Hi everybody As I don't need Windows anymore I freed the small partition where the rest of Win was and installed SUSE 10.0 (downloaded version on CD's) a second time: I wanted to have another SUSE install so that in case of "something happens" to my running system I still have a bootable system. Install went fine, I could boot the new (second) SUSE and the "normal" too. Then I run YOU, among others installing the new kernel 2.6. But after reboot I get Error 18: selected cylinder exceeds maximum supported by BIOS I found some explanations about this message, saying "This error is returned when a read is attempted at a linear block address beyond the end of the BIOS translated area. This generally happens if your disk is larger than the BIOS can handle (512MB for (E)IDE disks on older machines or larger than 8GB on others.). In more practical terms this means the BIOS is unable to start executing the kernel because the kernel is not located within the block it can access at boot up time." (from http://wiki.linuxquestions.org/wiki/GRUB ) I understand that I should make a separate, small /boot partition in the first 1024 sectors. But what I wonder about is: - why does my "normal" SUSE 10 (same kernel, no separate /boot partition) boot without any poblem? - why did the new SUSE 10 boot easily before the YOU update? This makes me think, thant it can't be a problem of the BIOS nor of the disk size...??? Here are some details that could maybe help to locate the problem: - kernel used in both installations: 2.6.13-15.8-default "old" (running) Suse 10.0 GRUB is installed in the MBR "new" (non-booting) GRUB is installed in the boot sector of /dev/hda1 menu.lst (of "old System") looks like this: ----------- color white/blue black/light-gray default 0 timeout 8 gfxmenu (hd2,0)/boot/message title SUSE LINUX 10.0 root (hd2,0) kernel /boot/vmlinuz root=/dev/hdg1 vga=0x31a selinux=0 acpi=off resume=/dev/hde1 splash=verbose showopts initrd /boot/initrd title SUSE LINUX 10.0 (/dev/hda1) root (hd0,0) kernel /boot/vmlinuz root=/dev/hda1 vga=0x31a selinux=0 acpi=off resume=/dev/hde1 splash=verbose showopts initrd /boot/initrd title Diskette chainloader (fd0)+1 title Failsafe -- SUSE LINUX 10.0 root (hd2,0) kernel /boot/vmlinuz root=/dev/hdg1 vga=normal showopts ide=nodma apm=off acpi=off noresume selinux=0 nosmp noapic maxcpus=0 edd=off 3 initrd /boot/initrd --------------- fstab (of running system) looks like this: /dev/hdg1 / reiserfs acl,user_xattr 1 1 /dev/hda7 /eigene reiserfs acl,user_xattr 1 2 /dev/hde3 /home reiserfs acl,user_xattr 1 2 /dev/hda6 /opt reiserfs acl,user_xattr 1 2 /dev/hde2 /srv reiserfs acl,user_xattr 1 2 /dev/hda3 /tmp reiserfs acl,user_xattr 1 2 /dev/hda2 /usr reiserfs acl,user_xattr 1 2 /dev/hda5 /var reiserfs acl,user_xattr 1 2 /dev/hda1 /suse10.0resc reiserfs acl,user_xattr 1 2 /dev/hde1 swap swap defaults 0 0 proc /proc proc defaults 0 0 sysfs /sys sysfs noauto 0 0 usbfs /proc/bus/usb usbfs noauto 0 0 devpts /dev/pts devpts mode=0620,gid=5 0 0 /dev/dvd /media/dvd subfs noauto,fs=cdfss,ro,procuid,nosuid,nodev,exec,iocharset=utf8 0 0 /dev/cdrecorder /media/cdrecorder subfs noauto,fs=cdfss,ro,procuid,nosuid,nodev,exec,iocharset=utf8 0 0 /dev/hdb4 /media/zip auto noauto,user 0 0 /dev/fd0 /media/floppy subfs noauto,fs=floppyfss,procuid,nodev,nosuid,sync 0 0 /dev/hdg2 /fotosave reiserfs acl,user_xattr 1 2 -------------- where /dev/hda1 (/suse10.0resc) is the partition of the new non-booting install Can anybody tell me what was going wrong with the new install after YOU? thanks for help. Daniel -- Daniel Bauer photographer Basel Switzerland professional photography: http://www.daniel-bauer.com special interest site: http://www.bauer-nudes.com
On Sunday 02 April 2006 07:20, Daniel Bauer wrote: <snipped and trimmed> /dev/hda1 /suse10.0resc reiserfs acl,user_xattr 1 2 /dev/hda2 /usr reiserfs acl,user_xattr 1 2 /dev/hda3 /tmp reiserfs acl,user_xattr 1 2 /dev/hda5 /var reiserfs acl,user_xattr 1 2 /dev/hda6 /opt reiserfs acl,user_xattr 1 2 /dev/hda7 /eigene reiserfs acl,user_xattr 1 2 /dev/hde1 swap swap defaults 0 0 /dev/hde2 /srv reiserfs acl,user_xattr 1 2 /dev/hde3 /home reiserfs acl,user_xattr 1 2 /dev/hdg1 / reiserfs acl,user_xattr 1 1 /dev/hdg2 /fotosave reiserfs acl,user_xattr 1 2 Hi Daniel, Can you please post /suse10.0resc/etc/fstab? regards, Carl
Am Sonntag, 2. April 2006 18:23 schrieb Carl Hartung:
On Sunday 02 April 2006 07:20, Daniel Bauer wrote: <snipped and trimmed> /dev/hda1 /suse10.0resc reiserfs acl,user_xattr 1 2 /dev/hda2 /usr reiserfs acl,user_xattr 1 2 /dev/hda3 /tmp reiserfs acl,user_xattr 1 2 /dev/hda5 /var reiserfs acl,user_xattr 1 2 /dev/hda6 /opt reiserfs acl,user_xattr 1 2 /dev/hda7 /eigene reiserfs acl,user_xattr 1 2
/dev/hde1 swap swap defaults 0 0 /dev/hde2 /srv reiserfs acl,user_xattr 1 2 /dev/hde3 /home reiserfs acl,user_xattr 1 2
/dev/hdg1 / reiserfs acl,user_xattr 1 1 /dev/hdg2 /fotosave reiserfs acl,user_xattr 1 2
Hi Daniel,
Can you please post /suse10.0resc/etc/fstab?
regards,
Carl
here it is: /dev/hda1 / reiserfs acl,user_xattr 1 1 /dev/hdg1 /suse10/ reiserfs acl,user_xattr 1 2 /dev/hda7 /suse10/eigene reiserfs acl,user_xattr 1 2 /dev/hdg2 /suse10/fotosave reiserfs acl,user_xattr 1 2 /dev/hde3 /suse10/home reiserfs acl,user_xattr 1 2 /dev/hda6 /suse10/opt reiserfs acl,user_xattr 1 2 /dev/hde2 /suse10/srv reiserfs acl,user_xattr 1 2 /dev/hda3 /suse10/tmp reiserfs acl,user_xattr 1 2 /dev/hda2 /suse10/usr reiserfs acl,user_xattr 1 2 /dev/hda5 /suse10/var reiserfs acl,user_xattr 1 2 /dev/hde1 swap swap defaults 0 0 proc /proc proc defaults 0 0 sysfs /sys sysfs noauto 0 0 usbfs /proc/bus/usb usbfs noauto 0 0 devpts /dev/pts devpts mode=0620,gid=5 0 0 /dev/dvd /media/dvd subfs noauto,fs=cdfss,ro,procuid,nosuid,nodev,exec,iocharset=utf8 0 0 /dev/cdrecorder /media/cdrecorder subfs noauto,fs=cdfss,ro,procuid,nosuid,nodev,exec,iocharset=utf8 0 0 /dev/hdb4 /media/zip auto noauto,user 0 0 /dev/fd0 /media/floppy subfs noauto,fs=floppyfss,procuid,nodev,nosuid,sync 0 0 -------- Daniel Bauer photographer Basel Switzerland professional photography: http://www.daniel-bauer.com special interest site: http://www.bauer-nudes.com
On Sun, 2006-04-02 at 13:20 +0200, Daniel Bauer wrote:
- why does my "normal" SUSE 10 (same kernel, no separate /boot partition) boot without any poblem? - why did the new SUSE 10 boot easily before the YOU update?
When you install a package, where exactly it gets written on the partition isn't predetermined. The YOU update included an updated grub and/or kernel. It seems likely that before the update, they were placed where the BIOS could read them, but after the update they got placed elsewhere. If you have an old BIOS things like this can happen. This is why you should have a separate /boot partition entirely located within the 1024 cylinder limit, to prevent things like this from happening. If you have a partition that extends beyond it, it may work or it may not, it's up to the file system. On the other hand, you may also want to check for an updated BIOS that doesn't have this limitation
On Sunday 02 April 2006 12:33, Anders Johansson wrote:
When you install a package, where exactly it gets written on the partition isn't predetermined. The YOU update included an updated grub and/or kernel. It seems likely that before the update, they were placed where the BIOS could read them, but after the update they got placed elsewhere. If you have an old BIOS things like this can happen.
