Re: gui versus kickstart capabilites

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

 



On 10/10/2014 11:30 AM, Chris Lumens wrote:
IIRC, I read somewhere that functionality should be implemented first in
kickstart and then in the gui.
We try to do things this way, yeah, especially for complicated new stuff
where figuring out the UI might take a while.

Thus, everything I can do in a kickstart
install I should be able to do in a gui install. Correct?
No.
Actually, I am OK with this and thinking about it I should have expected it.

OK, now let me reverse that.  I believe that everything (all functionality)
I can do in a gui install I should be able to do in a kickstart install.
Correct?
Yes.  Remember that the UI now internally is generating a kickstart
file.


Now this is the answer I was interested in and the answer I wanted.

When I do a gui install and I will be using BTRFS "partitioning", I can reuse (don't format) subvols such as home on /home or perhaps www on /srv/www. But, for the rootfs ("/") and /boot, they need to be "freshly formatted." I can reuse the subvol names and disk space by first deleting the subvol and then allocate it again with the same name. This will work if /boot is on a subvol or if it is a directory on the rootfs.

Now, what I would like is to have this same capability to "reuse" /boot and rootfs subvols when I do a kickstart install. The only way I can currently do this is to go out-of-band of the kickstart process (bootup in rescue mode and delete the subvols and then reboot to do the kickstart install).

If the gui is generating a kickstart file, how does it specify that subvol is to be deleted and then reallocated?

Why do I want this in kickstart. On my virtual systems I will reinstall over an existing system many times while keeping some data. And, more than once I have needed to reinstall on real hardware when doing a kickstart install.

I have previously submitted a couple of relatively simple patches for anaconda and pykickstart. Th update to pykickstart adds the "--reclaim" option to the btrfs command. The update to anaconda adds code to kickstart.py which checks to see if --reclaim was specified. If it was not, then processing will be as it is now. If it was AND the subvol currently exists, then it is deleted first followed by the allocate. Here is a summary of what gets changed:
Gene Czarcinski (1):
  v3 update anaconda to add reclaim

 pyanaconda/kickstart.py | 8 ++++++++
 1 file changed, 8 insertions(+)

Gene Czarcinski (1):
  v3 update pykickstart to add reclaim to btrfs command

 pykickstart/commands/btrfs.py | 5 +++++
 1 file changed, 5 insertions(+)
Why isn't this something that should work?

Gene

_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/anaconda-devel-list




[Index of Archives]     [Kickstart]     [Fedora Users]     [Fedora Legacy List]     [Fedora Maintainers]     [Fedora Desktop]     [Fedora SELinux]     [Big List of Linux Books]     [Yosemite News]     [Yosemite Photos]     [KDE Users]     [Fedora Tools]
  Powered by Linux