Re: reproducible filesystem corruption with 2.0.37pre4

Theodore Y. Ts'o (tytso@MIT.EDU)
Tue, 12 Jan 1999 00:16:45 -0500


Date: Sat, 9 Jan 1999 00:14:24 +0000 (WET)
From: Carlos Fonseca <cmfonsec@ualg.pt>

I have been experiencing silent filesystem corruption with 2.0.37pre4,
always on the same partition (sdb1, >2GB) of the same harddisk. Basically,
the filesystem is supposed to be clean, but running e2fsck on it reports
summary information errors. Once, the machine froze up completely in X
(I am not on a network, so could not check whether that was still active).
I am attaching the output of e2fsck corresponding to that freeze, and
another one. Block group 17 seems to be always involved, but that may have
to do with just how full the partition is.

Sounds like a bad block, probably located in the inode table of block
group 17. Upgrade to a newer e2fsprogs (at least version 1.13; 1.14
will probably will be available by the time you read this), and then run
e2fsck -c; this will force a badblocks scan, which will hopefully find
the bad block(s), and cause e2fsck to attempt to fix the filesystem by
avoiding to use the bad block(s).

- Ted

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/