fedora 8:booting problem
 

hey
i installed fedora 7 from the dvd that i downloaded from the fedora site.
Everything went nice through the installation but when i reboot
the following message comes up before loading:

"Module signature verification failed"
"insmod:error inserting '/lib/ohci-hcd.ko:key was rejected by service"
"Unable to load SELinux Policy "
and then it says panic and stops

What might the problem?

So not fedora 8 then (see title)?
the module signiture has expired.
disable selinux.

http://www.linuxquestions.org/questi...nstall-554637/

yep sorry for the missleading title.

The problem is i am trying to install either of the two versions on my p.c and none of them works grrrr.

I tried to reinstall F.7 and add the selinux=0 command but the problemas reappear before even finishing the installation.

If you can help me on this one it would be a great deal!

The usual error i encounter is " Error rpmdb : Fatal error panic!!"

I installed Fedora 8 once but not with full success , after finishing the installation some errors occured on the screen
while it was restarted which i could not read. Then i said ok lets work with that but the installation was full of problems,
yum was crushing all the time and i thing with a common error like the "rpmdb".

Is any of these problems coming from my hardware (which i thing not) because is the only way of explaining it since i tried 4-5 different
cd's and dvd's alla checked and passed through integrity tests!

Is there a way of installing either of them fully through the internet ??

The only reason I've gotten errors during any installation was because of a bad CD/DVD burn. Why don't you try burning Fedora again but at the slowest possible speed? I know it'll be time consuming, but it's worth it.

Brandon

The rpm database is corrupt. As Mr MyHeartPumpsFreon says, reburn the DVD at lowest speed.

hmmm

Thats why i mentioned in my previous post that i have used 4-5 different cd's .
i have burnt them in different cd-writers, i downloaded the from different mirrors, i have tried mostly everything
. I make avery time the test that says in the begining before installation and it passes it every time.
But my guess is also that is a cd-dvd problem so i will give it another 2-3 shots!

Thanks a lot.

ok i tried once again and the problem still remains
I burned a couple of cd's again, and i also used an old one with which i installed fedora 7 that i am using this moment.
The problems are always around this rpmdb an the installer seems to fail.
Even with the old dvd that did not have anu problems still the same message appears.

Now i am using my old installation in an old hard drive and i have problems with yumex again with this rpm thing.

My only thought is that something remained on the old hard disk concerning this rpmdb and it causes the problem.
How could i make a full format of the hard disk using fdisk???

i tried some things by my self with fdisk
and this are the results

Command (m for help): p

Disk /dev/sda2: 98 MB, 98703360 bytes
255 heads, 63 sectors/track, 12 cylinders, total 192780 sectors
Units = sectors of 1 * 512 = 512 bytes

Device Boot Start End Blocks Id System
/dev/sda2p1 63 192779 96358+ 83 Linux

Command (m for help):

the problem is that my hard drive is NOT 98MB but 80GB.
how can i format it???

ok here it is againn excuse me because i am a bit rushy..

i suppose the rest of the hard drive is on /dev/sda3

because i get


[root@noname sbin]# ./fdisk /dev/sda3
Device contains neither a valid DOS partition table, nor Sun, SGI or OSF disklabel
Building a new DOS disklabel. Changes will remain in memory only,
until you decide to write them. After that, of course, the previous
content won't be recoverable.


The number of cylinders for this disk is set to 7382.
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)
Warning: invalid flag 0x0000 of partition table 4 will be corrected by w(rite)

Now can a put them together again and make a single partition for the whole thing?
And will that help with the problems with the installation??

On consideration:

Try the install again, choose the option to install to the entire drive. Check the created partitions... that they occupy the entire drive. Preferably use a custom scheme, create boot, swap, and root, but do not use LVM. Watch for errors.

If firstboot fails after that, boot from the DVD to rescue mode, and run "fdisk -l" to find the partitions: do they agree with what was done in anaconda? Then run fsck on each partition (except swap).

Then get back to me.

i followed your instruction and here is the result

At first the installing went ok once again but when it started rebooting it showed some error messages after leaving anaconda
which i could not see cause it was going fast,but it wasnt about rpmdb this time.
is there any log file that i can find them in???

I splited my hard disk in
/sda1 /boot 4GB
/sda2 / 70GB
/sda3 swap 2 GB

but there is always some space that remains unused about 500 MB.

Then the first boot didnt work out it freezed on
"Uncompressing Linyx....Starting the kernel"

Afterwards i rebooted from the rescue mode of the cd

i run fdisk and the partitions where correct.
then i run fsck on the two partitions and this is the result

fsck on /boot
recovering journal
clean 32/1025024 files somethin...blocks

fsck on /
recovering journal
clean bla/blabla files ....... blocks

The appropriate logfiles are like this:
/var/log/boot.log
/var/log/syslog*
/root/firstboot.log

It would mostly help to see the entire fsck output.

