I can not go to # prompt

Please see http://wdb1.sco.com/kb/showta?taid=105411 for fixing root filesystem problems

The problem here is that when I power on the PC, and after typing <enter> when "boot:" appears, just before I can see the (Type CRTL+D or password ....) to enter in single user mode I receive the following:

fsstat: root file system needs checking.
OK to check the root file system (/dev/root)? (y/n)

if I answer "yes" I get the message posted in my first post and if I answer "no" I get nothing, so independently of my answer I cannot enter, the system does not respond anymore.

I have used the emergency disks to boot up to single user mode and I tried /etc/fsck -ofull <Enter> but I receive an error saying that /etc/checklist could not be found so I tried fsck -s -ofull -D  /dev/hd0root and this found some problems so I chose to repair them and the final message I receive is that the filesystem was modified. At this point everything seems to be ok but when I reboot I have the same problem as nothing had been done.

Thanks

After you get into single user mode, did you following the procudure in:
     http://wdb1.sco.com/kb/showta?taid=105411

If you still have problem, used the emergency disks to boot up to single user mode , then
run "divvy" to find out all the filesystem device (Warming, do NOT make any change to
the partion table, just write down the partion name, otherwise you will LOST data!)

then do a fsck to all the filesystem, this will involved a lot off human intervention, if you are
 not sure what to do you need to get someone know SCO (NOT other Unix favour!) to do it
for you.

       

Have you tried booting the system without any operator intervention? The file "ioctl.syscon" may be corrupted.  You may want to boot from your boot/root floppies and go thru your filesystem checks first.  After doing that do a reboot on the machine and let it boot on its own.  when the boot: prompt comes up "DO NOT" press enter--let it time out, which should be about 60-seconds.  By doing the fsck's from floppy above the system should not stop and ask about doing a filesystem check.  If the decision sceen about pressing <CTL>-D or entering the system password to go to maintenance mode comes up "DO NOT" do anything here either--let it time out--should be about 30-seconds.  When the date and time query comes up "DO NOT" enter anything here either--let it time out.  If the system comes up to a login you should login as root then go to the etc directory and do the following:

     # ls -l ioctl.syscon

you should get something like:

     lrwxrwxrwx  1 root     sys   44 Aug 12 2004 ioctl.syscon > /var/opt/K/SCO....

If you see the "l" at the beginning of the permissions then cd to the path shown after the greater than symbol then "remove" the ioctl.syscon file at that location.

if the permissions above start with a dash (-rw...) then remove the ioctl.syscon file from the etc directory.

After removing the ioctl.syscon file do a shutdown and reboot.  You should be able to boot as you normally would.  The file /etc/ioctl.syscon will be recreated during the boot process.  

This is a known "gothca" on openserver 5.  The system only checks for the presence of the file at boot time it does not check it for corruption.  If the file is not found a new one is created.  But if it is there the system will just go on--however if the file is corrupt what you have described above can happen.

Good Luck

yuzh,

I followed the procedure step by step and after reboot I was able to enter normally, without any warning or error message. I thought all my problems had gone but when I reboot the computer for a second time I had exactly the same problem.

sppalser,

I want to probe your solution but at this time the only way to enter the system is using the emergency floppies, I cannot boot the machine on its own because it hangs up after listing the devices during the boot process.

So, do you guys think this problem can be solved?? I have used all the utilities to verify system integrity and in theory there have been fixed all the problems by these utilities but after a reboot the problem is present again. What else can I do??

Thanks!

Should have mentioned as well, when booted on the boot/root set, did you check your filesystems for free space?  Full root filesystems can cause a lot of grief.

Good Luck

Well, I finally was able to get the system come up using the "unix.safe" as you pointed and I followed all steps in the SCO's knowledge center's article but once again all this was unuseful. Tomorrow I'm going to open the machine to make a good cleaning as you suggest.

I have checked the filesystem for free space and this is not the problem. And in relation to your question about my method to reboot, I have been doing this using "init 6" or "reboot" but also due to my problem I have been doing a lot of cold boots when the machine hangs up.

Tomorrow I'll let you know the situation after cleaning the machine.

also check /var/adm/messages file to see if any errors.

I cleaned the machine, all cards and cables but still no luck. I checked /var/adm/messages and there aren't errors. At this time the only way to get the system come up is using the "unix.safe" alternate kernel, any other ideas???

Is it something about the kernel??

Well I think it is a issue with the kernel cause continuing with different tests I have discovered that if the kernel is rebuilded, for example, if I change the nic configuration and let the system to rebuild the kernel and set it as default to boot, when I reboot I can enter the system normally, without any error message. I can reboot as many times as I want and everything is ok, but sadly I also discovered that if I turn off the machine, when I power on it again I'm in the beginning, I have the same problem.

Any comment would be appreciated!!

Boot up your box again, use scoadmin to reconfig you the kernel (get rid of all the devices
which is no longer in your box), relink the kernel, and reboot.

if the kernel is clean up and you still have trouble, the next thing to look at your hardware,
eg mainborad etc.

Well I think I'm going to close this question, I can't figure out what the problem is but I have had the machine working for the last two days without problems and as I said in a previous post I can reboot normally but when I shutdown I have the problem again. However I have now the recipe to come up the system again; I have to enter using "unix.safe" then I relink the kernel and finally I reboot the system.

I know this is not the best solution but I really don't know why I have this problem everytime I shutdown the system and assuming that I'm not going to be shuting down the system frequently I can live with this.

Thank you guys, I'm going to split the points because I have learned a lot from both of you.