Description of problem: During the installation of Fedora Core 4 (using Anaconda) I got: Assertion (head_size <= 63) at disk_dos.c:580 in function prob_partition_for_geom() failed after clicking [ignore] Assertion (sectors <= 63) at disk_dos.c:590 in function prob_partition_for_geom() failed after clicking [ignore] Assertion ((C * heads + H) * sectors + S == A at disk_dos.c:593 after clicking [ignore] The first message reappears... After clicking [ignore] repeatdly, the same messages return a couple of times and then I can go to the partitions, following to the rest of the configuration normally. At the end these messages return, the partitions seem to be created (the West HD seems unformated to Windows), but the installation aborts with the dumping I intend to attach here (as soon as I find how). I have tried to do the partitions automaticatically, manually, and deleting the old partitions (only in my second HD) as mentioned in bug <a href=https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=138419#c150>138419, comment #150</a>. My system is a HP Pavilion 514n with the original HD 60Gb (Samsung SV0602H). I added 512Mb RAM and a Western Digital 80Gb (WDC WD800JB) connected as 2nd slave (oddly, the Western HD could only be connected as slave in the first IDE; I couldn't find any combination of jumpers that could make it get along with the Samsung). My plan is to maintain Windows while I learn how to use Fedora, for emmergencies. I intend to install Fedora on the West HD and also put my data on it. I hope Windows will be able to "see" them, since it will be a FAT 32 partition. If this is possible, I can work under Fedora, but I can return to Windows if I get stuck. My CD-ROM/burner combo (PHILIPS CDD5301) is the 2nd master, from where I booted with the Fedora CD#1. I am a Fedora first user, who is tired of Windows stuff (with which I have lived since the version 3.11). I had more than enough. Please, help me out of Windows!!! Thank you in advance. How reproducible: Every time Steps to Reproduce: 1. Boot with CD#1, Fedora Core 4 2. Starting installation with Anaconda 3. Accepting only partitioning of the second HD (the first has WinXP Home Edition) 4. Accepting all the other defaults 5. Start installation Actual results: Dump, saved to a floppy (attached to this report). Expected results: Fedora should finish installed on the second HD, having boot dual with WinXP on the first HD (which is partitioned by HP disk recovery) Additional info: I tried to submit this bug through the normal form, but Internet Explorer 6 showed an script error in your page: Invalid sintax at line 228, character 13 in https://bugzilla.redhat.com/bugzilla/easy_enter_bug.cgi
Created attachment 125661 [details] Dump from Anaconda installer
It is duplicated in https://bugzilla.redhat.com/bugzilla/show_bug.cgi? id=184032 because, after the submission of this report I received a message that "Product name is not valid. Go back and try again". I did, but then I realized that the report was created.
*** Bug 184032 has been marked as a duplicate of this bug. ***
(In reply to comment #0) I mispelled the error message. Where it appears: prob_partition_for_geom() failed Read probe_partition_for_geom() failed
I disconnected the Western Digital from my computer, but I still got the same message. The weird geometry is coming from the Samsung HD, which has the HP partition and WinXP installed. Since this is the disk I intend NOT to touch, I do not understand why Anaconda refuses to install Fedora in this computer. Is there a workaround if I disconnect the Samsung HD, plug the Western HD as master, install Fedora in it, and make a dual boot later? How? Could it be that the problem is the dual boot installer that is trying to modify /dev/hda? Anaconda says the dual boot information would be written there, but this hda has the HP factory partition that; from Windows, it is adviced to not touch it or else... :-) There are two partitions: one is FAT32 containing the "recovery" information, and the other is NTFS containing WinXP and my files. I'm less than beginner in Linux - step by step instruction, please.
*** This bug has been marked as a duplicate of 138419 ***
(In reply to comment #6) Chris, I found the bug 138419 before, and I also left a message there. I tried what is recommended there, but it did not help me. All the proposed workarounds there "fix" the wrong partitions by deleting them. Well, I cannot do that for two reasons: 1. It is a HP Pavilion and the partition is made with HP recovery disks. When I tried to install a regular Windows, the computer became unstable. There are some special (and hidden) Windows adaptations that only HP knows. 2. Since I have a second HD in my computer, I do not intend to touch the HP system. Anaconda's bug, however, complains about this HD that I do not intend to use, and is preventing me to install Fedora in the other. Can someone help me out?