[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: another seriously corrupt ext3 -- pesky journal



In message <20030821202507 GD5035 ti19>, "Bill Rugolsky Jr." writes:
> On Thu, Aug 21, 2003 at 03:28:03PM -0400, Erez Zadok wrote:

> Erez,
> 
> If I understand Ted correctly, the contents of the journal inode, not the
> contents of the journal, are replicated.  In particular, there is a copy
> of the block pointers that specify where the journal is located.  The JBD
> code takes care of ensuring that the data in the journal is not garbage.
> The problem is when the journal block pointers (i.e., the journal meta-data)
> that identify where the journal lives (since it is just a file, identified
> by a list of block pointers) get corrupted.
> 
> Regards,
> 
> 	Bill Rugolsky

Ah, that makes sense.

So if I understood it, we simply reserve double the space in the superblock
to store this duplicate inode info.  This can be done by userland tools at
format time, and the kernel doesn't need to care that there's a duplicate
journal inode info on disk, b/c that inode info will never change.  Is it
correct that the journal inode info never changes? (no a/m/ctime, size
changes, etc.)?

In that case, why not put a copy of this journal inode info in every copy of
the superblock?  It's a one-time effort at mkfs time and have little to no
impact at run time.  It costs a bit more space, but that's a small price to
pay for increased reliability.

Thanks,
Erez.




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]