Port-arm archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Proposal for making raspberry pi and similar systems easier to install



Am 06.03.2013 02:27, schrieb Christos Zoulas:
In article 
<CAJcb3fqqOu598GOWbep06xE4zXd6uqdubdzojjesYHL+ZnMSeQ%mail.gmail.com@localhost>,
Andy Ruhl  <acruhl%gmail.com@localhost> wrote:

Filesystem         Size       Used      Avail %Cap Mounted on
/dev/ld0a          558M       547M       -17M 103% /
tmpfs              624K       188K       436K  30% /dev
/dev/ld0e           52M       7.5M        44M  14% /boot
kernfs             1.0K       1.0K         0B 100% /kern
ptyfs              1.0K       1.0K         0B 100% /dev/pts
procfs             4.0K       4.0K         0B 100% /proc

It's true that something probably should be done.

In the specific case of the Raspberry Pi, it would be nice to just
boot a regular sysinst from an SD card, then do the install to a USB
disk of some sort (since you can get ones with large storage). At that
point I wouldn't mind leaving a small SD card in the machine whose
sole purpose is to boot  and mount root from the USB disk. Or I
suppose the hardware could possibly be hacked to boot directly from
USB.

But for other devices which don't have a nice console, I completely
agree. What you said is easier than trying to get a console and
netboot a ramdisk kernel or something. That is a large barrier to
entry.

It is trivial to fix. First remove "log" from the root partition and reboot.
(there is a bug that makes mount -u -o nolog / crash). After you reboot...

# shutdown // to single user

# disklabel -i ld0
A<enter>  // autosize
d<enter>
unused: <enter>
start: <enter>
end: $<enter>     // resize to physical
a<enter>
4.2BSD: <enter>
start: <enter>
end: $<enter>     // or to whatever you want to resize.
W<enter>
q<enter>

# resize_ffs /dev/rld0a // to the size of the partition you set.

power off, no syncing.
Don't add back -o log, it makes things crash on the nightly run. I've
been running with.

I tried that and got "BAD SUPER BLOCK: VALUES IN SUPER BLOCK DISAGREE
WITH THOSE IN FIRST ALTERNATE"-Errors. After resize_ffs i also tried
to check for errors but nothing could be found. Anyone else ran into
same trouble?

robert



Home | Main Index | Thread Index | Old Index