"fatal" compile - can't find compiled kernel
 

I have slogged through the kernel 2.4 'make menuconfig' process.
'make dep' seemed to go okay
'make bzImage', ditto
'make modules' had problems:
Message at end of 'make modules install

Warning device 0x0306 exceeds 1024 cylinder limit
Fatal: sector 51220658 too large for linear mode (Try 'iba32' instead)
make[1]: *** [install] Error 1
make[1]: Leaving directory 'usr/src/linux/arch/i386/boot'
make: *** [install] Error 2

can anyone help me with this?

PS:
Should I be able to find a compiled kernel by the name of bzImage, or has the above problem defeated the creation of this file? I am working with Red Hat Linux v7.0 and am upgrading to kernel 2.4. I am using a floppy disk to boot Linux. Is this where my kernel is? Is it also on my HD? If so, what is it called?


[Edited by rdaves@earthlink.net on 06-06-2001 at 01:01 AM]

Re:
 

While the error I'm thinking of is not in the place I would expect to see it, it; looks like a LILO error. I don't recall the simple fix, but I've been just making a (very) small /boot partition so that lilo can see it from boot. I've been doing this so long now that I haven't actually looked at how the new version of lilo works. Does newer lilo need some kernel support?

mcleodenine
 

I haven't got a clue if LILO requires kernel support or not.
I don't know if I understand your point. I have not gotten to a point where I would be using LILO. I was finishing up the compile of my kernel upgrade.

Re: mcleodenine
 

It looks to me like the kernel compile went ok as it appears that the error was generated after the make, after the make install, after the vmlinuz was copied to your /boot. You're doing this from the canned 'make bzImage' which does all the grunt work mentioned above for you. (Build/Backup/Copy/rerun lilo). How big is your first linux partition? IIRC, lilo can only see the first 1024 cylinders. It needs to see all the cylinders participating in a partition. If the boot partition has more than this lilo will spaz on you.

Here's my sorta standardized partition scheme (for a single non-LVMn non-RAID device)

/dev/hda1 /boot # Usually about 24MB, 2048byte inode
/dev/hda2 /swap
/dev/hda3 /root

re-reading your commands...
my kernel compiles go like this
make dep (on the first run)
make bzImage (well actually bzlilo cuz I am lazy)
make modules
make modules_install (note the underscore)
lilo (after we check our lilo.conf)


The kernel image you are looking for should be 'vmlinuz', but as I said your skrips should move your vmlinuz for you. I have had probs in the past where I d/l'd a new kernel that wanted to put the built parts in different dirs than SuSE uses (like '/'). Where did ya get the source?

mcleodenine
 

I either got the source (linux 2.4.0.tar.gz) from http://www.kernel.org, or http://www.redhat.com
By the way, you used "IIRC" or something like that in your post. What does that mean? I am really a newbie.

I checked my linux partitions. I have 3 Linux partitions:
-
23 MB - Start Sector 51,215,283, total sectors 4812
4212MB - Start Sector 51,263,478
70MB Swapfile - Start Sector 59,890,383
-
If you check the error message in my first post, the problem sector seems to be on the 23MB disk at sector 51,220,658. I don't know where this partition came from or why it is there or what is on it.



[Edited by rdaves@earthlink.net on 06-06-2001 at 08:06 AM]

Re: mcleodenine
 

IIRC == If I recall correctly. (IIRC...)

I guess the next option is to see if you can have a look at the /etc/lilo.conf and give us a hint about the mount points in /etc/fstab.

This is of course assuming that lilo is the problem...

Maybe wait and see if someone else has an idea.

mcleodnine
 

Well, no one else seems to be able to help. So, if you don't mind, I would like to continue trying to solve this enigma.
I am not sure we are looking at the right stuff, because my /etc/lilo.conf and /etc/fstab files, at this time, are the ones that are being used by the kernel v2.2.16.22. I am trying to install the kernel 2.4. I also don't understand why we are looking at lilo.conf, since I don't think (correct me if I'm wrong) this file has been created yet for kernel 2.4. But anyhow,just in case I am wrong, here is my /etc/lilo.conf/
-
boot=/dev/hda
map=/boot/map
install=/boot/boot.b
prompt
timeout=50
message=/boot/message
linear
default=linux

image=/boot/vmlinuz-2.2.16-22
label=linux
read-only
root=/dev/hda7

other=/dev/hda1
label=dos
-
and here is my /etc/fstab:
-
LABEL=/ / ext2 defaults 1 1
LABEL=/boot /boot ext2 defaults 1 2
/dev/cdrom /mnt/cdrom iso9660 noauto,owner,ro 0 0
/dev/cdrom1 /mnt/cdrom1 iso9660 noauto,owner,ro 0 0
/dev/fd0 /mnt/floppy auto noauto,owner 0 0
none /proc proc defaults 0 0
none /dev/pts devpts gid=5,mode=620 0 0
/dev/hda8 swap swap defaults 0 0
-
I am pretty much stuck at this point

Wait a minute... Did you boot with a 2.4.x kernel from the floppy? I was under the impression that you had to be running the kernel you want to build... (?)

Am booting from floppy
 

Sorry for the lack of communication. Up until last night, I was booting RedHat linux v7.0, kernel 2.2 from a floppy. Last night, I said "to heck with it" and finished up all the steps to get kernel 2.4 running. It actually ran, but with some red 'failure' notices during the boot. I think one of the problems is with the mixer, because on shutdown, I saw a failure notice that said 'aumix: error opening mixer', the other message was 'problem with NFS lockd'. It all went by so fast, it was difficult to capture these messages. I am guessing that these failures are the same ones that came up during the boot process. How do I log the boot process so I can look at the errors? Where do we go from here?

You can have a look at what hapened during the boot by either using 'dmesg' or having a look at /var/log/boot.msg . You might be able to glean some more information about what went wrong by having a look at /var/log/messages.