Date: Tue, 12 Dec 2000 02:25:17 -0800 (PST) From: jdb <jdb@detox.methadonia.net> To: questions@freebsd.org Subject: boot troubles. Message-ID: <200012121025.CAA24005@detox.methadonia.net>
next in thread | raw e-mail | index | archive | help
alas, i am not on -questions, so i suggest replying directly.
i've recently installed 4.2-RELEASE, which seems to display an odd quirk:
boot0 doesn't appear to work. i am rather unsure how to debug this problem,
so i'll give everyone as many details as I possibly can on the hopes that
perhaps this has been seen before. A quick search through the handbook
and FAQ turned up nothing (except a lot of useless links to 22.6 -- my
search terms were "installing boot blocks").
I have a Micronics Dual Fortress SM02 motherboard. The onboard IDE has
been disabled in the BIOS and no IDE drives are present. Instead, I am
booting from a Seagate ST15230N on SCSI ID 0 through an Adaptec 2940UW.
There is an ST19171W at SCSI ID 1. The drives work fine for installation,
and they boot and run fine if I boot from the installation floppies (using
'set currdev=disk1s1a:' followed by 'boot').
The actual symptom is that it's unclear whether boot0 ever loads. Should
boot0 display anything before it loads boot1? Literally the thing stops
in its tracks and never displays the expected boot prompt (F1=FreeBSD).
I have attempted to reinstall boot0 to no avail, using the command,
boot0cfg -B -v da0
Everything looks good as far as boot0cfg is concerned (meaning, it
displays the type 0xa5 slice at spot 1, and says it wants to boot that;
it is configured to use BIOS drive 0x80).
fdisk(8) displays some more interesting reports. by simply running 'fdisk',
it complains that "Figures below won't work with BIOS for partitions not
in cyl 1" and "Warning: BIOS sector numbering starts with sector 1." The
first one I'm thinking of simply ignoring, because it then suggests I use
values which match those extracted from the in-core disklabel. The second
is more interesting, since I remember something about partitions (er,
"slices") starting on the 1st cylinder, rather than the 0th. But I'm
honestly not sure what to make of it.
The exact output from fdisk(8) is:
# fdisk
******* Warking on device /dev/da0 *******
parameters extracted from in-core disklabel are:
cylinders=1616 heads=247 sectors/track=21 (5187 blks/cyl)
Figures below won't work with BIOS for partitions not in cyl 1
parameters to be used for BIOS calculations are:
cylinders=1616 heads=247 sectors/track=21 (5187 blks/cyl)
Media sector size is 512
Warning: BIOS sector numbering starts with sector 1
Information from DOS bootblock is:
The data for parition 1 is:
sysid 165,(FreeBSD/NetBSD/386BSD)
start 21, size 8382171 (4092 Meg), flag 80 (active)
beg: cyl 0/ sector 1/ head 1;
end: cyl 1023/ sector 21/ head 246
The data for partition 2 is:
<UNUSED>
The data for partition 3 is:
<UNUSED>
The data for partition 4 is:
<UNUSED>
I did *not* select dangerously dedicated disks during install-time.
As a note, it is rather unclear how to install bootblocks with FreeBSD.
While it would be pretty bad to have a user shoot itself in the foot with
such a powerful trigger, I suggest it might be documented anyway, if indeed
this is the problem.
Another note, I did manage to crash out of the FreeBSD install during the
post-install options menu, after I configured X and all that stuff (still
in the late sysinstall process, not after having rebooted). This may have
something to do with the problem.
Anyone have any ideas or suggestions?
Thanks.
- joel
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200012121025.CAA24005>