Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 03 Mar 2000 11:01:06 -0500 (EST)
From:      John Baldwin <jhb@FreeBSD.org>
To:        Brad Knowles <blk@skynet.be>
Cc:        freebsd-stable@FreeBSD.org
Subject:   RE: Crash on boot when disks are present?
Message-ID:  <200003031601.LAA00700@server.baldwin.cx>
In-Reply-To: <v0422080bb4e401fc570d@[195.238.1.121]>

next in thread | previous in thread | raw e-mail | index | archive | help

On 02-Mar-00 Brad Knowles wrote:
> Folks,
> 
>       This one is so weird that I don't have the foggiest clue how I'd 
> go about searching the archives for answers.  Searching for "crash on 
> boot" netted me way more hits than I care to try to count.
> 
> 
>       I've got a FreeBSD 3.4-STABLE system (cvsup'ed a couple of weeks 
> ago) that I've been doing some tests with and it has been working 
> pretty well up until now (after I got the issues of which PCI slots I 
> can plug SCSI controllers into ;-).
> 
>       Anyway, the vendor field engineer for our disk drive array (one 
> of those expensive mainframe-style refrigerator-size models) came in 
> this morning to try to fix some problems we're having with it not 
> recognizing all the RAM that is installed, and among other things he 
> did a firmware upgrade of the controllers, and while we were at it we 
> also deleted all the existing RAID groups and created some new ones, 
> with new logical units associated, etc....
> 
>       I've gone in and made sure that the BIOS has been changed (yet 
> once again) to try to boot off the internal Quantum Atlas-IV drive 
> that is attached to the AIC-7890 controller, because the external 
> drive array does not have the OS installed, etc....  Unfortunately, 
> when I try to boot with the drive array turned on, it crashes hard 
> (during the bootloader phase, I think), displaying the following:
> 
> F1    DOS
> F2    FreeBSD
> F5    Drive 1
> 
> Default: F2
> 
> -
> int=00000000  err=00000000  efl=00030083  eip=00000004
> eax=000004e3  ebx=0000097f  ecx=0000dc00  edx=000034be
> esi=0000dc00  edi=00000004  ebp=00000000  esp=000004ce
> cs=c800  ds=c800  es=9e7d    fs=0000  gs=0000  ss=9e7b
> cs:eip=d4 00 00 00 00 00 00 00-00 00 00 00 00 00 00 00
> ss:esp=00 00 00 00 00 00 00 00-00 00 00 00 00 00 00 00

That's very odd indeed.  Well, let see here.  0xd4 0x00 is
AAM with a base of 0, which results in a divide by zero.
That explains the fault at least. However, it looks like you
have trashed memory.  Recently, I fixed a bug in the loader
for emulating one of the BIOS calls used by some RAID adapters.
Could you try using the boot floppies from the latest 3.x-snap
on releng3.freebsd.org instead?  Thanks.

>       I hope I got this right -- I don't have a serial console attached 
> to the machine, so I had to type this in manually.
> 
> 
>       Anyway, when I turn off the external drive array, it boots fine. 
> I know I need to expand /var to accomodate /var/crash, but I don't 
> think that this would actually generate a crash dump, for which this 
> would be useful.

No, this is the loader itself faulting, BSD itself isn't actually up
and running yet, so we can't write to /var/crash at this point.

>       Other than having the guy come back out to "upgrade" the 
> controller firmware version back to what it was, does anyone have any 
> other advice that they can give me as to what I might be able to do 
> to debug the problem?
> 
> 
>       Thanks!

-- 

John Baldwin <jhb@FreeBSD.org> -- http://www.FreeBSD.org/~jhb/
PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc
"Power Users Use the Power to Serve!"  -  http://www.FreeBSD.org/


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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200003031601.LAA00700>