This is something I have not been able to find much info on so far - maybe I am searching on the wrong keywords... anyway.... I am using SuSE7.3, vanilla full install. When I want to reboot the system, I logout, and click the Shutdown button. That presents me two options - Shutdown or Reboot. To the right of Reboot is a dropdown box. In previous versions of SuSE Linux I could click that dropdown and select one of my LILO options such as WinNT or SuSE Safe mode. Well now in 7.3 that box is still there, but there is nothing in it. I cannot pre-select how I want the system to restart... I have to wait and pick it from the boot menu after the BIOS has doen it's thing. What option have I missed (messed with?) to empty this menu out? How can I enable it or fill it back up with the LILO options? C.
On Monday 25 February 2002 21:15, Clayton Cornell wrote:
This is something I have not been able to find much info on so far - maybe I am searching on the wrong keywords... anyway....
I am using SuSE7.3, vanilla full install. When I want to reboot the system, I logout, and click the Shutdown button. That presents me two options - Shutdown or Reboot. To the right of Reboot is a dropdown box. In previous versions of SuSE Linux I could click that dropdown and select one of my LILO options such as WinNT or SuSE Safe mode. Well now in 7.3 that box is still there, but there is nothing in it. I cannot pre-select how I want the system to restart... I have to wait and pick it from the boot menu after the BIOS has doen it's thing.
What option have I missed (messed with?) to empty this menu out? How can I enable it or fill it back up with the LILO options?
C.
What do you get if you do "/sbin/lilo -q", and what do you have in your /opt/kde2/share/config/kdm/kdmrc under the heading [Shutdown] Those two places are the most likely culprits. kdmrc will determine if lilo is to be used at all, and with what map file, and lilo -q is what fills the dropdown with values. If either is incorrect, things will fail regards Anders
I am using SuSE7.3, vanilla full install. When I want to reboot the system, I logout, and click the Shutdown button. That presents me two options - Shutdown or Reboot. To the right of Reboot is a dropdown box. In previous versions of SuSE Linux I could click that dropdown and select one of my LILO options such as WinNT or SuSE Safe mode. Well now in 7.3 that box is still there, but there is nothing in it. I cannot pre-select how I want the system to restart... I have to wait and pick it from the boot menu after the BIOS has doen it's thing.
What option have I missed (messed with?) to empty this menu out? How can I enable it or fill it back up with the LILO options?
What do you get if you do "/sbin/lilo -q", and what do you have in your /opt/kde2/share/config/kdm/kdmrc under the heading [Shutdown]
/sbin/lilo -q gives me the expected: Warning: BIOS drive 0x82 may not be accessible linux * failsafe nt memtest86 Well, OK, the warning was new to me, but the rest looks normal. In kdmrc the bit that looks relevant is: [Shutdown] HaltCmd=/sbin/halt LiloCmd=/sbin/lilo LiloMap=/boot/map RebootCmd=/sbin/reboot UseLilo=true
Those two places are the most likely culprits. kdmrc will determine if lilo is to be used at all, and with what map file, and lilo -q is what fills the dropdown with values. If either is incorrect, things will fail
Seems logical. Seems odd though that the dropdown box is like this (empty) right from the beginning of a default install. I have never seen anything in the box at all since I installed (and reinstalled) SuSE7.3. This is before I had a chance to get in and mess with things as I like to do - can't learn about it all if you don't screw things up a bt here and there ;-) I would expect that the default config as put on the SuSE CDs would work, and this would only get screwed up after I had been poking about and playing with things. C.
On Monday 25 February 2002 21:43, Clayton Cornell wrote:
/sbin/lilo -q gives me the expected: Warning: BIOS drive 0x82 may not be accessible linux * failsafe nt memtest86
Well, OK, the warning was new to me, but the rest looks normal.