Run:

fsck -aC /dev/sda1
fsck -aC /dev/sda2

ok here is the deal:

I have a new hard disk exactly the same as before,right ou tof the box!
I start to install from one of my F7 dvds that i had before and was working.
i passed it through the test and started installing.
i splitted the disk as before

when it was about to finish it crashed on "installing the bootloader"
Ok i start a new installation with another dvd which i downloaded. Exactly the same thin happened.
Then i make an installation with the first dvd but with text mode.

I did this before because is more easy to see the error, and indeed the usual error ocured during the installation.
It said 'rpmdb-Fatal region error detected, but it still continued to install stuff and the error stayed on screen.
When it was finished this time the bootloader was installed ok.
So i rebooted, the booting failed with the mesage
"kernel panic-not syncinc:VFS:Unable to mount root fs on unkniwn_block(0,0)"

Then i tried to go to rescue mode and after 2-3 tries i managed to do it.
so following your instructions again i did, fdisk and it was ok.
But the fsck command came up with different things than before:

fsck /dev/sda1 (the /boot)
/boot1 was not cleanly unmounted,check forced
pass1 : Checking inodes,blocks and sizes
pass2:Checking directory and stucture
pass3 checking directory and connectivity
pass4 cheking reference counts
pass5 checking group summary information

/boot 1 32/512512 files(3.1%-contiguous) 17585/512064 blocks

then i did fsck again and it said:

/boot1 clean 32/512512 files 17585/512064 blocks

Then the same for the other partition
fsck /dev/sda3
/1 was not cleanly unmounted,ckeck forced
ok here is the deal:

I have a new hard disk exactly the same as before,right ou tof the box!
I start to install from one of my F7 dvds that i had before and was working.
i passed it through the test and started installing.
i splitted the disk as before

when it was about to finish it crashed on "installing the bootloader"
Ok i start a new installation with another dvd which i downloaded. Exactly the same thin happened.
Then i make an installation with the first dvd but with text mode.

I did this before because is more easy to see the error, and indeed the usual error ocured during the installation.
It said 'rpmdb-Fatal region error detected, but it still continued to install stuff and the error stayed on screen.
When it was finished this time the bootloader was installed ok.
So i rebooted, the booting failed with the mesage
"kernel panic-not syncinc:VFS:Unable to mount root fs on unkniwn_block(0,0)"

Then i tried to go to rescue mode and after 2-3 tries i managed to do it.
so following your instructions again i did, fdisk and it was ok.
But the fsck command came up with different things than before:

fsck /dev/sda1 (the /boot)
/boot1 was not cleanly unmounted,check forced
pass1 : Checking inodes,blocks and sizes
pass2:Checking directory and stucture
pass3 checking directory and connectivity
pass4 cheking reference counts
pass5 checking group summary information
/1*****File system was modified**************
/1 103620/18513920 files(0,1%non-contuguous),1164711/18512904 blocks

I cant really understand if it is a matter of the dvd's or the hardware.

Thanks for the help anyway|!

Not normally the only message that goes with this error.
The usual solution is to rebuild the database:
http://www.netadmintools.com/art451.html

Need to see grub.conf in the drives boot partition.

What does this mean? You boot from the DVD, at the boot: prompt, you enter "linux rescue", and then what? It took 2-3 goes before it would boot?

I need to see the output!

fdisk -l
cat /boot/grub/grub.conf

If you have a removable media handy, you can mount that and pipe the outputs to a text file on the media. eg:

mount -t vfat /dev/fd0 /media/floppy
fdisk -l > /media/floppy/fdisk.txt
cat /mnt/hda1/boot/grub/grub.conf > /media/floppy/grub.txt

Then you can just copy the text over. (Above commands will need to be modified to your exact setup.)

Lastly - what is this HW?
It is possible that you happen to have one of those rare systems which confuse anaconda. Have you tried running live distros?

everything seems to work right now that i formated my hard disk through gparted!
The install is ok and everything seems to work fine

But i am still having problems with yum and yumex

Sometimes its ok and sometimes crushes GOod time!

i get the usual error rpmdb: PANIC: fatal region error detected; run recovery
and i trie the usual thing with

removing the __db*
and then rpm --rebuilddb

but it doesnt rebuild it!
it crushes saying

"error: error(-30977) getting "&ãÉwo2Œ¯„M¶‹¸f½Ä±{" records from Filemd5s index
rpmdb: PANIC: fatal region error detected; run recovery"
and a hundred messages like this with a different part after "getting"

is there a way of reinstalling yum from the start or something like this???