Manually cleaning a DB2 Managed Clustered File System

我一个lab system不work了。 我想清除一下。

完全参考了：http://www-01.ibm.com/support/docview.wss?uid=swg21447939

现在的情况

=========

[root@s2 ~]# df -h /gpfs2

Filesystem            Size  Used Avail Use% Mounted on

/dev/gpfs2             11G  614M   11G   6% /gpfs2

[root@s2 ~]# mmgetstate

 Node number  Node name        GPFS state

------------------------------------------

       1      s2               active

[root@s2 ~]# mmlsnsd

 File system   Disk name    NSD servers                                    

---------------------------------------------------------------------------

 gpfs2         gpfs1nsd     (directly attached)      

开始

=====

[root@s2 ~]# mmumount all -a

Mon Mar 23 19:08:51 GMT 2015: mmumount: Unmounting file systems ...

[root@s2 ~]# mmlsnsd

 File system   Disk name    NSD servers                                    

---------------------------------------------------------------------------

 gpfs2         gpfs1nsd     (directly attached)      

[root@s2 ~]# mmdelfs gpfs2

All data on the following disks of gpfs2 will be destroyed:

    gpfs1nsd

Completed deletion of file system /dev/gpfs2.

[root@s2 ~]# mmlsnsd

 File system   Disk name    NSD servers                                    

---------------------------------------------------------------------------

 (free disk)   gpfs1nsd     (directly attached)      

[root@s2 ~]# mmshutdown -a

Mon Mar 23 19:10:39 GMT 2015: mmshutdown: Starting force unmount of GPFS file systems

Mon Mar 23 19:10:44 GMT 2015: mmshutdown: Shutting down GPFS daemons

s2:  Shutting down!

s2:  'shutdown' command about to kill process 4664

s2:  Unloading modules from /lib/modules/2.6.39-300.17.2.el6uek.x86_64/extra

s2:  Unloading module mmfs26

s2:  Unloading module mmfslinux

s2:  Unloading module tracedev

Mon Mar 23 19:10:53 GMT 2015: mmshutdown: Finished

[root@s2 ~]# mmdelnsd gpfs1nsd

mmdelnsd: Processing disk gpfs1nsd

[root@s2 ~]# mmlsnsd

mmlsnsd: No disks were found.

[root@s2 ~]# mmdelnode -a

Verifying GPFS is stopped on all affected nodes ...

mmdelnode: Propagating the cluster configuration data to all

  affected nodes.  This is an asynchronous process.

mmdelnode: Command successfully completed

[root@s2 ~]# mmlscluster

mmlscluster: This node does not belong to a GPFS cluster.

mmlscluster: Command failed.  Examine previous error messages to determine cause.