[Lustre-discuss] MMP: lustre-discuss at lists.lustre.org

DaMiri Young damiri at unt.edu
Thu Jun 7 14:31:09 PDT 2012

Hi All,
We are recovering from a Lustre 1.6 error. We believe this error 
resulted from
several drives failing simultaneously. We replaced the failed drives and 
down, upgraded the controller & enclosure firmware on the S2A9900.

After getting the appliance back online we are now trying to run e2fsck 
on all
20 luns to verify. 4 of the 20 luns report:
--------------------------- begin e2fsck (e2fsck 1.40.11.sun1) -----
e2fsck: MMP: invalid magic number in header while trying to open 
The superblock could not be read or does not describe a correct ext2
filesystem.  If the device is valid and it really contains an ext2
filesystem (and not swap or ufs or something else), then the superblock
is corrupt, and you might try running e2fsck with an alternate superblock:
     e2fsck -b 8193 <device>

As per another post:
it seems 'tune2fs -f -E clear-mmp /dev/device' can be used to clear the 

My questions are:
1. Is the a recommended course of action here?
2. Does this increase the chance of data loss?

I'll be happy to provide more info upon request. Please advise.

Best Regards,

More information about the lustre-discuss mailing list