From rfg@tristatelogic.com  Fri Jul 11 19:24:22 2008
Return-Path: <rfg@tristatelogic.com>
Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34])
	by hub.freebsd.org (Postfix) with ESMTP id 4EFE4106564A
	for <FreeBSD-gnats-submit@freebsd.org>; Fri, 11 Jul 2008 19:24:22 +0000 (UTC)
	(envelope-from rfg@tristatelogic.com)
Received: from segfault-outgoing-helo.tristatelogic.com (112.171-60-66-fuji-dsl.static.surewest.net [66.60.171.112])
	by mx1.freebsd.org (Postfix) with ESMTP id 3FDB98FC14
	for <FreeBSD-gnats-submit@freebsd.org>; Fri, 11 Jul 2008 19:24:22 +0000 (UTC)
	(envelope-from rfg@tristatelogic.com)
Received: by segfault.tristatelogic.com (Postfix, from userid 1237)
	id 96CCD1142D; Fri, 11 Jul 2008 12:06:23 -0700 (PDT)
Message-Id: <20080711190623.96CCD1142D@segfault.tristatelogic.com>
Date: Fri, 11 Jul 2008 12:06:23 -0700 (PDT)
From: Ronald F. Guilmette <rfg@tristatelogic.com>
Reply-To: Ronald F. Guilmette <rfg@tristatelogic.com>
To: FreeBSD-gnats-submit@freebsd.org
Cc:
Subject: 7.0-RELEASE install dies (probably same problem as PR 113160)
X-Send-Pr-Version: 3.113
X-GNATS-Notify:

>Number:         125516
>Category:       i386
>Synopsis:       [install] 7.0-RELEASE install dies
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    gavin
>State:          open
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Fri Jul 11 19:30:01 UTC 2008
>Closed-Date:    
>Last-Modified:  Wed Dec 03 12:45:02 UTC 2008
>Originator:     Ronald F. Guilmette
>Release:        FreeBSD 7.0-RELEASE i386
>Organization:
Infinite Monkeys & Co. LLC
>Environment:
	7.0-RELEASE i386
>Description:
	This looks like the exact same problem as in PR 113160.  (I think that
	PR should be reopened.  See below.)

	I have an older Athlon XP system based on an ASUS A7N266-VM/AA
	motherboard, and when trying to install from the 7.0-RELEASE disk #1,
	the install process gets down past the Stage 3 boot and starts to boot
	the install kernel, which then recognizes all of the hardware, but
	then, just after it has done all of that, the process dies rather
	mysteriously and just puts up the standard mountroot (brief)
	instructions followed by a mountroot> prompt.

>How-To-Repeat:
	Just try to boot from a 7.0-RELEASE CD on an ASUS A7N266-VM/AA based
	system and I do believe you will see the problem.  I have tried to
	pin down the problem by using multiple different CDs and multiple
	different CD ROM drives (because I thought at first that the problem
	might be related to the CD or CD drive) and that just isn't the problem.
	It is clear that the real problem is 7.0's interaction with this
	specific motherboard.

>Fix:
	I believe that PR 113160 should be reopened and a proper sort of fix
	should be developed.  Just telling people to diable the 15M-16M
	memory hole is *not* a solution, in particular on _this_ motherboard,
	where ASUS, in its infinite wisdom, decided to "customize" the BIOS
	by, among other things, removing any such option.  To be clear, on
	this motherboard (ASUS A7N266-VM/AA) there simply *is* no option in
	the BIOS to disable the 15M-16M memory hole.  (I searched every one
	of the BIOS menus, twice, and it just ain't there.)  So as of now,
	all owners of A7N266-VM/AA based systems are screwed... none of us
	can get 7.0-RELEASE installed on a blank/fresh A7N266-VM/AA system.

	Surely there _must_ be a way to work around the presence of a memory
	hole at 15M-16M, yes?  I mean how hard can that be?  (I hate to have
	to just throw this motherboard in the trash!)

	Why is it that with a 6.1-RELEASE CD (and on the exact same hardware),
	I _can_ get down all the way to the Install menu with no crashing?
	What changed between 6.1 and 7.0 as regards to that 15M-16M range?
>Release-Note:
>Audit-Trail:
State-Changed-From-To: open->feedback 
State-Changed-By: gavin 
State-Changed-When: Fri Jul 11 20:06:16 UTC 2008 
State-Changed-Why:  
To submitter:  Firstly, is there any chance you could get a copy 
of the boot messages?  If you can't obtain them through a serial 
console, uploading good quality digital photos should be enough. 
Secondly, are you able to confirm if 6.2 and 6.3 CDs boot on 
this machine?  Downloading the 40 meg bootonly ISOs will be 
a sufficient test, and the results may well go a long way to 
figuring out exactly what was changed to cause this issue. 


