com
Cc: freebsd-gnats-submit@FreeBSD.ORG
Subject: Re: misc/21214: BTX HALTED before loader
Date: Fri, 15 Sep 2000 21:08:16 +0200 (SAST)

 > >Number:         21214
 > >Category:       misc
 > >Synopsis:       BTX HALTED before loader
 
 > >Description:
 > I have an external raid controller/array that I'm connecting to 2
   Dell Poweredge 2450's with Adaptech 2940 U2W controllers.  Channel
   2 of the raid array is mapped to logical drive 0 on id 8, and
   channel 3 of the raid array (4 channels 0-contoller cache, 1 drive
   array, 2 host 1, 3 host 2) is mapped to logical drive 1 on id 10.
   The first host will boot fine when connected to channel 2, but the
   second host will fail when connected to channel 3 of the raid array.
   I get the following error after hitting F1 FreeBSD (FreeBSD boot
   manager):
 > 
 > int=00000000 err=00000000 efl=00030246 eip=00001654
 > eax=00000000 ebx=0000038e ecx=00000000 edx=00000000
 > esi=00009d3e edi=00001a3e ebp=00000396 esp=0000038a
 > cs=ce00 ds=0040 es=9d3e fs=0000 gs=0000 ss=9d3e
 > cs:eip=f7 f1 33 d2 8a 4e f6 f2-f1 3d ff 03 76 03 b8 ff
 > ss:esp=00 00 3f 00 00 00 00 00-00 00 02 00 df 09 00 ce
 > BTX HALTED
 
 It would probably be useful if you could send the following output
 
     dd if=/dev/mem bs=1024 skip=816 count=16 | gzip > foo.gz
 
 (best send it to me directly rather than the whole list).  A division
 by zero exception is apparently happening, and the output may throw
 some light on this.
 
 --
 Robert Nordier
 
 rnordier@nordier.com
 rnordier@FreeBSD.org
 
State-Changed-From-To: open->closed 
State-Changed-By: rnordier 
State-Changed-When: Fri Sep 15 16:35:07 PDT 2000 
State-Changed-Why:  
From further information gathered by the originator, it seems most 
likely this is a hardware or firmware problem. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=21214 

From: Robert Nordier <rnordier@nordier.com>
To: "David W. Stevens" <dwstevens@conduit-it.com>
Cc: freebsd-gnats-submit@FreeBSD.org
Subject: Re: misc/21214: BTX HALTED before loader
Date: Sat, 16 Sep 2000 01:31:28 +0200 (SAST)

 Thanks for the feedback.  The kind of error made me wonder if it
 wasn't a more general problem than just BTX, and I think it makes
 sense to close the PR in the light of the additional info.
 
 --
 Robert Nordier
 
 rnordier@nordier.com
 rnordier@FreeBSD.org
 
 
 David W. Stevens wrote:
 
 > I seem to have gotten around the problem by removing the mapping to the
 > second logical drive in the array and splitting the first logical drive into
 > multiple partitions then mapping them to the separate channels.  Both
 > systems now boot fine.  I'm wondering if it's not my RAID hardware that was
 > causing the problem.  I submitted a report to my raid vendor and they were
 > able to reproduce the same type of error on windows 2000 and solaris 2.7, so
 > I can be pretty sure that it was something in the controller itself.  Though
 > sometimes win2k would boot fine and see the drive (that's what was throwing
 > me off and made me think it was something with FreeBSD).
 > 
 > If you still want the output from the dd command below let me know and I'll
 > get it too you.  I'm able to reproduce the error anytime now so if you want
 > to play with it let me know.
 > 
 > Thanks for your help Robert, it's good to know that support is out there!
 > 
 > Dave
 > --
 >  David W. Stevens
 >  CTO, VP - Internet Development
 >  Conduit Internet Technologies, Inc.
 >  dwstevens@conduit-it.com
 >  800.493.5045
 >  814.867.8248 fax
 > 
 >  Conduit-IT: Powering Successful eBusiness
 >  http://www.conduit-it.com/
 > 
 > 
 > > From: Robert Nordier <rnordier@nordier.com>
 > > Date: Fri, 15 Sep 2000 21:08:16 +0200 (SAST)
 > > To: dwstevens@conduit-it.com
 > > Cc: freebsd-gnats-submit@FreeBSD.ORG
 > > Subject: Re: misc/21214: BTX HALTED before loader
 > > 
 > >>> Number:         21214
 > >>> Category:       misc
 > >>> Synopsis:       BTX HALTED before loader
 > > 
 > >>> Description:
 > >> I have an external raid controller/array that I'm connecting to 2
 > > Dell Poweredge 2450's with Adaptech 2940 U2W controllers.  Channel
 > > 2 of the raid array is mapped to logical drive 0 on id 8, and
 > > channel 3 of the raid array (4 channels 0-contoller cache, 1 drive
 > > array, 2 host 1, 3 host 2) is mapped to logical drive 1 on id 10.
 > > The first host will boot fine when connected to channel 2, but the
 > > second host will fail when connected to channel 3 of the raid array.
 > > I get the following error after hitting F1 FreeBSD (FreeBSD boot
 > > manager):
 > >> 
 > >> int=00000000 err=00000000 efl=00030246 eip=00001654
 > >> eax=00000000 ebx=0000038e ecx=00000000 edx=00000000
 > >> esi=00009d3e edi=00001a3e ebp=00000396 esp=0000038a
 > >> cs=ce00 ds=0040 es=9d3e fs=0000 gs=0000 ss=9d3e
 > >> cs:eip=f7 f1 33 d2 8a 4e f6 f2-f1 3d ff 03 76 03 b8 ff
 > >> ss:esp=00 00 3f 00 00 00 00 00-00 00 02 00 df 09 00 ce
 > >> BTX HALTED
 > > 
 > > It would probably be useful if you could send the following output
 > > 
 > > dd if=/dev/mem bs=1024 skip=816 count=16 | gzip > foo.gz
 > > 
 > > (best send it to me directly rather than the whole list).  A division
 > > by zero exception is apparently happening, and the output may throw
 > > some light on this.
 > > 
 > > --
 > > Robert Nordier
 > > 
 > > rnordier@nordier.com
 > > rnordier@FreeBSD.org
 > > 
 > 
 
>Unformatted:
