[3649] in Release_7.7_team
Re: Next question from Ji-yong Wang
daemon@ATHENA.MIT.EDU (Greg Hudson)
Sat Dec 21 10:10:50 2002
From: Greg Hudson <ghudson@MIT.EDU>
To: William Cattey <wdc@mit.edu>
Cc: release-team@mit.edu
In-Reply-To: <F4857C80-14A4-11D7-9587-000393995C5C@mit.edu>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Date: 21 Dec 2002 10:10:47 -0500
Message-Id: <1040483447.19068.22.camel@error-messages.mit.edu>
Mime-Version: 1.0
On Sat, 2002-12-21 at 00:27, William Cattey wrote:
> Is it intended that an update that cannot complete due to lack of space
> will be allowed to begin, and then abort mid-stream?
Yes, but the update will abort before any potentially harmful changes
are made to the machine. I agree that it would be psychologically more
agreeable to do the space check before printing the "Beginning update"
message, but that's not easy. Space and dependency checks are done by
the RPM library as part of preparing a transaction to run, so it all has
to happen within rpmupdate.
