Brian Ashe wrote: > -------- Original Message -------- > Subject: Re: mxt_scan_bios > Date: Sat, 9 Feb 2002 01:13:23 -0500 > From: Brian Ashe <[EMAIL PROTECTED]> > Reply-To: [EMAIL PROTECTED] > Organization: Dee-Web Software Services, LLC. > To: [EMAIL PROTECTED] > References: <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> > > Steve, > > On Saturday 09 February 2002 12:06, you said something about: > > But now that I think about it (see source code below) it seems that > > mxt_scan_bios is probably not the culprit at all. > > > > I think the routine just returns that error code and the caller records > > that there is no mxt memory and goes on with its next task. Which is, > > from the looks of my listings when I successfully boot from the floppy, > > the Swap file. That might bear some relation to my earlier thoughts on > > the matter - namely disk partitioning anomalies. If this is correct, > > then the actual crap-out point is either in the routine that calls > > mxt_scan_bios, or in the swap routine, before it has a chance to print > > out its message. > > > > Anyone following me, here? > > > > Well, I don't want you to start getting lonely... ;) > > When you upgraded, did you choose ext3 for the partitions (the old > ext2 ones)? > > Have you checked /boot for corruption? <- Seems likely read on. > > What does your partition layout look like? > > I recently did a dual boot set-up with RH7.2 (using ext3) and Win98 > that had > some unusual problems with disk corruption (this is why I bring this > up). I > have been meaning to look into it more so I can get some reproducible > details > and submit a bug on it (but alas, life interferes too much with my good > intentions). > > Here is what I did and the results... > > Created the following partition layout: > /boot ~30MB first partition > C:/ ~2GB second partition > Extended for the rest > D:/ ~1GB > /usr > /home > /var > /tmp > swap > (and a few others all logical) > > Installed Win98 > Installed RH7.2 > > All was fine at first (both booted and ran). > > Then I put some thing on the D: drive and it all went screwy(I didn't > immediately see that this was the problem). It didn't immediately die or > anything, but it did run out of drive space on C: within a couple of > minutes > (well that was the ever so intelligent message from Windows anyway). It > claimed I had used up 12GB on C: (it's only an 8GB drive). > I booted using my boot floppy to Linux (neither was booting after the > write > to D:) and started to look into it. My /boot was toast. Everything was > gone > except kernel.h and even that was unreadable. > So I restored /boot (reinstalled the kernel rpms) and decided to use LILO > instead of GRUB (not a big fan of GRUB and was suspicious of it). > This got me back to booting Linux but not Win98. So I reinstalled > Win98 (no > format just overwrite) and then back to the boot floppy, rerun LILO and > everything seemed OK again (boot wise). > So back to Win98 again. As I started doing stuff I noticed the files I > had > previously put on D: were gone. I decided to restore them and then the > connection started being made, as the previous problem started up again. > So this time as I went through the cycle described above I decided to > eliminate the D: drive (just removed the partition and left it as free > space) > and it has been fine ever since. > > Notes and observations: > This was the first time I used RH7.2 with ext3 on a dual boot with > Win98 that > had more then a C: drive. I had upgraded a 7.1 to 7.2 but kept the > ext2 on a > Win98 with C: and D: and had no problems, as well as several 7.2 with > ext3 > with Win98 with only a C: each running fine. I cannot say (as I > haven't had > time to play) if there is a real connection here, but it is the > direction I'm > leaning. > I have never had any problems with ext3, but several people have noted > that > when using ext3 with a squid cache it would suffer disk corruption. > Perhaps > there is some strange mojo in ext3 that is causing these unusual > occurrences, > but I am far from a file system expert to be able to accurately say > (at least > until I waste a day playing with this due to curiosity ;). Or perhaps > GRUB > fudges the partition table beyond repair under these circumstances (I > keep > trying it but have always reverted back to LILO for some reason or > another). > > Hopefully all of this will give you something to try or some ideas > about what > is happening. Good luck. > > > <big snip>
