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

Re: [NFS] Re: Is ext3 kernel nfsd friendly?



On Monday January 29, radtke conectiva com wrote:
> Andreas Dilger wrote:
> 
> > I've been using knfsd on ext3 (2.2.17 kernel) for quite some time without
> > problems.  However, this is the older NFS code, and not the major changes
> > that went into NFS for 2.2.18, so YMMV.
> 
> I'm using NFS with 2.2.19pre7 + ext3 without any problems yet, quite stable.

Ofcourse, "apparently stable" is quite different from "absolutely
reliable"...

I haven't looked at ext3 or played with it but in order for a
filesystem to work with knfsd (in it's current form) it must:

  Support the read_inode super_operation and 
   1/ recognise if the inode number refers to a currently active inode
       or not
   2a/ if not, must return an is_bad_inode() inode.
   2b/ if so must  provide a completely filled-in inode (nfs doesn't for example.
                ext2 does, ext3 almost certainly does).

   3/ The "delete_inode" super_operation must cope gracefully with
      being given an is_bad_inode.

   4/ Should keep a "generation" number in the inode-on-disk, should
      assigned a "new" value to this when the inode is allocated, and
      should store this value in the i_generation field of the "struct
      inode".


Before I started looking at knfsd, ext2 did not do 1 (or 2a) or 3, and
I think there was something funny with 4, but I'm not sure.
If ext3 does all these, it should work fine.  If not, I suspect they
would be fairly straight forward to add.

NeilBrown





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