So the installer starts ok, enter, enter etc, unlock crypto, calculate dependencies, then it tries to copy install.img to the system (to where?) and complains about an unspecified error.
the logs are also not so clear as to what fs needs room. There is plenty.
ERROR: failed to calculate free space for None does make it even more vague.

I read:
The /boot filesystem is used to store the kernel and initial ramdisk images that form the core of the Fedora operating system. required by several utilities including grub, kernel and preupgrade. 

This means that a BIOS workaround is abused for installation.
This is the 21st century.
Why are we fixed to /boot?
Why do we assume that we waste that much room in there?

/, /var etc may have more room.

The "failed to calculate free space for None" really has nothing to do with anything.  This is the real error message:

anaconda 12.46 exception report
Traceback (most recent call first):
  File "/usr/lib/anaconda/isys.py", line 173, in umount
    rc = _isys.umount(what)
  File "/usr/lib/anaconda/storage/formats/fs.py", line 600, in unmount
    rc = isys.umount(self._mountpoint, removeDir = False)
  File "/usr/lib/anaconda/storage/formats/fs.py", line 790, in teardown
    return self.unmount(*args, **kwargs)
  File "/usr/lib/anaconda/storage/__init__.py", line 1855, in umountFilesystems
    device.format.teardown()
  File "/usr/lib/anaconda/storage/__init__.py", line 1052, in umountFilesystems
    self.fsset.umountFilesystems(ignoreErrors=ignoreErrors, swapoff=swapoff)
  File "/usr/lib/anaconda/yuminstall.py", line 834, in run
    self.anaconda.id.storage.umountFilesystems()
  File "/usr/lib/anaconda/yuminstall.py", line 1585, in doInstall
    rc = self.ayum.run(self.instLog, cb, anaconda.intf, anaconda.id)
  File "/usr/lib/anaconda/backend.py", line 291, in doInstall
    return anaconda.backend.doInstall(anaconda)
  File "/usr/lib/anaconda/dispatch.py", line 200, in moveStep
    rc = stepFunc(self.anaconda)
  File "/usr/lib/anaconda/dispatch.py", line 123, in gotoNext
    self.moveStep()
  File "/usr/lib/anaconda/gui.py", line 1195, in nextClicked
    self.anaconda.dispatch.gotoNext()
  File "/usr/lib/anaconda/iw/progress_gui.py", line 79, in renderCallback
    self.intf.icw.nextClicked()
  File "/usr/lib/anaconda/gui.py", line 1216, in handleRenderCallback
    self.currentWindow.renderCallback()
SystemError: (2, None)

When you hit this error, can you please switch over to tty2 by pressing ctrl-alt-f2, run the following commands, and attach their output to this bug report?  Thanks

mount
umount /mnt/sysimage/var

I guess so.
Meanwhile I used the yum method to upgrade to F12.
No big deal?

Well, the fact that you've upgraded means that you now won't really be able to upgrade with anaconda since there will be no packages to change.  I'm not optimistic about your chances to be able to reproduce, and if you can't reproduce it then we may have to close this bug report as INSUFFICIENT_DATA unless someone else here can reproduce.

Hmm.
What info do you really need?
Maybe I can provide it any other way?

The basis is the weird idea to use /boot to put install.img and more stuff.
/boot is an evil BIOS workaround. Do not expand it's use any further.
Please find a different spot or make it flexible enough to use whatver place that has room.

Then there's the bug message itself that is very hard to understand.
Make it say what came before and after and what it wants.
Not just a stack dump that even with all the info that  was in the log doesn't give you enough clue(s).

Then the handling of the error: There is no way of retrying or recovering.

The shutdown happens in some other screen. No clear message to the user what state the system is in. (yes I could figure it out but not everybody works like I do)

Upon reboot we see fsck complain about superblocks with times in the FUTURE. These should be avoided. Not everybody lives in the USA.

The statement

The "failed to calculate free space for None" really has nothing to do with
anything.  This is the real error message:

is a puzzling.

I just reported all stuff that appeared on my screen and was put in the logs.
I did not program this, nor requested install.img to be parked in /boot. (see my rant 1 message up)

If they are unrelated I should file another bug?
Please say so and specify what I should tell in there.

> Hmm.
> What info do you really need?
> Maybe I can provide it any other way?

I really need what I mentioned in comment #4.

> Then there's the bug message itself that is very hard to understand.
> Make it say what came before and after and what it wants.
> Not just a stack dump that even with all the info that  was in the log doesn't
> give you enough clue(s).
>
> Then the handling of the error: There is no way of retrying or recovering.

If it were an expected error that we could provide a way of retrying, then there would be a nicer looking error message with useful buttons.  But it's not an expected error - it's an exception, which means a bug, which means we don't know what buttons to provide or how to retry.  And the information required varies from bug to bug so we can't grab everything we'd ever need or it would be too unwieldy.

Could I fake the upgrade by downgrading some rpm so that the process will work as in a full upgrade? (not too damaging but working...)

That seems likely to cause further problems and make things more confused, but feel free to give it a try and let us know if you reproduce the original problem.

What rpm would be most suitable?

(In reply to comment #12)
> What rpm would be most suitable?  

Some rpm high up in the application stack, like firefox or so, so that you don't break the basic functionality of your system.

That might destroy some my data.
Any other takers that are less risky?

(In reply to comment #14)
> That might destroy some my data.
> Any other takers that are less risky?  

Removing firefox the application will not remove your firefox profile which has
your bookmarks, etc. It will only remove the firefox binary and accompanying libs, etc.

Clearing all need info's, so as to clear anaconda team need info.

*** This bug has been marked as a duplicate of bug 548592 ***

Which one has the lower number?
So what does it show?