Re: -current AMD64 unable to mount root after buildworld/kernel reboot

On Thu, July 12, 2007 00:45, Garrett Cooper wrote:
> Gelsema, P (Patrick) - FreeBSD wrote:
>> Hi,
>> I have got the following issue.
>> I have installed Current June AMD64 snapshot without any issues. I can
reboot and the machine comes back up without any issues.
>> /etc/make.conf is copied from /usr/share/examples/etc/make.conf and the
only changed made are:
>> cputype?=athlon64
>> cflags= -O2 -fno-strict-aliasing -pipe (only removed the # in front of the
>> line)
>> coptflags=-O -pipe (only removed the # in front of the line)
>> and the settings regarding cvsup are enabled. yes and
>> cvsup3.nl.freebsd.org, and only the standard-cvsupfile is uncommented.
using the /usr/share/examples/cvsup/standard-supfile with no changes.
First thing I do is a make update (as of this evening) after which I
execute the following commands;
>> #chflags -R noschg /usr/obj
>> #rm -rf /usr/obj
>> #make cleandir
>> #make cleandir
>> #make -j8 buildworld
>> #make -j8 buildkernel
>> #make -j8 installkernel
>> #make -j8 installworld
>> #mergemaster
>> #shutdown -r now
>> After this the machine reboots, starts FreeBSD and detects everything
as
>> it ought to. However the moment it tries to mount root from
>> ufs:/dev/da0s1a it seems to fail and asks for;
>> manual root filesystem specification. When I type in manually
>> ufs:/dev/da0s1a it continues with the following error;
>> ELF interpreter /libexec/ld-elf.so.1 not found
>> init: /bin/sh on /etc/rc terminated abnormally, going to single user mode.
>> Enter full pathname of shell or RETURN for /bin/sh.
>> Hitting enter I get;
>> ELF interpreter /libexec/ld-elf.so.1 not found
>> init: single user shell terminated, restarting.
>> Enter full pathname of shell or RETURN for /bin/sh.
>> When rebooting with kernel.old everything goes ok and I am able to log on.
>> Any help much appreciated.
>> Rgds,
>> Patrick
>
>     You're most likely missing some key components in specifying your
> disk controller.
>
>     Do you have SCSI, SATA, PATA? If you have SCSI/SATA, did you perhaps
> install your system with GEOM enabled?

SCSI, Adaptec 39320D controller with Seagate SCSI disk. I used GENERIC
kernel without any modifications. Controller is detected as aic7902 which
is part of GENERIC.

I swapped Harddisks and now I get the following error when building the
kernel ( I did get this on the other harddisk as well):

panic: ufs_dirbad: /usr: bad dir ino 2638106 at offset 0: mangled entry
cpuid = 1
KDB: enter: panic
[thread pid 9704 tid 100056 ]
Stopped at    kdb_enter+0x31:leave
db>

what could be causing this?

Patrick

>     Booting into the backup kernel (single user mode), will be extremely
> helpful in determining your issue, in particular what disk configuration
you're using.
>
> -Garrett