Date: Thu, 16 Aug 2012 02:27:44 +0400 From: Lev Serebryakov <lev@FreeBSD.org> To: current@freebsd.org Subject: r239290: boot problem Message-ID: <1857132939.20120816022744@serebryakov.spb.ru>
next in thread | raw e-mail | index | archive | help
Hello, Current.
I have typical NanoBSD installation: 8GiB CF card patitioned with
MBR into 4 slices. Slices 1 and 2 is used for code (active only
one at any time) and 3 is used for configs and 4 for some mutable
data. Slices 1 and 2 have only one BSD partition: `a'.
boot0 in MBR is "standard" boot0.
Any upgrade is dd'ing of new code slice in place and switching active
slice (partition) with "gpart set". It works Ok for many years.
Now, I've tried to upgrade system to CURRENT r239290 (for latest
changes and experiments with my network traffic and schedulers). This
new code slice was dumped into slice 2 (NB!) and slice 2 becomes
active.
"reboot" command in loader doesn't work at all. It only prints
"Rebooting..." and waits forever.
After this upgrade system stops to boot. Loader doesn't see any
slices but first one (with old system) and stops boot process because
could not find kernel. When it is instructed to load kernel from first
slice, it could do this, but could not continue boot -- it reboots on
"boot" command after ~10 seconds of inactivity!
Here is output of gpart command on "old" system (r239228):
=======================================================================
root@gateway:/root # gpart show
=> 63 16006977 ad0 MBR (7.6G)
63 8000433 1 freebsd (3.8G)
8000496 63 - free - (31k)
8000559 8000433 2 freebsd [active] (3.8G)
16000992 3024 3 freebsd (1.5M)
16004016 2016 4 freebsd (1M)
16006032 1008 - free - (504k)
=> 0 8000433 ad0s1 BSD (3.8G)
0 16 - free - (8.0k)
16 8000417 1 !0 (3.8G)
=> 0 8000433 ad0s2 BSD (3.8G)
0 16 - free - (8.0k)
16 8000417 1 !0 (3.8G)
=======================================================================
Here is console log of process, including my typing:
=======================================================================
1 Seconds to automatic boot. Press Ctrl-P for entering Monitor.
F1 FreeBSD
F2 FreeBSD
F6 PXE
Boot: F2
/boot/config: -h -S115200
Consoles: serial port
BIOS drive C: is disk0
BIOS 639kB/523264kB available memory
FreeBSD/x86 bootstrap loader, Revision 1.1
(root@vmware-c-32.home.serebryakov.spb.ru, Wed Aug 15 12:55:36 MSK 2012)
can't load 'kernel'
Type '?' for a list of commands, 'help' for more detailed help.
OK lsdev -v
cd devices:
disk devices:
disk0: BIOS drive C:
disk0s1: FreeBSD 3906MB
disk0s1a: Unknown 3906MB
pxe devices:
OK set rootdev=disk0s1a:
OK load /boot/kernel/kernel
not found '/boot/kernel/kernel'
OK load disk0s1a:/boot/kernel/kernel
disk0s1a:/boot/kernel/kernel text=0x43b620 data=0x3a7ec+0x30690 syms=[0x4+0x5ab70+0x4+0x7de03]
OK show
LINES=24
boot_serial=YES
comconsole_pcidev=
comconsole_port=1016
comconsole_speed=115200
console=comconsole
currdev=disk0s2a:
interpret=OK
kernelname=disk0s1a:/boot/kernel/kernel
loaddev=disk0s2a:
prompt=${interpret}
rootdev=disk0s1a:
OK boot
\
> BOOM! REBOOTED!
=======================================================================
--
// Black Lion AKA Lev Serebryakov <lev@FreeBSD.org>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1857132939.20120816022744>