322 Implementing the IBM General Parallel File System (GPFS) in a Cross-Platform Environment
Delete a storage pool
The following considerations apply when you delete a storage pool:
User Storage pools can be deleted by deleting the last disk in the storage pool; system
storage pool cannot be deleted.
Only one storage pool can be deleted at a time with the mmdeldisk command.
Pools containing data cannot be deleted. You must either delete the data or migrate data
to separate pool before deleting the disk.
To delete system storage pool you must delete the entire file system. However, if you have
multiple disks in the system storage pool, they can be deleted if the remaining disks have
enough space to hold existing metadata.
Working with files in storage pools
To change a file storage pool assignment, consider the following information:
root user may change a file’s assigned storage pool by issuing the mmchattr -P
The default method is to migrate the data immediately (use the -I yes parameter).
If the parameter -I defer is specified, the command does not move the existing data to
the new storage pool immediately and is deferred until a later call to either the
mmrestripefs or mmrestripefile command.
To rebalance or to restore files in a storage pool, consider the following information:
root user may rebalance or restore all files in a file system by issuing the
Specify the -P option (uppercase P), to repair only files that are assigned to the specified
Specify the -p option (lowercase p), to repair the file placement within the storage pool.
Alternately, you may issue the mmrestripefile -p command to repair the placement of
specific files within the storage pool.
Files that are assigned to one storage pool, but with data in a separate pool, will have their
data migrated to the correct pool.
8.1.3 File sets
A file set is a subtree of a file system namespace that in many respects behaves like an
independent file system. File sets provide a means of partitioning the file system to allow
administrative operations at a finer granularity than the entire file system.
Figure 8-2 on page 323 shows an example of several file sets, as follows:
The Fileset_u1 starts from /dir4/dir_b directory and has /user1/ as the parent
The Fileset_u2, which starts at /user2 directory and has as its children the data1, data2,
data3, and data4 directories.
Note: To use a user-defined storage pool, a placement policy has to be defined. By
default, all files are written in the system storage pool, and when it is filled, an
error occurs even if the user-defined storage pools are empty.
Chapter 8. Information lifecycle management (ILM) 323
Figure 8-2 File sets
Consider the following information about file sets:
File sets can be used to define quotas on both data blocks and inodes.
The owning file set is an attribute of each file and can be specified in a policy to control the
– Initial data placement
– Replication of the file’s data
Consider the following information about file sets implementation:
– When the file system is first created, only one file set exists, which is called the root file
set. The root file set contains the root directory and any system files, such as quota
– As new files and directories are created, they automatically become part of the parent
directory’s file set.
– The file set to which a file belongs is largely transparent for ordinary file access.
However, to display the containing file set and the other attributes of each file, use the
mmlsattr -L command
– When upgrading an existing file system to the latest version of GPFS that supports file
sets, all existing files and directories are assigned to the root file set.
Consider the following information about file set namespaces:
– A newly created file set consists of an empty directory for the root of the file set, and it
is initially not linked into the file system's namespace.
– A newly created file set is not visible to the user until it is attached to the namespace by
issuing the mmlinkfileset command.
– File sets are attached to the namespace with a special link called a
– Only one junction is allowed per file set, so that a file set has a unique position in the
namespace and a unique path to any of its directories.