From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1b) Gecko/20020722
Description of problem:
This was an attempt to install the (Null) beta on a spanish Toshiba Satellite
I chose a new install of the Null beta, in Spanish(Spain) and with the Spanish
keyboard selected. Everything worked fine from the beginning (except from
ISOLINUX failing to determine the block size, and assuming 0. The installation
then went through without problems until the installer hanged towards the end
of the package installation, while installing postfix (this may not be significant).
The system was responsive, though: I could move the mouse pointer, and switch
to the consoles at F1 through F5. When switching back to F7 (the installer
graphical display) only gray and white areas where shown (where the text used to
be during installation). I can still move the mouse pointer, but nothing happens.
There were no error messages at any console (F1 through F5). From the bash
shell, I killed the anaconda process which had a parameter specifying the cdrom
drive (as shown from the output of the "ps" command). At this point the CDROM
(the second one) was ejected, and the console at F1 said that the installer
program terminated abnormally, and it was shutting the system down. I had to
turn the machine off manually.
Version-Release number of selected component (if applicable):
When you run ps see if anything else is running (like the %post of the postfix
I tried to reproduce it and look for the %post in the process list,
but this time the installation worked without problems on the same machine.
I wasn't able to reproduce the bug.
In case this is relevant, I am adding this comment. After the installation
worked correctly, I booted the system and played around a bit, and then I
realized that "postfix" had not been selected during system installation. I
later on added postfix with the package tool and it got installed fine.
reporter (incorrectly) changed bug status from NEEDINFO to MODIFIED. The
MODIFIED state means that *code* has been modified, not that the bug report has.
Since you weren't able to reproduce this, I'm going to go ahead and close this
bug. If you see the same problem with the next release, then reopen it.