Journal abortion in ext3 and its relation with remounting

[Theodore Ts'o]

On Sun, Jan 20, 2013 at 07:39:24PM +0530, shubham wrote:
> I was looking at the code of ext3 file system and found some strange
> implementation there :
> 
> Can someone let me know the validity of below statements and with
> reasoning :
> 
> 1. I found that it might also happen that journal is aborted but not
> re-mounted
> 2. Journal gets aborted but it might be possible to mount it in
> read-write mode.
> 3. Can we write some data on the partition where journal is aborted.

Sorry, I'm not at all sure what you're asking.  Can you go into more
detail about what you're concerned about?

Since I'm guessing English is not your first language, perhaps you can
demonstrates with a series of commands which triggers what you think
could happen?  Or point at the explicit C code that for which you have
questions?

> One more question, how unlinked inode is handled by ext3 ?

Handled in what way?   Again, what are you wondering about?

	   	       	      	       	   - Ted