Description of problem:
Found while debugging a workaround that we wanted to suggest to a customer to speed up pending self-heals.

How reproducible:
Always.

Steps to Reproduce:
1.Get the list of gfids that need heal from .glusterfs/indices/xattrop

2.Do an aux mount of the volume:
glusterfs --aux-gfid-mount --volfile-id=<volanme> --volfile-server=server-I.P. /path/to/mount 

3. For each gfid, do a dummy getfattr:
getfattr -n user.no_such_attr /path/to/mount/.gfid/gfid-string

Actual results:
Background self-heal is not triggered from the mount.

Expected results:
The getxattr should trigger a background heal.