Yep, but I think kdm's liloinfo barfs on the warning. It runs lilo -q and tries to parse the output, and I suspect this is throwing it. At least that's all I can think of, since your kdmrc looks just like mine. I suggest you investigate it, by for instance running lilo in very verbose mode and seeing why it's warning you //Anders
On Monday 25 February 2002 8:52 pm, Anders Johansson wrote:
Yep, but I think kdm's liloinfo barfs on the warning. It runs lilo -q and tries to parse the output, and I suspect this is throwing it. At least that's all I can think of, since your kdmrc looks just like mine. I suggest you investigate it, by for instance running lilo in very verbose mode and seeing why it's warning you
Interesting thread. I originally had a 30Gb drive with Windows on it (hda), and installed a second 4Gb one with Suse 7.1 (hdd). I could boot into both OSs OK. Then I took that out and replaced it with a 40Gb IBM drive. I had problems partitioning this, and had to use IBM's utility to "clear" the drive. Then I used Partition Magic to format it, and put 7.2 on it. Since then I haven't visited Windows again, but last month I tried to boot into it for a game, and couldn't, so I didn't bother. Then last week I decided to install 7.3, and went through and put lilo on the MBR. But it couldn't boot, and I had to install again and put lilo onto a floppy. lilo -q gives the same reading as Clayton's: Warning: BIOS drive 0x82 may not be accessible lilo -T geom=0x82 gives: bios=0x82, cylinders=1024, heads=255, sectors=63 EDD packet calls allowed I note that dmesg gives the following info: hda: 60036480 sectors (30739 MB) w/1916KiB Cache, CHS=3737/255/63, UDMA(66) hdd: 80418240 sectors (41174 MB) w/380KiB Cache, CHS=79780/16/63, UDMA(33) and later: hda: hda1 hda2 < hda5 hda6 hda7 hda8 hda9 hda10 hda11 > hdd: [PTBL] [5005/255/63] hdd1 hdd2 hdd3 hdd4 < hdd5 hdd6 hdd7 hdd8 hdd9 hdd10 hdd11 hdd12 hdd13 > Neither of these matches what lilo says, and I'm not clear why two similar sized disks should have such different CHS numbers. I think (correct me if I'm wrong) that the [PTBL] [5005/255/63] for hdd relates to the CHS via the formula (C x H) / 255, ie (79780 x 16) / 25 = 5005 - this is something to do with LBA?? Anyway, It'd be great to be able to shed a bit of light on this - somehow the drive geometry doesn't match somewhere, although the machine seems to be working fine, apart from the lack of being able to dual-boot. Kevin
On Tuesday 26 February 2002 07:48 am, Kevin Donnelly wrote: ----------snip----------------
Interesting thread. I originally had a 30Gb drive with Windows on it (hda), and installed a second 4Gb one with Suse 7.1 (hdd). I could boot into both OSs OK. Then I took that out and replaced it with a 40Gb IBM drive. I had problems partitioning this, and had to use IBM's utility to "clear" the drive. Then I used Partition Magic to format it, and put 7.2 on it. ----------snip---------------- lilo -q gives the same reading as Clayton's: Warning: BIOS drive 0x82 may not be accessible
lilo -T geom=0x82 gives: bios=0x82, cylinders=1024, heads=255, sectors=63 EDD packet calls allowed
---------snip--------------------
I also get that message but I just assumed that lilo only "guaranteed" the first 2 disks to be bootable (hda and hdb) which in my case are both on the first ide controller. I actually do use a boot partition on the third disk as well, lilo is just warning that it may not work. dh
On Tuesday 26 February 2002 16:48, Kevin Donnelly wrote:
On Monday 25 February 2002 8:52 pm, Anders Johansson wrote:
Yep, but I think kdm's liloinfo barfs on the warning. It runs lilo -q and tries to parse the output, and I suspect this is throwing it. At least that's all I can think of, since your kdmrc looks just like mine. I suggest you investigate it, by for instance running lilo in very verbose mode and seeing why it's warning you
Interesting thread. I originally had a 30Gb drive with Windows on it (hda), and installed a second 4Gb one with Suse 7.1 (hdd). I could boot into both OSs OK. Then I took that out and replaced it with a 40Gb IBM drive. I had problems partitioning this, and had to use IBM's utility to "clear" the drive. Then I used Partition Magic to format it, and put 7.2 on it.
