[26604] in Athena Bugs
Re: Subtle logic error in installer's analysis of pre-existing
daemon@ATHENA.MIT.EDU (andrew m. boardman)
Tue Sep 13 23:55:53 2005
Message-Id: <200509140355.j8E3tDck016608@pothole.mit.edu>
To: Bill Cattey <wdc@mit.edu>
In-Reply-To: Your message of "Mon, 12 Sep 2005 16:35:15 EDT."
<1126557315.22857.68.camel@tokata.mit.edu>
Date: Tue, 13 Sep 2005 23:55:13 -0400
From: "andrew m. boardman" <amb@mit.edu>
X-Spam-Score: 1.041
X-Spam-Level: * (1.041)
X-Spam-Flag: NO
cc: bugs@mit.edu
Errors-To: bugs-bounces@mit.edu
> It took me a moment to realize that there was a bug here:
Happily, there's not, though the actions of the installer are perhaps
less than clear if one is used to the standalone installer which creates
a separate /boot. (Which, at this point, is done so that partitioning on
public machines looks largely the same.)
> Install Athena, blow Athena away with Windows, but leave the
> /boot partition as the first partition on the disk.
> [...]
> The installer sees the /boot partition, and rather than realizing
> its part of a normal Athena disk, says there's something already there
> and asks if you want to boot it.
Some more background after talking to Bill: when he did the MS-Windows
install, it saw the bootable partition and volunteered to not overwrite
it, filling the rest of the disk with NTFS. The subsequent Athena
install downsized the NTFS partition and installed itself in one logical
partition, not creating or using a /boot partition at all, thus leaving
/dev/hda1 around as an orphan to confuse the issue.
