[Gluster-devel] GlusterFS vs xfs w/ inode64 (was: Missing files?)

[Anand Avati]

> Any light on this issue?
>
> To sum up, I need to know if GlusterFS is capable of dealing with an xfs
> filesystem mounted with the inode64 option, or if it will be shortly.
>
> Sorry but it's an urgent matter to me.
>

This is a limitation of libfuse exporting fuse nodeid as 'unsigned
long' which is 4byte wide on 32bit systems (even though the
kernel/libfuse protocol use a u64 as the datatype). This will be fixed
in 2.1 of glusterfs where we will be removing dependency on libfuse
and having an in-tree fuse protocol interpreter. You can expect pre
releases by September.

(An alternative is to use a hacked version of libfuse where you
re-typedef fuse_ino_t to unsigned long long and do the right typecast
while pulling out the nodeid from the protocol headers -- and
recompile glusterfs with the new libfuse headers)

Avati