This sounds familiar. I have done a similar thing wiht my system. I posted a while ago with drive problems after formatting a new drive to FAT32 using Win2k. I have to wonder if my Linux install is seeing the new drive as 0x82. I am hoping to play around with the drives this weekend (if I have time) using this suggestion I was given then (and haven't had time to try yet): -=-=-=-=-=-=-=-=-===- repartition the disk using linux fdisk and create a partition table with the following entries: 01 extended (full disk size) 05 logical1 (20G) make this type 0b 06 logical2 (20G) make this type 0b 07 logical3 (17G) make this type 0b after fdisking the disk (assume its hdb) do the following dd if=/dev/zero of=/dev/hdb5 bs=512b count=2 and do the same with /dev/hdb6 and /dev/hdb7 sync sync init 6 and format the drives from windows2K using Fat32. -=-=-=-=-==-=-=-=-=-=-
lilo -T geom=0x82 gives: bios=0x82, cylinders=1024, heads=255, sectors=63 EDD packet calls allowed
With lilo -T geom=0x82, I get: bios=0x82, cylinders=4865, heads=255, sectors=63 EDD packet calls allowed
I note that dmesg gives the following info: hda: 60036480 sectors (30739 MB) w/1916KiB Cache, CHS=3737/255/63, UDMA(66) hdd: 80418240 sectors (41174 MB) w/380KiB Cache, CHS=79780/16/63, UDMA(33) and later: hda: hda1 hda2 < hda5 hda6 hda7 hda8 hda9 hda10 hda11 > hdd: [PTBL] [5005/255/63] hdd1 hdd2 hdd3 hdd4 < hdd5 hdd6 hdd7 hdd8 hdd9 hdd10 hdd11 hdd12 hdd13 >
dmsg on my system gives me: hda: 16514064 sectors (8455 MB) w/418KiB Cache, CHS=1027/255/63, (U)DMA hdb: 117266688 sectors (60041 MB) w/1819KiB Cache, CHS=7299/255/63, UDMA(66) hdc: 78165361 sectors (40021 MB) w/1024KiB Cache, CHS=77545/16/63, UDMA(66) and of course.... my old friend: hdb: dma_intr: status=0x51 { DriveReady SeekComplete Error } hdb: dma_intr: error=0x84 { DriveStatusError BadCRC } Looking at this now, it looks like the info for 0x82 matches one of the partitions on my 60Gig drive (hdb). I have been having problems accessing this drive from Linux (all partitions on hdb are fine when accessing from Windows2k, but Linux can only see the first partition)... thus the re-fdisk and re-format instructions I included above. So... the big experiment... archive the data I have on my 60Gig drive, do the steps to fdisk and format... try reading the drive, check what lilo -q has to say after hdb is supposedly fixed. Any other comments of suggestions before I launch into this?? C.
Hi Good point from David - this may be so. On Wednesday 27 February 2002 6:11 am, Clayton Cornell wrote:
and of course.... my old friend: hdb: dma_intr: status=0x51 { DriveReady SeekComplete Error } hdb: dma_intr: error=0x84 { DriveStatusError BadCRC }
A bad cable? I had this with a CD-RW until I changed the cable.
So... the big experiment... archive the data I have on my 60Gig drive, do the steps to fdisk and format... try reading the drive, check what lilo -q has to say after hdb is supposedly fixed. Any other comments of suggestions before I launch into this??
Well, the thread has gone silent! I offer you best wishes in your experiment - please report back, and then I might summon up the nerve to do something similar! Kevin
and of course.... my old friend: hdb: dma_intr: status=0x51 { DriveReady SeekComplete Error } hdb: dma_intr: error=0x84 { DriveStatusError BadCRC }
A bad cable? I had this with a CD-RW until I changed the cable.
Thought so originally. I went out to my local computer shop and spent quite a few Euros on some high quality drive cables... or so they claimed. Anyway, they are the third new set of cables in 6 months and the problem is still there.... but on a different drive now... the new drive that I am going to reformat this weekend. I am hoping that the problem will go away then.
Well, the thread has gone silent! I offer you best wishes in your experiment - please report back, and then I might summon up the nerve to do something similar!
Yah.. I'm keeping this thread in my SuSE list folder. I'll send something of a result on Sunday night. Later C.
Well, I re-fdisked my 60Gig drive and finally have access to all partitions. No errors so far.... and now there is a noticeable absence of DMA errors. Thing is, lilo -q still gives the same information: -=-=-=-=-=-= lilo -q Warning: BIOS drive 0x82 may not be accessible linux * failsafe nt memtest86 -=-=-=-=-=- lilo -T geom=0x82 bios=0x82, cylinders=4865, heads=255, sectors=63 EDD packet calls allowed -=-=-=-=-=-=- I have no idea now which drive is 0x82. Anyone have any suggestions? Also, still no solutions to the Reboot dropdown menu missing it's LILO options. LILO seems to be OK, and boots everything fine from a clean boot. The nice little SuSE/LILO menu pops up when I boot, and I can select my various boot options. The bit that still does not work is the reboot menu in the X login screen. Not critical, just annoying, and I am pedantic enough to want to figure this one out. C.
On Saturday 02 March 2002 10:11, Clayton Cornell wrote:
Well, I re-fdisked my 60Gig drive and finally have access to all partitions. No errors so far.... and now there is a noticeable absence of DMA errors.
I spoke too soon. They are back. Sigh. I did some messing about with my system over the last couple hours, and it actually seems to be a problem with my 3rd IDE port on my motherboard. I swapped cables around, and the DMA errors didn't follow the cables... I swapped drives around, and the DMA errors stayyed with what ever drive was connected to the 3rd IDE port. Odd... annoying... wish it would go away. C.
participants (4)
-
Anders Johansson -
Clayton Cornell -
David Herman -
Kevin Donnelly