Red Hat Bugzilla – Bug 74844
Installer gives runtime error right after starting up
Last modified: 2007-04-18 12:46:59 EDT
From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.0.0) Gecko/20020530 Description of problem: Good CD1 (did a verify on it and it passed... and it installs great on other machines) that crashes after the "running anaconda (the install program)" message (or whatever the exact wording of it is) with a runtime error. See "Additional Information" section for details. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1.boot install CD 2.Hit enter (default install) or try "linux text", "linux noprobe" and "linux lowres" 3.anaconda runs and crashes with runtime error (see Additional Information section for details) Actual Results: anaconda runs and crases with runtime error (see Additional Information section for details) Expected Results: Install program should progress normally and allow me to install. Additional info: I get the following Python runtime error messasage from Python: Traceback (most recent call last): File "/usr/bin/anaconda", line 384 in ? from installclass import DefaultInstall File "/usr/lib/anaconda/installclass.py", line 27 in ? from constants import BETANAG ImportError: cannot import name BETANAG install exited abnormally (etc) reboot (etc) Please note, this is the real RHL 8.0 and not a beta or a leaked release canidate... the MD5checksums match up to those for RHL 8.0. Oddly enough, I installed all three betas for 8 (including Null) and never ran into this problem. FIC motherboard running an AMD Duron with VIA chipset (can get more sprecifics if requested) and a Voodoo 3 PCI (not AGP) video card. I'm willing to try various command line options to the installer but I've already tried all common methods (mentioned in the install screens) and it runtime errors regardless.
Ok, more to the story... to clarify... I DID DO a media check on CD 1 and it passed. Upon futher examination, an alternate console (either ALT-F3 or ALT-F4, I don't remember which) showed <3> decompression error (or something like that) over and over... so for some reason my brand new Philips CDRW2400 services burner couldn't read the disk. I tried two other disks (both PASSed the media check) and they did the same thing. To get the blasted machine installed... I decided to do a network install... by taking the first CD1 (as well as CD2 and CD3) and created an install server on another machine in the house. Then I used the same CD1 to boot from and used the "linux askmethod" boot prompt option and did an NFS install. Since it didn't read much of the CD, I was able to install. I guess I really am not looking for resolution to this problem because it appears beyond the control of the anaconda installer... BUT it is good to note that just because a CD passes the media check (yes, even on the target install machine), that doesn't mean that the install machine will be able to successfully read the CD for an install.