Hi Anders (grovel grovel) :-) With all due respect (boatloads!) I don't think this is Daniel's problem. First, he describes it as "the small partition." Since it is the first primary on hda and it previously held a Windows installation, it probably has the lowest cylinder range of all the partitions on that drive. Add to this the fact that he was able to boot the installation before applying the updates and it looks (to me, anyway) more like the kernel update, itself, got mangled. It is possible that the installer became confused by the de facto 'mirror' installation residing on hdg. In my mind, in aggregate, all of these factors combine to rule out the liklihood that this is a true "out of BIOS address range" scenario. And, from what he's written, it also looks like he applied the kernel patch concurrently with the other patches. That is almost always a recipe for trouble like this. I'd like to know what the system 'looks like' to the other installation before I formulate or recommend a course of action. That's why I've asked Daniel to post the other fstab. Hope all is going well for you! regards, Carl
Am Sonntag, 2. April 2006 18:33 schrieb Anders Johansson:
On Sun, 2006-04-02 at 13:20 +0200, Daniel Bauer wrote:
- why does my "normal" SUSE 10 (same kernel, no separate /boot partition) boot without any poblem? - why did the new SUSE 10 boot easily before the YOU update?
When you install a package, where exactly it gets written on the partition isn't predetermined. The YOU update included an updated grub and/or kernel. It seems likely that before the update, they were placed where the BIOS could read them, but after the update they got placed elsewhere. If you have an old BIOS things like this can happen.
This is why you should have a separate /boot partition entirely located within the 1024 cylinder limit, to prevent things like this from happening. If you have a partition that extends beyond it, it may work or it may not, it's up to the file system.
I could try this, but I am afraid of adding an additional partition because, as you can see, I have other patitions on /dev/hda used by the working system. What happens to /dev/hda2-7, including the "extended partition" /dev/hda4, when I split /dev/hda1 in two partitions (/boot and /)? Then they will get other numbers, don't they? And what will my now working installation "think" of this? Sorry, if these questions are stupid, but I'd rather be too careful than a bit too little...
On the other hand, you may also want to check for an updated BIOS that doesn't have this limitation
I don't want to do this, as my "normal" system just runs perfect and also here I am afraid that with a change or update I could ruin something... Daniel -- Daniel Bauer photographer Basel Switzerland professional photography: http://www.daniel-bauer.com special interest site: http://www.bauer-nudes.com
So, I finally got the second install running, too... Don't know exactly where the error came from, but this is what I did: According to a PM from Carl I edited the fstab's of both the existing system and the new "fallback" system and removed in each all partitions that belong to the other. So now every system has a fstab mentioning only it's own partition(s). Then I did a complete new install of the "fallback" system on /dev/hda1. Again according to Carl's PM I did not install GRUB again (it is already installed in the MBR from the running system), I just said "install GRUB to floppy disk", but haven't used that floppy later. Just to see if it works I rebooted after the install finished. As I already had edited menu.lst before I could easily boot the new system from there. I have been so far before, too... Then I splitted up the YOU updates, to see where the failure could have arised. After each "part" downloaded and installed by YOU I re-booted to see if it still works. These are the parts: 1) unchecked everything but the new kernel 2) unchecked everything but GRUB-update 3) unchecked everything but Yast2 liby2util, perl-bootloader, yast2-booloader, yas2-users, yast2-packagemanager, insserv 4) unchecked everything but xorg-x11-server, x11-tools, xorg-x11. xorg-x11-libs, xorg-x11-server 5) all the rest So, I guess, I did a lot of unnecessary re-booting, but at least the new system now works perfectly, too... I have no idea why it didn't work last time, could be: - kernel update should be done separately, and/or - two GRUB installs (one in MBR, one in boot sector of /dev/hda1) don't work together, or - fstab containing mountpoints of other system is not liked by YOU regards Daniel -- Daniel Bauer photographer Basel Switzerland professional photography: http://www.daniel-bauer.com special interest site: http://www.bauer-nudes.com
participants (3)
-
Anders Johansson -
Carl Hartung -
Daniel Bauer