Responsible-Changed-From-To: freebsd-i386->gavin 
Responsible-Changed-By: gavin 
Responsible-Changed-When: Fri Jul 11 20:06:16 UTC 2008 
Responsible-Changed-Why:  
Track 

http://www.freebsd.org/cgi/query-pr.cgi?pr=125516 

From: Gavin Atkinson <gavin.atkinson@ury.york.ac.uk>
To: bug-followup@FreeBSD.org
Cc:  
Subject: Re: i386/125516: 7.0-RELEASE install dies (probably same problem as
 PR 113160)
Date: Sat, 12 Jul 2008 18:42:57 +0100 (BST)

 Feedback from submitter:
 
 ---------- Forwarded message ----------
 Date: Fri, 11 Jul 2008 18:47:19 -0700
 From: Ronald F. Guilmette <rfg@tristatelogic.com>
 Subject: Re: i386/125516: 7.0-RELEASE install dies (probably same problem as PR 113160)
 
 
 In message <200807112010.m6BKAsnC085231@freefall.freebsd.org>, you wrote:
 
 >Synopsis: 7.0-RELEASE install dies (probably same problem as PR 113160)
 >
 >State-Changed-From-To: open->feedback
 
 See below regarding this change of state.
 
 >State-Changed-By: gavin
 >State-Changed-When: Fri Jul 11 20:06:16 UTC 2008
 >State-Changed-Why:
 >To submitter:  Firstly, is there any chance you could get a copy
 >of the boot messages?
 
 If you give me a workable methodology to do that... i.e. capture ALL OF
 THEM... then I will certainly try.
 
 >If you can't obtain them through a serial console,
 
 Given the other alternative you proposed, I _would_ like to try to do
 this, but I don't know how.  Can you help to guide me?
 
 I have just skimmed this:
 
     http://www.freebsd.org/doc/en/books/handbook/serialconsole-setup.html
 
 but it fails to answer two key questions:
 
     1) How does one get set to use COM1 as the serial consol during a boot
        from an install CD?  (Must I get to a boot prompt and then type in
        the command 'set console="comconsole"' and the 'boot'?)
 
     2) Assuming that the only thing approximating a serial "dumb terminal"
        that I have on hand is another UNIX box which also is running FreeBSD
        and which also has a serial port, and assuming that I have obtained
        and installed an appropriate nukll modem serial cable between the
        two serial ports on the two machines, then what command or commands
        should or must I execute on the second system in order to talk to
        the serial port (like a dumb terminal) _and_ capture its output to
        a disk file?
 
 > uploading good quality digital photos should be enough.
 
 This was not my first prefrence, but it is obviously faster than the
 alternative.  But I am not fast enough to capture everything as it all
 goes scollling by.  Stiil I hope these photos may be helpful, particularly
 the final one:
 
  	http://www.tristatelogic.com/freebsd/p7110151.jpg
  	http://www.tristatelogic.com/freebsd/p7110152.jpg
  	http://www.tristatelogic.com/freebsd/p7110154.jpg
 
 
 >Secondly, are you able to confirm if 6.2 and 6.3 CDs boot on
 >this machine?
 
 I just did, yes.  Apparently, I _can_ get past booting and do an entire
 install (onto a blank harddrive) from a set of 6.3-RELEASE CDs.  No problem.
 I just can't seem to make that happen on this system for 7.0-RELEASE.
 
 Perhaps in cases where the md memory disk cannot be successfully set up,
 _somebody_ should be writting _some_ sort of message to the console so
 that we all can have a prayer of figuring out what exactly went wrong.
 
 >the results may well go a long way to
 >figuring out exactly what was changed to cause this issue.
 
 I hope so.
 
 
 Regards,
 rfg
 
 
 P.S.  May I ask why you elected to close this PR?  I am having trouble under-
 standing how that is appropriate, given that the problem has not actually been
 solved.
 
 P.P.S.  I will be away from my desk until at least next Thrusday, but if
 you have more questions, I will be happy to respond when I return.

