echo dbfs | nohup ${ORACLE_HOME}/bin/dbfs_client dbfs@q4dbfs -o direct_io,allow_other /home/oracle/q4dbfs &

--echo后面的dbfs是dbfs的密码
--dbfs@q4dbfs是dbfs的用户和tns名
--/home/oracle/q4dbfs是mount的目录

Applies to:
Oracle Server - Enterprise Edition - Version: 11.2.0.1 to 11.2.0.3 - Release: 11.2 to 11.2
Linux x86
Linux x86-64

Description
After upgrade Linux kernel (e.g. 2.6.32-100.26.2.el5) to UEK (e.g. 2.6.32-300.10.1.el5uek) DBFS is not working.
Likelihood of Occurrence

1) OS upgrade from 2.6.32-100.26.2.el5 to: 2.6.32-300.10.1.el5uek (Linux UEK Kernel).

2) After boot the Linux server using the UEK kernel, DBFS reports the next problems:

$> nohup dbfs_client -o allow_other dbfs@EUSPRD /mnt_dbfs < ~/passwordfile.f &

$> df -k
df: `/mnt_dbfs': Resource temporarily unavailable

3) But, if the Linux box is booted again with the previous kernel release (non-UEK) then DBFS has not issues.


Possible Symptoms
1) DBFS fails after system boots using an UEK kernel.
2) DBFS development team confirmed:

There is some issue in fuse in recent UEK kernel releases (e.g. 2.6.32-300.7.2.el5uek).

It's fixed in newer kernel (v2.6.32-300.11.1) which is due for a release. 
fuse user and library rpm is also due for a release.

Workaround or Resolution
Therefore, please continue using the previous kernel release 
((boot the system with the previous kernel release (non- UEK))) for your DBFS configuration, 
until the new kernel v2.6.32-300.11.1 is released.