Re: [edlug] Failing hdd and mystery crash

[Mike Moran <mike_moran@xxx.xxx.xxx>]

Alistair K Phipps wrote:
> On Fri, 5 Mar 2004 09:35:17 +0000
> Mike Moran <mike_moran@xxx.xxx.xxx> wrote:
>
>
> > into that damn inode :-) So, is there some option I can pass to one of 
> > the ext2 family of programs which will allow me to check that the 
> > badblocks inode has been updated? Also, is the "bad magic" message 
> > given by e2dump (as mentioned previously) when I try to use it to dump 
> > the number 2 inode something to worry about?
>
>
> dumpe2fs -b /dev/hdXn
>
> Where /dev/hdXn is the ext2/3 partition, should output the list of bad
> blocks on that partition.

Right. Thanks very much for that.

> Bad magic usually means that the device you specified does not contain
> a ext2/3 filesystem. 

I think this was because I wasn't using dumpe2fs (running this instead 
produced no "Bad Magic").

Well, it seems I'm getting to the bottom of the problem (if any). If I 
check one partition (hda9 in this case), without specifying a block 
size, then I get 3 bad blocks. However, if I get dump2fs to tell me the 
block size, and pass this to badblocks, then I get no bad blocks 
detected. So, it would seem the block size is important; meaning I was 
probably just too stupid to think of specifying it before.

One final thing: when I use ide-smart (vesrion 1.4) on hda, I get all 
passes. However, when I use smartctl (version 2.1, using "smartctl -a 
/dev/hda9") I get the following near the end:

Smart Error Log Read failed: Input/output error
Smartctl: Smart Errorlog Read Failed
Smart Error Log Read failed: Input/output error
Smartctl: Smart Self Test log Read Failed

I also get a "hda: drive_cmd:" error logged in /var/log/messages, like I 
have seen before. This means I was possibly mistaken before and it was 
the checking of the disk with smartctl which was actually *causing* the 
hda errors. This does leave me wondering why smartctl is causing these 
errors. Is it a smartctl error, or something wrong with the disk?

Anyway, in summary, thanks for all the help; I'm reasonably happy for 
now that the disk is ok and it's my assumptions that led me up the 
garden path or I'm just going to have to wait for it to fail before I 
definitely know. As HAL wisely said, "It can only be attributable to 
human error."

--
Mike

-
----------------------------------------------------------------------
You can find the EdLUG mailing list FAQ list at:
http://www.edlug.org.uk/list_faq.html