From: "Ronald F. Guilmette" <rfg@tristatelogic.com>
To: Gavin Atkinson <gavin@FreeBSD.org>
Cc: bug.followup@FreeBSD.org, FreeBSD-gnats-submit@FreeBSD.org,
    freebsd-i386@FreeBSD.org
Subject: Re: i386/125516: 7.0-RELEASE install dies (probably same problem as PR 113160)
Date: Tue, 22 Jul 2008 21:46:17 -0700

 Hello,
 
 Can you please tell me what is happening with this bugreport?  I sent
 you another reply (see directly below) a few days ago, but I have not
 heard back from you, nor from anyone else about this bug.  Also, most
 worrying is that the cc: address you put into your last message to me
 (bug.followup@FreeBSD.org) seems to be undeliverable (!)  Please see
 the full bounce message that I am also including below also.
 
 Please let me know if any progress can be made on this bug and/or if
 there is anything else that I ought to be providing in the way of useful
 additional information.  (As I've noted already, if I can't get a fix for
 this, then I'm basically going to have to throw away the ASUS A7N266-MV/AA
 motherboard that started all this.)
 
 Thanks in advance.
 
 
 Regards,
 rfg
 
 =============================================================================
 To: Gavin Atkinson <gavin@FreeBSD.org>
 cc: bug.followup@FreeBSD.org
 Subject: Re: i386/125516: 7.0-RELEASE install dies (probably same problem as PR 113160) 
 In-reply-to: Your message of Sat, 12 Jul 2008 19:05:28 +0100.
              <20080712184259.L2347@ury.york.ac.uk> 
 Date: Sat, 19 Jul 2008 18:48:10 -0700
 Message-ID: <931.1216518490@tristatelogic.com>
 From: "Ronald F. Guilmette" <rfg@tristatelogic.com>
 
 
 In message <20080712184259.L2347@ury.york.ac.uk>, you wrote:
 
 >> 	http://www.tristatelogic.com/freebsd/p7110151.jpg
 >> 	http://www.tristatelogic.com/freebsd/p7110152.jpg
 >> 	http://www.tristatelogic.com/freebsd/p7110154.jpg
 >
 >For info, you should be able to press scroll lock and then scroll back up 
 >through the messages with pageup/pagedown.
 
 OK.  Do you need that extra info?  Or are the pictures above sufficient?
 
 >Can you try booting, but escape to the loader prompt and enter the 
 >following:
 > 	set hw.ata.atapi_dma=0
 > 	boot
 >
 >and see if you get the same symptoms?
 
 OK, I tried that.  Just to be on the safe side, I tried it both with and
 without quoting the zero ("0") because I wasn't sure if the quotes
 were needed.  Anyway, either way, I am still getting the exact same
 kind of failure at the exact same place.
 
 So?  What now?  I really want to get this working.
 
 >> P.S.  May I ask why you elected to close this PR?...
 >
 >I haven't closed this PR, it's currently in "feedback" state, i.e. was 
 >waiting for feedback from you.
 
 OK.  You got my feedback now.  (Sorry it took so long but I was out of town.)
 =============================================================================
 
 
 Now the bounce message for <bug.followup@FreeBSD.org>:
 
 
 ------- Forwarded Message
 
 Return-Path: MAILER-DAEMON
 Delivery-Date: Sat Jul 19 19:04:25 2008
 Return-Path: <>
 X-Original-To: rfg@tristatelogic.com
 Delivered-To: rfg@tristatelogic.com
 Received: from mx2.freebsd.org (mx2.freebsd.org [69.147.83.53])
 	by segfault.tristatelogic.com (Postfix) with ESMTP id 284F71142D
 	for <rfg@tristatelogic.com>; Sat, 19 Jul 2008 19:04:25 -0700 (PDT)
 Received: from hub.freebsd.org (hub.freebsd.org [IPv6:2001:4f8:fff6::36])
 	by mx2.freebsd.org (Postfix) with ESMTP id D9F30150EB1
 	for <rfg@tristatelogic.com>; Sun, 20 Jul 2008 02:04:24 +0000 (UTC)
 	(envelope-from "")
 Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34])
 	by hub.freebsd.org (Postfix) with ESMTP id DC159106567C
 	for <rfg@tristatelogic.com>; Sun, 20 Jul 2008 02:04:24 +0000 (UTC)
 	(envelope-from "")
 Received: by mx1.freebsd.org (Postfix)
 	id DB0768FC16; Sun, 20 Jul 2008 02:04:24 +0000 (UTC)
 Date: Sun, 20 Jul 2008 02:04:24 +0000 (UTC)
 From: MAILER-DAEMON@FreeBSD.ORG (Mail Delivery System)
 Subject: Undelivered Mail Returned to Sender
 To: rfg@tristatelogic.com
 Auto-Submitted: auto-replied
 MIME-Version: 1.0
 Content-Type: multipart/report; report-type=delivery-status;
 	boundary="BC79A8FC12.1216519464/mx1.freebsd.org"
 Message-Id: <20080720020424.DB0768FC16@mx1.freebsd.org>
 
 This is a MIME-encapsulated message.
 
