Re: another seriously corrupt ext3 -- pesky journal

Re: another seriously corrupt ext3 -- pesky journal

[Mike Fedyk]

On Mon, Aug 18, 2003 at 02:13:54PM -0400, Theodore Ts'o wrote:
> On Mon, Aug 18, 2003 at 12:39:46PM -0400, Erez Zadok wrote:
> > The power failure on Thursday did something evil to my ext3 file system (box
> > running RH9+patches, ext3, /dev/md0, raid5 driver, 400GB f/s using 3x200GB
> > IDE drives and one hot-spare).  The f/s got corrupt badly and the symptoms
> > are very similar to what Eddy described here:
> > 
> > 	https://www.redhat.com/archives/ext3-users/2003-July/msg00015.html
> > 
> > That is, nearly everything I try results in and error such as
> > 
> > 	"Invalid argument while checking ext3 journal for /dev/md0"
> 
> What probably happened is that the power failed while you were writing
> out the inode table, and the memory failed before the DMA engine and
> hard drive did, since DRAM tends to be more sensitive to voltage drops
> that other parts of the system.  As a result, random garbage got
> scribbled all over the disk.  (Ted 's observation: PC Class hardware
> is sh*t.)
> 
> Normally, this isn't a problem, since the ext3 journal contains full
> backups of recently written data blocks.  (As opposed to filesystems
> that use soft update or logically journaled filesystems, which are
> even more fragile in the face of cheap hardware that scribble random
> garbage on power failure.)  However, this is not true when the first

After reading this comment about logical journaling (and reading a thread
about htree and ext3 journaling a while ago, and how logical journaling
could make the problems go away there), it made me think of reiserfs 3.6,
which I have been using recently over the past six months or so. 
                               
IIUC, reiserfs uses a logical journal, and it only does meta-data
journaling, but not to the level of ext3's writeback mode.          
                               
There are patches to add ext3-like data journaling to reiserfs that are
planned to be submitted for one of the next couple 2.4-pre kernels.
                               
I'd be very interested in how reiserfs (and reiserfsck!) might handle a
similar scenario. 
                               
Mike                  

Re: another seriously corrupt ext3 -- pesky journal

[Oleg Drokin]

Hello!

On Mon, Aug 18, 2003 at 01:16:09PM -0700, Mike Fedyk wrote:

> IIUC, reiserfs uses a logical journal, and it only does meta-data
> journaling, but not to the level of ext3's writeback mode.          

You probably meant ext3's jordered mode.
ext3's writeback mode is somewhat like reiserfs' default mode of operations
without extra journaling patches.

> There are patches to add ext3-like data journaling to reiserfs that are
> planned to be submitted for one of the next couple 2.4-pre kernels.

True.
Also Chris should be working on porting those to 2.6. At least I hope so.

> I'd be very interested in how reiserfs (and reiserfsck!) might handle a
> similar scenario. 

We do not have journal inode. If you get your journal header or superblock
corrupted, you need to run "reiserfsck --rebuild-sb" and answer some questions
about what was your FS look like (amount of questions depends on how much
journal header and superblock were damaged).
All other corruptions (bitmaps and stuff) should be handled by usual reiserfsck modes.
(reiserfsck --rebuild-tree, if corruption was not trivial)

Bye,
    Oleg