Thank you Brian. I knew this saga would find a knowledgeable person sooner or later but it wasn't easy to make that happen. I kept having to omit details to shorten the posts down to where someone would read it. So if you were by some amazing coincidence to have looked at all my posts here for the last three days you would have discovered that YES, I did the ext3 thing as part of the upgrade, a detail that I myself had forgotten about in all that ensued! OK. What would corruption in /boot look like? Here's what I find doing a listing:total 8483 drwxr-xr-x 4 root root 1024 Feb 7 01:46 . drwxr-xr-x 20 root root 4096 Feb 9 06:04 .. -rw-r--r-- 1 root root 512 Feb 7 01:23 boot.0300 -rw-r--r-- 1 root root 512 Jul 6 2000 boot.0302 -rw-r--r-- 1 root root 5824 Jun 24 2001 boot.b -rw-r--r-- 1 root root 612 Jun 24 2001 chain.b drwxr-xr-x 2 root root 1024 Feb 7 14:42 grub -rw-r--r-- 1 root root 325093 Feb 6 23:25 initrd-2.4.7-10BOOT.img -rw-r--r-- 1 root root 332039 Feb 7 01:01 initrd-2.4.7-10debug.img -rw-r--r-- 1 root root 330775 Feb 6 23:09 initrd-2.4.7-10.img -rw------- 1 root root 1048576 Feb 6 23:01 .journal lrwxrwxrwx 1 root root 14 Feb 7 00:44 kernel.h -> kernel.h-2.4.7 -rw-r--r-- 1 root root 405 Feb 7 00:44 kernel.h-2.4.7 drwxr-xr-x 2 root root 12288 Jul 6 2000 lost+found -rw------- 1 root root 14848 Feb 7 01:46 map -rw-r--r-- 1 root root 23108 Jun 24 2001 message -rw-r--r-- 1 root root 21282 Jun 24 2001 message.ja lrwxrwxrwx 1 root root 20 Feb 6 23:09 module-info -> module-info-2.4.7-10 -rw-r--r-- 1 root root 13598 Sep 6 16:39 module-info-2.4.7-10 -rw-r--r-- 1 root root 13598 Sep 6 15:43 module-info-2.4.7-10debug -rw-r--r-- 1 root root 640 Jun 24 2001 os2_d.b lrwxrwxrwx 1 root root 19 Feb 6 23:09 System.map -> System.map-2.4.7-10 -rw-r--r-- 1 root root 435039 Sep 6 16:39 System.map-2.4.7-10 -rw-r--r-- 1 root root 255994 Sep 6 15:26 System.map-2.4.7-10BOOT -rw-r--r-- 1 root root 454347 Sep 6 15:43 System.map-2.4.7-10debug -rwxr-xr-x 1 root root 2828866 Sep 6 15:43 vmlinux-2.4.7-10debug lrwxrwxrwx 1 root root 16 Feb 6 23:09 vmlinuz -> vmlinuz-2.4.7-10 -rw-r--r-- 1 root root 0 Jul 6 2000 vmlinuz- -rw-r--r-- 1 root root 802068 Sep 6 16:39 vmlinuz-2.4.7-10 -rw-r--r-- 1 root root 671411 Sep 6 15:26 vmlinuz-2.4.7-10BOOT -rw-r--r-- 1 root root 1025338 Sep 6 15:43 vmlinuz-2.4.7-10debug -rw-r--r-- 1 root root 0 Jul 6 2000 vmlinuz---help Now that you mention it, a lot looks strange here, the strangest being that boot.b is from my OLD installation. How do I fix that? Do you see anything else here? What's up with all these zero-length files? Partition layout: [root@sleepingbear boot]# /sbin/fdisk /dev/hda The number of cylinders for this disk is set to 2482. There is nothing wrong with that, but this is larger than 1024, and could in certain setups cause problems with: 1) software that runs at boot time (e.g., old versions of LILO) 2) booting and partitioning software from other OSs (e.g., DOS FDISK, OS/2 FDISK) Command (m for help): p Disk /dev/hda: 255 heads, 63 sectors, 2482 cylinders Units = cylinders of 16065 * 512 bytes Device Boot Start End Blocks Id System /dev/hda1 * 4 1275 10217308+ c Win95 FAT32 (LBA) /dev/hda2 1 3 24066 83 Linux /dev/hda3 1276 2482 9695227+ 5 Extended /dev/hda5 1276 1471 1574338+ 83 Linux /dev/hda6 1472 1602 1052226 83 Linux /dev/hda7 1603 2042 3534268+ 83 Linux /dev/hda8 2043 2091 393561 82 Linux swap /dev/hda9 2092 2482 3140676 83 Linux Partition table entries are not in disk order Command (m for help): Let's see. /dev/hda1 is Drive C: /dev/hda2 is /boot /dev/hda4 (not shown) is drive E: /dev/hda5 is /home /dev/hda6 is / /dev/hda7 is /usr /dev/hda8 is swap /dev/hda9 is /usr/local Anyway, if you or anyone else can ge me going with this info, I owe you one. Steve _______________________________________________ Redhat-list mailing list [EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