 - --BC79A8FC12.1216519464/mx1.freebsd.org
 Content-Description: Notification
 Content-Type: text/plain; charset=us-ascii
 
 This is the mail system at host mx1.freebsd.org.
 
 I'm sorry to have to inform you that your message could not
 be delivered to one or more recipients. It's attached below.
 
 For further assistance, please send mail to postmaster.
 
 If you do so, please include this problem report. You can
 delete your own text from the attached returned message.
 
                    The mail system
 
 <bug.followup@FreeBSD.org>: host hub.freebsd.org[2001:4f8:fff6::36] said: 550
     5.1.1 <bug.followup@FreeBSD.org>: Recipient address rejected: User unknown
     in local recipient table (in reply to RCPT TO command)
 
 - --BC79A8FC12.1216519464/mx1.freebsd.org
 Content-Description: Delivery report
 Content-Type: message/delivery-status
 
 Reporting-MTA: dns; mx1.freebsd.org
 X-Postfix-Queue-ID: BC79A8FC12
 X-Postfix-Sender: rfc822; rfg@tristatelogic.com
 Arrival-Date: Sun, 20 Jul 2008 02:04:24 +0000 (UTC)
 
 Final-Recipient: rfc822; bug.followup@FreeBSD.org
 Original-Recipient: rfc822;bug.followup@FreeBSD.org
 Action: failed
 Status: 5.1.1
 Remote-MTA: dns; hub.freebsd.org
 Diagnostic-Code: smtp; 550 5.1.1 <bug.followup@FreeBSD.org>: Recipient address
     rejected: User unknown in local recipient table
 
 - --BC79A8FC12.1216519464/mx1.freebsd.org
 Content-Description: Undelivered Message
 Content-Type: message/rfc822
 
 Received: from segfault-outgoing-helo.tristatelogic.com (112.171-60-66-fuji-dsl.static.surewest.net [66.60.171.112])
 	by mx1.freebsd.org (Postfix) with ESMTP id BC79A8FC12
 	for <bug.followup@FreeBSD.org>; Sun, 20 Jul 2008 02:04:24 +0000 (UTC)
 	(envelope-from rfg@tristatelogic.com)
 Received: from segfault-nmh-helo.tristatelogic.com (localhost [127.0.0.1])
 	by segfault.tristatelogic.com (Postfix) with ESMTP id 32F9311423;
 	Sat, 19 Jul 2008 18:48:10 -0700 (PDT)
 To: Gavin Atkinson <gavin@FreeBSD.org>
 cc: bug.followup@FreeBSD.org
 Subject: Re: i386/125516: 7.0-RELEASE install dies (probably same problem as PR 113160) 
 In-reply-to: Your message of Sat, 12 Jul 2008 19:05:28 +0100.
              <20080712184259.L2347@ury.york.ac.uk> 
 Date: Sat, 19 Jul 2008 18:48:10 -0700
 Message-ID: <931.1216518490@tristatelogic.com>
 From: "Ronald F. Guilmette" <rfg@tristatelogic.com>
 
 
 In message <20080712184259.L2347@ury.york.ac.uk>, you wrote:
 
 >> 	http://www.tristatelogic.com/freebsd/p7110151.jpg
 >> 	http://www.tristatelogic.com/freebsd/p7110152.jpg
 >> 	http://www.tristatelogic.com/freebsd/p7110154.jpg
 >
 >For info, you should be able to press scroll lock and then scroll back up 
 >through the messages with pageup/pagedown.
 
 OK.  Do you need that extra info?  Or are the pictures above sufficient?
 
 >Can you try booting, but escape to the loader prompt and enter the 
 >following:
 > 	set hw.ata.atapi_dma=0
 > 	boot
 >
 >and see if you get the same symptoms?
 
 OK, I tried that.  Just to be on the safe side, I tried it both with and
 without quoting the zero ("0") because I wasn't sure if the quotes
 were needed.  Anyway, either way, I am still getting the exact same
 kind of failure at the exact same place.
 
 So?  What now?  I really want to get this working.
 
 >> P.S.  May I ask why you elected to close this PR?...
 >
 >I haven't closed this PR, it's currently in "feedback" state, i.e. was 
 >waiting for feedback from you.
 
 OK.  You got my feedback now.  (Sorry it took so long but I was out of town.)
 
