Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 22 Jun 1997 18:58:37 +1000
From:      Bruce Evans <bde@zeta.org.au>
To:        bde@zeta.org.au, tweten@ns.frihet.com
Cc:        freebsd-mobile@FreeBSD.ORG, freebsd-questions@FreeBSD.ORG, freebsd-stable@FreeBSD.ORG
Subject:   Re: NEC 6030X, FreeBSD 2.2.2, and the CDROM
Message-ID:  <199706220858.SAA27527@godzilla.zeta.org.au>

next in thread | raw e-mail | index | archive | help
>>The panic is before /etc/fstab is read.
>
>I disagree, because changing the contents of /etc/fstab (from wd1 based 
>contents provided by the CD install to wd0 based contents) prevents the panic.

The kernel shouldn't be paniced by anything in fstab.  It isn't even used
until after the root file system has been used to start init.

>When I ran the install CD, it made an /etc/fstab built around wd1.  I have 
>since gone into single user mode and changed it to be built around wd0.  Also, 
>the following tests were done with the CD drawer in place.
>
>Entering "1:wd(0,a)/kernel" results in the following immediately after the 
>last device probe:
>
>	Changing root device to wd1a

Are you sure?  The kernel actually prints "changing" entirely in lower case,
and entering wd(0,a) should have forced wd0a here.

>	swapon: /dev/wd0b: Device not configured
>	Automatic reboot in progress ...
>	Can't open /dev/rwd0a: Device not configured
>	/dev/rwd0a: CAN'T CHECK FILE SYSTEM.
>	/dev/rwd0a: UNEXPECTED INCONSISTENCY: RUN fsck MANUALLY.

THis shows that you don't really want wd0a.

>Entering "0:wd(1,a)/kernel" results, immediately after the last device probe, 
>in an infinite loop of:
>
>	Error: D:0x81 C:0 H:0 S:0

This message is printed by the boot loader.  No device probes are possible
if it occurs.  Entering 0: should have forced D:0x80 or no error here.

>Entering nothing results (immediately after the last device probe) in:
>
>	panic: cannot mount root
>
>	Syncing disks ... done
>	Automatic reboot in 15 seconds - press a key on the console to abort

This error is normal when the FreeBSD drive number passed to the kernel
does not match the actual boot drive number.  Perhaps you don't need to
do anything special when the CD draw is in place.  0:wd(1,a)kernel might
be necessary otherwise.

Bruce



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199706220858.SAA27527>