e2fsck aborts when invalid indirect block is encountered
Hi, All.
During a recent read-only checking of an corrupted ext3 file system, I found a strange behaviour of e2fsck: when an inode has an invalid indirect block number, e2fsck aborts with the following message:
e2fsck 1.39 (29-May-2006)
Pass 1: Checking inodes, blocks, and sizes
Inode 12 has illegal block(s). Clear? no
Illegal block #-1 (4294967295) in inode 12. IGNORED.
Error while iterating over blocks in inode 12: Illegal indirect block found
e2fsck: aborted
You can reproduce it with this code snippet:
#!/bin/sh
dev=/dev/sde
mnt=/mnt
mkfs.ext3 -F $dev
mount $dev $mnt
dd if=/dev/zero of=$mnt/file bs=1M count=1 umount $dev
debugfs -w -R 'sif file block[IND] 0xFFFFFFFF' $dev e2fsck -f -n $dev
Doing a fixing without -n option can safely delete this bad blocknum.
My question is: Is this behaviour a bug or intended?
Quoted 5 lines · Chen Huan
During a recent read-only checking of an corrupted ext3 file system, I found a strange behaviour of e2fsck: when an inode has an invalid indirect block number, e2fsck aborts with the following message:
e2fsck 1.39 (29-May-2006)
Please retest with a new version of e2fsprogs. The current release version is 1.41.14, and the work-in-progress for version 1.42 is available via Git.
Quoted 21 lines
Pass 1: Checking inodes, blocks, and sizes
Inode 12 has illegal block(s). Clear? noIllegal block #-1 (4294967295) in inode 12. IGNORED.
Error while iterating over blocks in inode 12: Illegal indirect block found
e2fsck: abortedYou can reproduce it with this code snippet:
#!/bin/sh
dev=/dev/sde
mnt=/mntmkfs.ext3 -F $dev
mount $dev $mnt
dd if=/dev/zero of=$mnt/file bs=1M count=1 umount $dev
debugfs -w -R 'sif file block[IND]0xFFFFFFFF' $dev e2fsck -f -n $dev
Quoted 7 lines
Doing a fixing without -n option can safely delete this bad blocknum.
My question is: Is this behaviour a bug or intended?
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Cheers, Andreas
Quoted 10 lines · Andreas Dilger
Quoted 5 lines · Chen Huan
During a recent read-only checking of an corrupted ext3 file system, I found a strange behaviour of e2fsck: when an inode has an invalid indirect block number, e2fsck aborts with the following message:
e2fsck 1.39 (29-May-2006)Please retest with a new version of e2fsprogs. The current release version is 1.41.14, and the work-in-progress for version 1.42 is available via Git.
The problem remains in e2fsck 1.42-WIP (02-Jul-2011)
Quoted 37 lines
Quoted 22 lines
Pass 1: Checking inodes, blocks, and sizes
Inode 12 has illegal block(s). Clear? noIllegal block #-1 (4294967295) in inode 12. IGNORED.
Error while iterating over blocks in inode 12: Illegal indirect block found
e2fsck: abortedYou can reproduce it with this code snippet:
#!/bin/sh
dev=/dev/sde
mnt=/mntmkfs.ext3 -F $dev
mount $dev $mnt
dd if=/dev/zero of=$mnt/file bs=1M count=1 umount $dev
debugfs -w -R 'sif file block[IND]0xFFFFFFFF' $dev e2fsck -f -n $devQuoted 7 lines
Doing a fixing without -n option can safely delete this bad blocknum.
My question is: Is this behaviour a bug or intended?
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.htmlCheers, Andreas
Quoted 10 lines · Chen Huan
Hi, All.
During a recent read-only checking of an corrupted ext3 file system, I found a strange behaviour of e2fsck: when an inode has an invalid indirect block number, e2fsck aborts with the following message:
e2fsck 1.39 (29-May-2006)
Pass 1: Checking inodes, blocks, and sizes
Inode 12 has illegal block(s). Clear? no
Quoted 5 lines
Illegal block #-1 (4294967295) in inode 12. IGNORED.
Error while iterating over blocks in inode 12: Illegal indirect block found
e2fsck: abortedMy question is: Is this behaviour a bug or intended?
This is deliberate. There are two reasons for this:
(1) e2fsck -n is currently defined as being equivalent to (a) opening the file system read-only, and (b) answering 'no' to all questions asked by e2fsck.
(2) In the case where you *aren't* doing an e2fsck -n, there are certain situations where not correcting an error means (a) that its dangerous to try to fix anything afterwards, and that (b) many of the diagnostics issued by e2fsck can not be relied upon.
We could change things so that e2fsck doesn't abort in the case where -n is specified, which would avoid the problem 2a, but it breaks the definition of 1b. It also doesn't solve the problem listed in 2b. Still, if it's really annoying it's something I could consider changing. Let me sleep on it.
- Ted