Linux

Ask something
Discussion Locked

Inode on SCO Server Marked Bad.

0 helpful
7 replies

by chaik99

Closed to new repliesThis discussion is closed. It stays readable, but it takes no new replies.

Hi, I need some help ASAP. The SCO Unixware server crashed early morning.The error message is "hardware error" condition has been detected. There are whole bunch of error after that. I am not able to use the keyboard anymore at that time, so have topower it off & on. I did this two times, finally get back in.The second time I reboot the server, there are messages stated that root file system inode xxxx marked bad. Altogether there are four inode been marked bad, then there is one message pop out stated on the disk driver, block xxx "remapped". Now the server is up and running. I am thinking if any one can give me some opinion or good link to do more research on inode & how to fix/replace the primary disk on the server.
Thanks
chaik

7 replies

by jbullard

Be careful...even though the system rebooted, you may not be able to access the data in the files where the indoes were marked as bad. You may want to unmount the drive and run a FULL fsck. As far as a link goes try http://uw7doc.sco.com/ODM_FSadmin/fssag-11.html
(this is for the inode problem but you can use the navigation links to get to other useful pages).

by johan

With the program ncheck (look in manual pages) you can discover which files are connected to the bad marked inodes. May be this can help you any further. Generally a problem like yours is due to a bad spot on the harddisk residing in one of the superblocks of the filesystems. If you have SCSI you can scan for it with the /etc/scsibadblk program, with IDE you can use /etc/badblk. Don't mix them, you will get a lot deeper into trouble.

by gloria.weimern

As the machine is booting hit the appropriate keystrokes to enter SCSISelect Utilities (usually Ctrl+A). Then select SCSI Disk Utilities. Select the ID line for the hard disk (Usually 0). Select Verify Disk Media.
This utility will look for bad spots, try to move the data, and update the media error map.

This discussion is closed to new replies.