Description of problem:
Anaconda crashed in handleRenderCallback reading disk 3 during an upgrade
install of RHL 9.0 over RHL 7.1. Just before the crash, disk 3 was prompted,
and OK was pressed on the prompt. The disk was read for several seconds, then
the crash happened. At that point, Red Hat Linux 9.0 install aborted.
Version-Release number of selected component (if applicable):
Every time. Bug was reproduced twice using the same steps.
Steps to Reproduce:
1. Boot RHL 9.0 disk 1 on a machine with RHL 7.1 installed.
2. Indicate an upgrade install on original root partition.
3. Select defaults otherwise.
4. Do as directed by prompts through disks 1 and 2.
5. Insert disk 3 when prompted. Press Ok.
RHL 9.0 install aborted.
RHL 9.0 should have continued.
See attached dump.
Created attachment 91182 [details]
Dump from Anaconda.
Can you attach the /etc/lilo.conf.rpmsave file?
I plan on attaching /etc/lilo.conf.rpmsave some time next week, week of 28
April - 2 May 2003. This week, 21-25 May, I am in class, and do not have
access to the office Linux computer that experienced the bug.
Actually, it occurs to me that I no longer have /etc/lilo.conf.rpmsave from RHL
9.0, and it is not present on my RHL 8.0 installation. Sorry. You may wish to
consider performing the steps described to reproduce the problem, at which
point presumably a meaningful /etc/lilo.conf.rpmsave would be generated. Hope
RH9.0 upgrade from RH7.1
I just had the same thing happen yesterday. Anaconda crashed at the package
configuration step, after packages were installed. I'll attach the anaconda
dump. Unfortunately, I don't have the lilo.conf.rpmsave as I decided to do a
clean install (i had backups :-)
I did grab this from the screen when anaconda dropped out:
---- snip ----
XLIB:unexpected async reply (sequence 0x33cab2)!
The program 'anaconda' received an X Window System error.
The error was 'BadCursor'.
(Details: serial 3394227 error-code 6 request-code 95 minor-code 0)
---- snip ----
I had chosen to change my bootloader from Lilo to Grub. When I rebooted I got
only an LI, so I think the boot.map was toast. I went in linux single and poked
around a bit, then decided to clean install.
Created attachment 92201 [details]
dump from anaconda
Made some cleanups to hopefully fix this