NetBSD-Users archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
NetBSD on Dedibox SC Gen2
Hello,
for the record here is a short howto install NetBSD on Dedibox SC Gen2.
Dedibox is a dedicated server offer by the french compagny online.net
(I've no interests in this compagny - I just happen to rent a
server here since yesterday :).
The Dedibox SC Gen2 is a Dell XS11-VX8 server with a VIA Nano U2250 CPU,
2GB Ram and a 500GB SSHD drive. The XC 2015 is probably the same hardware
with a bigger disk or a plain SSD.
The provider's remote management allows (re)installation of various
linux distribtions, and FreeBSD 10.x. It also allows a rescue mode
(I guess a PXE boot with ramdisk) for theses OSes.
The probleme is that there's no console access, only SSH access to
the installed or rescue OS.
To install NetBSD I did the following (based on instructions from
https://devnullblg.wordpress.com/2014/04/18/openbsd-installation-on-a-dedibox-sc-gen2/):
boot the Ubuntu distrib in rescue mode, ssh to it, get a root shell
(sudo -s), and install qemu:
apt-get update && apt-get -y install qemu
download a NetBSD install iso (you need the complete install iso; as
you won't have network access to download the sests):
wget http://ftp.fr.netbsd.org/pub/NetBSD-daily/netbsd-7/201509170410Z/images/NetBSD-7.0_RC3-amd64.iso
(of course you'll have to adjust the URL).
then boot NetBSD in qemu from the iso image, with the HD attached to the
real hard disk:
qemu-system-x86_64 -no-kvm -hda /dev/sda -cdrom NetBSD-7.0_RC3-amd64.iso -curses -boot d
from here you can do a NetBSD install on the hard disk.
But there is *ONE BIG ISSUE*: qemu is buggy with disks larger than 128GB,
and this causes NetBSD to write at the wrong places above 128GB,
causing the resulting install to be unbootable.
So: create a NetBSD MBR partition (type 169) which covers the whole
disk, but then allocate only partitions below the 128GB limits in the
disklabel. You'll be able to create of resize partitions later.
I did just create a /, /var and swap partitions of a few GB, and
created the big /home later.
Don't forget to create a non-root user which is part of the wheel group,
and enable sshd.
Before exiting sysinst you have to configure network in your new
installation. For this:
drop to shell
mount /dev/wd0a /mnt
chroot /mnt
create /etc/ifconfig.wm0 with your ip address and netmask
create /etc/mygate with your gateway
(you got these informations when you ordered your dedibox, or you
can find them in the linux rescue boot with ifconfig and route commands).
edit /boot.cfg to boot with acpi disabled by default (boot -2).
run 'halt -p' to gracefully shutdown qemu.
Then, from the web management you can reboot to normal mode.
After a few minutes, your server should respond to ping, and you
should be able to ssh in.
If things go wrong:
the FreeBSD rescue mode allows you to mount your NetBSD partitions:
mount /dev/ada0s1p1 /mnt
(and eventually mount the /var partition on /mnt/var is you created one).
Then you can look at /mnt/var/run/rc.log, /mnt/var/run/dmesg.boot and
/mnt/var/log/messages to find for some hints.
--
Manuel Bouyer <bouyer%antioche.eu.org@localhost>
NetBSD: 26 ans d'experience feront toujours la difference
--
Home |
Main Index |
Thread Index |
Old Index