mistakes in the doc

http://smartmontools.sourceforge.net/badblockhowto.html

Second Step: we need to find the block size of the file system
(normally 4096 bytes for ext2):

root]# tune2fs -l /dev/hda3 | grep Block
Block count: 29119820
Block size: 4096

Note: the fractional part of 0.125 indicates that this problem
LBA is actually the second of the eight sectors that make up
this file system block.

Fourth Step: we use debugfs to locate the inode stored in this
block, and the file that contains that inode:

Obviously missing Third step. And conversion of
first sector from hexa to decimal, and then conversion into FS
block number.

Now everything is back to normal: the sector
has been reallocated. Compare the output just below to similar
output near the top of this article:

There is no similar block at top of article.

***

The page lacks a few things:

- consideration about RAID (note: RAID1 implementation by mdadm
varies with version: raid1 by 0.9 is not at all the same as for 1.2;
1.x versions can be migrated to raid5; I don't know if it's possible
with 0.9 - gotta test it some day); equations on how to transfert a
disk-sector offset into a filesystem sector

- consideration about crypto: which crypto layers keep sectors in
place, and which ones don't

- easily perform manual test at user level using dd if=/dev/urandom
of=/mnt/disk/foo untill the disk is full (take care; the system may
dislike a full disk), then read the file again, to perform write
then read test on whole the free space, without using dangerous
tools like badblocks. I sometimes use dychotomy with this: first
file uses only 99% of free space; then a second file for 99% of
remaining space. Read second small file, then delete it at once;
this reduces the chance of full disk on system critical partitions
(like /tmp or /var). dd can easily perform a raw disk read; but non
destructive write test is more tricky. Doing this test twice is a
good indicator; a disk can be damaged in a way that ECC works fine;
but considering that a double urandom is unlikely to produce twice
the same sequence at the same place, a bad sector will probably be
detected with this method. This lacks write test where files are;
this test is not perfect; it's an easy alternative to check live
disk sanity without destroying the FS.

- considerations about hidden FS.

--
>o_/ DEMAINE Benoît-Pierre (aka DoubleHP) http://benoit.demaine.info/
If computing were an exact science, IT engineers would'nt have work \_o<

"So all that's left, Is the proof that love's not only blind but deaf."
(FAKE TALES OF SAN FRANCISCO, Arctic Monkeys)

------------------------------------------------------------------------------
Time is money. Stop wasting it! Get your web API in 5 minutes.
www.restlet.com/download
http://p.sf.net/sfu/restlet

_______________________________________________
Smartmontools-support mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/smartmontools-support