by Hi,
This is the system:
Linux 2.0.27 ELF (* No modules *)
Pentium 100
32mb Ram
2 x 4.3 GB segate HD
2940 scsi controller
device mount Blocks inodes
/dev/sda1 / 466518 120832
/dev/sdb1 /usr/local/news 383750 205200
/dev/sdb2 /var/spool/news 3301902 3784704
/dev/sda2 /var/spool/news/alt 3475842 1855952
The Problem.
frequently (* > 1/week *) the system gets a
Jan 8 08:26:46 news kernel: Kernel panic: EXT2-fs panic (device 08:12):
ext2_read_inode: unable to read i-node block - inode=3711252,
block=3711029
What I tried.
I have removed the filesystems, reformated with the scsi format utility
recreated the file systems, re-installed the system. I had assumed the
problem was pointing to a bad location on the hard drive but the block
number in not consistant for the errors. The machine it running INN and
no news stays longer then a week, so id it were a bad block we would see
the panic for the the same bad block every week. (* does e2fsck add bad
blocks to the bad block file.*)
The questions
What are your guesses to try next?
Is there a way to set up a bad block redirection area?
How is the bad block file created?
Is there a utility to scan for bad blocks that is nondestructive to the
file system?
Please respond by e-mail, and thanks in advance
Vince
--
Vince Di Scipio
Network Administration
The College of Wooster
Wooster Oh 44691