System description: WS 630 prototype Dual 800 MHz CPUs 128 MB ECC RDRAM 40 MB IDE HDD Matrox G400 AGP Factory download correctly connects to the network and downloads the RPMS. The installation fails with an anaconda error (line 426) with manual install from HDD or CDROM and with kickstart install. The anaconda dump is
Created attachment 2275 [details] ANACDUMP.txt
Created attachment 2276 [details] PCI.TXT
The anaconda dump is attached along with the results of cat /proc/pci.
Is this hardware generally available? If not, what's the possibilty of sending it to our Durham site so we can reproduce this prior to release? Time is SHORT -- RC2 is 21-Aug. Please reply ASAP.
I sent the system to Doug Ledford I checked the Airborne status and it arrive at RedHat on 14 August D. Long signed for the system. Any question please contact me directly. Rogelio_Noriega
The traceback is because the kernel (via uname) reports the arch is i?86 (yes ?). RPM tries to match up packages with the architecture, so it finds no matches, causes the problem you saw. There are additional issues related to this which are kernel related. I'm assigning this bug to the kernel team.
There are many potential problems here. Intel says that our current kernel, lacking FXSAVE/FXRESTOR, will simply not work reliably with the P4. I patched procfs to return 6 instead of ? (UGLY patch) and will do the same for the uname system call.
*** This bug has been marked as a duplicate of 16294 ***