Bug 85833 - Anaconda handles missing packages poorly
Summary: Anaconda handles missing packages poorly
Alias: None
Product: Red Hat Public Beta
Classification: Retired
Component: anaconda
Version: phoebe
Hardware: i386
OS: Linux
Target Milestone: ---
Assignee: Jeremy Katz
QA Contact: Mike McLean
Depends On:
TreeView+ depends on / blocked
Reported: 2003-03-08 20:19 UTC by Zenon Panoussis
Modified: 2007-04-18 16:51 UTC (History)
0 users

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Last Closed: 2003-03-11 17:33:46 UTC

Attachments (Terms of Use)

Description Zenon Panoussis 2003-03-08 20:19:34 UTC
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.3b) Gecko/20030210

Description of problem:
Normally, if anaconda doesn't find an rpm that is selected for installation, or
finds one that is corrupted, it returns an error message with the alternatives
"cancel" and "try again". 

On errors on mtr-gtk-0.52-2.i386.rpm, lynx- and
ncftp-3.1.5-4.i386.rpm this behavious fails. Anaconda reports "fatal error" and
only offers the option to reboot. 

In my case, these files (and many others) were indeed corrupted and they
installed fine when I replaced them with good copies. The problem is not in the
packages themselves, but in anaconda's error handling: on these particular files
it does not offer the option to try again. Result: the installation has to be
started all over from scratch and annoyance reaches red level after the third time. 

Version-Release number of selected component (if applicable):

How reproducible:
Didn't try

Steps to Reproduce:
1. Ruin one of the packages in question
2. Try an installation that includes that package.

Additional info:

Comment 1 Zenon Panoussis 2003-03-08 20:40:33 UTC
Note: Using NFS and the graphical installer.

Comment 2 Jeremy Katz 2003-03-11 17:33:46 UTC
If they're corrupted on the server and we don't find out until rpm is unpacking
them, there's unfortunately not much we can do :(

Comment 3 Zenon Panoussis 2003-03-12 00:41:24 UTC
Well yes, there is: make sure that anaconda asks instead of just crashing. It
does that already on other corrupt packages, so why not on these three too? To
me, it looks like a forgotten if/then error check, nothing much more. 

Note You need to log in before you can comment on or make changes to this bug.