asked on

Reset root password for UnixSco 7.1.3 on IBM SCSI HDD

The X Admin change the root password, I have no other user account to access the server and no emergency disk, it's UNIXSco 7.1.3 working in IBM machine SCSI HDD. from the original CD same as the installed on the server I installed on another normal IDE computer as password reset testing scenario. I create the emergancy disk by using "# /sbin/emergency_disk diskette1" command from this testing machine and it give me "Creation of the Emergency Recovery boot floppy was successful" and I use it, it give me HDD is sane then I select option 1 to mount the file system then I went to shell and change both passwd and shadow files and then rebuild them again and I success to reset the password, but when I start on the server, cause I have no access so I have to push the shutdown or reset, I used same emergency Disk which I create it in the testing PC but on the server it give me HDD not sane and it didn't give me the first option for mount the file system,then I choose directly to access the shell and I tried to execute the fsck, mknod and mount but all not working, it says "UX:ksh: ERROR: mknod: not found" or in mount it says "Cannot open special: /dev/hda1" or "Cannot open special: /dev/hd0root" or "Cannot open special: /dev/root" or for fsck it says "UX:fsck:ERROR: cannot open vfstab" when I start the server again it works fine and no HDD error on the Hardware log on BIOS, I wanna know First: why HDD not sane? is it cause I press shutdown? or ther is other reason?, Second: can I use the same emergency disk from the testing IDE HDD PC to reset password on the server SCSI?, Third: how to make the Server SCSI sane to enable the first booting option "mount file system"? otherwise how to mount the file system manually? cause all steps after this I know and try it and it works fine.

Luqman Shantal

Confirm whether /etc/vfstab is available. If it is not present, restore it from the backup . .

http://manuals.fujitsu-siemens.com/fgm/pcl42a20-sol-cd2_6_2007/documents/manuals/En/GFS/gfsae/gfs00539.htm

omarfarid

Can you clarify what unix version / platform you are using and what server H/W?

yuzh

Please have a look at my ansewer in http:Q_20194019.html

> why HDD not sane?

for Unix machine, you should never just turn off the power of a Unix box, it can cause more problem, always login as root and shultdown the system properly.
man shultdown
to learn more details.

BTW, with SCO Unix, you can use "haltsys" to shultduwn the box.

Why you need to run mknod?
If your emergency FDs is created from the same box, the kernel should have no problem with all the HW devices for the box.

nasiradin

ASKER

Hi msconfig:

I limit the problem in small area. and what you are talking about is totally different from whati'm facing.
----------------------------------
Hi omarfarid:

Please read the title of this post.

----------------------------------
Hi yuzh:

Thanx for reply. all what you said I know and already as i wrote in my above question i did it and it works , my problem now is when I boot the server it give me message the HDD not sane, them how to mount it? please read my problem again.

yuzh

you need to run fsck to all filesystem, then mount the root
filesystem.

you can run "divvy" command to find out how the HD was partioned, and then fsck to the root filesystem, then get rid of the root password.

also have a look at the following doc:
"Lost Root Password SCO Unix":
http://aplawrence.com/Unixart/lost_root_password.html

nasiradin

ASKER

Dear All,

Thanx for cooperation, but I found the solution by my self.

yuzh

Please post your solution, and PAQ this question.

Thank you!
yuzh

ASKER CERTIFIED SOLUTION

nasiradin

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial