tech-toolchain archive

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

Re: creating FFS2 file-system with makefs on linux



The -x and -F options just help ensure that all the files and file permissions in the image are the right ones. Without that you could find that files in your image have strange owners and permissions.

I can legitimately avoid the use of an mtree specfile because all I do is extract official NetBSD sets to begin with.

Indeed I could find strange perms e.g. `/var/db/postfix/` with UID and GID 91 instead of 12. I am not sure where UID and GID 91 came from, as I don't have any `postfix` user on my host system so it wouldn't have been able to map it anyways.

Anyhow, extracting the sets with `--numeric-owner` did the trick. It works both with GNU tar and bsdtar, from GNU/Linux. I got a resulting guest system that looks pretty decent to me now.

Is resize_root supposed to give results at the first run already? While seeing the boot message about resizing /, should I therefore be able to check the size of `/` right away, or I am supposed to reboot afterwards?

resize_root (the shell script in /etc/rc.d) should reboot the system for you. I bet you don't have all of the following in your /etc/rc.conf.

     resize_root=YES
     resize_root_flags=-p
     resize_root_postcmd="/sbin/reboot -n"

Lloyd, fellow, weasel-or-dragon, DUDE! I wish I had the full howto from the start. This works.

Once rebooted one can remove those options from rc.conf and generate a template of the vdisk, and so the users won't have to go down that additional reboot. We're good.

Many thanks, really

Pierre-Philipp


Home | Main Index | Thread Index | Old Index