 
 
 - --BC79A8FC12.1216519464/mx1.freebsd.org--
 
 ------- End of Forwarded Message
 

From: "Ronald F. Guilmette" <rfg@tristatelogic.com>
To: Gavin Atkinson <gavin@FreeBSD.org>
Cc: bug.followup@FreeBSD.org, FreeBSD-gnats-submit@FreeBSD.org,
    freebsd-i386@FreeBSD.org
Subject: Re: i386/125516: 7.0-RELEASE install dies (probably same problem as PR 113160)
Date: Tue, 22 Jul 2008 23:54:11 -0700

 Whew!  I've just spent the last two hours, screweing around with this
 ASUS A7N266-VM/AA based system, trying to see if anything at all would
 make a difference, and now I have a lot more data points to add about
 this bug report.  (Not that any of this additional info will necessarily
 help.)
 
 First, the good news... It now appears to me that this bug is most probably
 _not_ the same at problem as was reported in PR 113160, i.e. failure to set
 up the "md" (memory disk) duing boot up from an install CD.
 
 Now the bad news... The problem is even worse than I first reported.
 
 Originally I only reported that an install from 7.0-RELEASE CDs was
 dying.  It now appears however that if one has a hard drive with
 7.0-RELEASE already fully installed on it, and if one then transplants
 that hard drive onto a ASUS A7N266-VM/AA based system, then you won't
 even be able to fully boot 7.0 from that drive.  (If you try this,
 the machine will mysteriously reboot itself, just at about the time
 when the spinning thingy should first be visible.  (I dunno.  Is that
 in the second stage boot loader?  Or is that the third?)
 
 Basically, even though I could not do an install on this specific system
 from 7.0-RELEASE CDs, I still did want to try to run 7.0-RELEASE on it,
 so I went to another Athlon XP 2000 system I have here, put a blank
 drive on that, repartitioned it, added BSD labels on the partition
 I wanted to use for 7.0-RELEASE and then did a totally fresh install.
 That all worked great.  No problem.  And I could boot from the drive
 right after that, no problem.
 
 Then I took that exact same drive over and installed it on my A7N266-VM/AA
 based system and tried to boot, whereupon I got the effect described above...
 it starts to boot (and makes it at least past the first stage) whereupon
 I hit the F2 key (to boot from the specific partition where I put 7.0-RELEASE)
 and then things kinda get started, but then a few seconds later the machine
 just reboots itself for no apparently good reason.  (This was repeatedly
 seen by me, many many times.  It is _not_ in any sense a transient problem.)
 
 I am repeatedly emphasizing that this is all happening on a ASUS A7N266-VM/AA
 based system, because I tried everything to pin down the problem, and the
 only logical conclusion I could come to in the end was that 7.0 just
 really really REALLY does not like this specific motherboard.  I pulled
 out all of the peripherial card (only two were in there, both just ethernet
 cards) and the problem remained.  I disconnected all drives except the
 one hard drive containing the copy of 7.0-RELEASE that I was trying to
 boot from and the problem remained.  I took the #1 and #2 memory modules
 out and put back (just) what had been #2 into the slot that #1 has been
 in and the problem remained.  I even took out the 300W power supply and
 put in a 400W supply and the problem still remained, exactly as before.
 
 The only elements that remaned constant across all of the tests I performed
 were the hard drive, the CPU and the motherboard.  I _know_ that the CPU
 is OK, because it has been running (and will again now be running)
 6.1-RELEASE for the past year and a half with no problems.  Also, I
 have several other Athlon XP 2000 systems here and I know that at least
 one other was able to boot and run 7.0-RELEASE without problems.  The
 exact same hard drive (Seagate 200GB PATA) booted 7.0-RELEASE just fine
 on a different one of my Athlon XP 2000 systems here, so that is not
 the problem.  The only think left which is suspect is the motherboard.
 
 Conclusion:  7.0-RELEASE just doesn't get along with the ASUS A7N266-VM/AA
 motherboard.  Like not at all.
 
 I'm still kinda hoping that somebody will tell me why.
 
 
 Regards,
 rfg
 
 
 P.S.  If it is the only way to get this solved, then I'm willing to ship
 the motherboard+cpu+heatsink/fan to anybody who might be willing to look
 into the problem further.
State-Changed-From-To: feedback->open 
State-Changed-By: gavin 
State-Changed-When: Wed Dec 3 12:43:19 UTC 2008 
State-Changed-Why:  
Feedback was received some time ago 

http://www.freebsd.org/cgi/query-pr.cgi?pr=125516 
>Unformatted:
