Re: ts209ii to ts219p hw upgrade w/ present hd config
- To: Martin Michlmayr <tbm@cyrius.com>
- Cc: debian-arm@lists.debian.org
- Subject: Re: ts209ii to ts219p hw upgrade w/ present hd config
- From: John Holland <john.holland@cellent-fs.de>
- Date: Tue, 04 Aug 2009 21:27:04 +0200
- Message-id: <1249414024.14019.32.camel@oklahoma>
- In-reply-to: <20090729163333.GF2450@deprecation.cyrius.com>
- References: <1248862732.8005.15.camel@oklahoma> <20090729163333.GF2450@deprecation.cyrius.com>
On Wed, 2009-07-29 at 18:33 +0200, Martin Michlmayr wrote:
> * John Holland <john.holland@cellent-fs.de> [2009-07-29 12:18]:
First I upgraded qcontrol to get support for the TS[12]19:
a) added the following line to /etc/apt/sources.list
deb http://people.debian.org/~tbm/orion lenny main
b) updated and upgraged
image was also upgraded to 2.6.30-1
Second I created a new mtd2 as described later by Martin
> > 1) backup vital data in the current system.
>
> Don't forget to make a backup of the flash.
>
Then created my backups of the current fs and the mtdblocks of the new
hardware TS219P.
> > 2) flash the mtd's as per Martins' site says here w/o any hds:
> > http://www.cyrius.com/debian/kirkwood/qnap/ts-219/unpack.html
> >
> > 3) reboot the system
> >
> > 4) turn off the system and put in the two disks from the current hw
> >
> > 5) boot the system and upgrade debian
> >
> > I think I'm missing a step in the list -> tell debian where to boot: in
> > my case a md, more specifically a raid 1 configuration.
>
> One step you're missing is to install the kernel for the TS-219P. The
> TS-209 is based on the Orion plaform whereas the TS-219P uses the
> Kirkwood platform, so you'll need the Kirkwood kernel.
>
> So, install the following kernel package on the TS-209:
>
> wget http://people.debian.org/~tbm/orion/old/linux-image-2.6.29-2-kirkwood_2.6.29-5_armel.deb
> sudo dpkg -i linux-image-2.6.29-2-kirkwood_2.6.29-5_armel.deb
>
> It will first install the kernel, then generate the ramdisk and then
> print an error saying that this kernel doesn't match your
> sub-architecture (that's okay - just ignore the error).
>
> Then you can download mtd1.bz2/mtd2.bz2, unpack them and put them in
> flash on the TS-219P (after making backups of flash).
>
> Of course, the mtd2 file assumed that your root device is sda2.
> You could repack the initramfs (mtd2) and change the root device:
>
> bzip2 -d mtd2.bz2
> mkdir i
> cd i
> zcat ../mtd2 | sudo cpio -i
> sudo perl -pi -e "s/sda2/md0/" conf/param.conf
> find . | sudo cpio --quiet -o -H newc > ../new-mtd2
> cd ..
> gzip -9 new-mtd2
> dd if=new-mtd2.gz of=new-mtd2 ibs=9437184 conv=sync
>
> When you write this to flash and then put the disk into the TS-219P,
> it should (hopefully) boot.
>
The system seemed to boot normally, but I couldn't reach it on the
network.
I put the disks in the original hardware and took a look at dmesg.1.gz
and quickly saw what is going wrong:
[42949385.390000] udev: renamed network interface eth0 to eth1
Searching /etc for instances of eth0 I found
/etc/udev/rules.d/70-persistent-net.rules containing two rules:
# Unknown net device (/class/net/eth0) (mv643xx_eth_port)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*",
ATTR{address}=="yy:yy:yy:yy:yy:yy", ATTR{type}=="1", KERNEL=="eth*",
NAME="eth0"
# Unknown net device (/devices/platform/mv643xx_eth_port.0/net/eth0)
(mv643xx_eth_port)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*",
ATTR{address}=="xx:xx:xx:xx:xx:xx", ATTR{type}=="1", KERNEL=="eth*",
NAME="eth1"
Where 'yy:yy:...' and 'xx:xx:...' are the mac addresses of the two
different machines.
Can I simply comment the above two lines to get things to work as
normally, or can I change the reference to eth1 to eth0? Perhaps,
someone has a better idea?
I do want to be able to boot from the original hardware using the same
disks as well as from the new...
Thanx in advance,
John Holland
___________________________________________
Cellent Finance Solutions AG
Firmensitz: Calwer Straße 33, 70173 Stuttgart
Registergericht: Amtsgericht Stuttgart, HRB 720743
Vorstand: Thomas Wild
Vorsitzender des Aufsichtsrats: Rudolf Zipf
Reply to: