[Lustre-discuss] uninit_bg

Andreas Dilger andreas.dilger at oracle.com
Tue Jun 1 09:29:00 PDT 2010


On 2010-06-01, at 08:04, Wojciech Turek wrote:
> I have a Lustre filesystem that has been created 3 years ago. I would
> like to enable uninit_bg feature on that filesystem devices.
> I found in tune2fs manual page that :
> --
> Enabling certain filesystem features may prevent the filesystem from
> being mounted by kernels which  do not  support  those  features.  In
> particular, the uninit_bg and flex_bg features are only supported by
> the ext4 filesystem.
> --
> 
> The ldiskfs of the filesystem is based on ext3 so I am not sure if it
> would be safe to enable that feature. However my other, newer Lustre
> filesystem has also ldiskfs based on ext3 file system and I can see
> that uninit_bg feature is enabled there (It had to be enabled by
> default at the filesystem creation time). So my question is, if it is
> safe to enable uninit_bg feature on already created ldiskfs (based on
> ext3) ?

This is one of the features we developed for Lustre ldiskfs that was later added upstream into ext4.  It is present in all ldiskfs modules for some years already.

Cheers, Andreas
--
Andreas Dilger
Lustre Technical Lead
Oracle Corporation Canada Inc.




More information about the lustre-discuss mailing list