This section describes special actions you must
take when working with file systems inside logical volumes. It addresses
the following topics:
|TIP: When dealing with file systems, you can use HP
SMH or a sequence of HP-UX commands. For most tasks, using HP SMH
is quicker and simpler. You do not have to explicitly perform each
of the following distinct tasks; rather, proceed from the HP SMH Disk
and File Systems area. HP SMH performs all the necessary steps for
Creating a File System
When creating either an HFS or VxFS file system
in a logical volume, you can use HP SMH or a sequence of HP-UX commands.
If you choose to use HP-UX commands directly, the following list describes
the subtasks for creating a file system.
create a new file system of a type other than HFS, you might need
to reconfigure the new type into the kernel. Normally, VxFS has been
configured into the kernel as part of the default configuration. See HP-UX System Administrator's Guide: Configuration Management for information on how to add a file system type.
Estimate the size required
for the logical volume. To estimate the size requirements for a logical
volume containing a file system, see “Setting Up Logical Volumes for File Systems”.
if there is sufficient disk space available in the volume group. Use
the vgdisplay command to calculate this information,
as described in “Creating a Logical Volume”.
If there is not enough space within
the volume group, you can add a disk to the volume group, as described
in “Adding a Disk to a Volume Group”.
the logical volume. Use lvcreate, as described
in “Creating a Logical Volume”.
Create the file system using the character device file.
# newfs -F fstype /dev/vg02/rlvol1
If you do not use the -F fstype option, then newfs creates a file system
based on the content of your /etc/fstab file.
If there is no entry for the file system in /etc/fstab, then the file system type is determined from the file /etc/default/fs. For information on additional options,
When creating a VxFS file system, file names will
be long automatically.
For HFS, use the -s or -l option to specify a file system with short or long file
names, respectively. By default, the length of file system names are
consistent with those of the root file system. Short file names are
14 characters maximum. Long file names allow up to 255 characters.
HP recommends using long file names for flexibility; files created
on other systems that use long file names can be moved to your system
without being renamed.
After you have created a file
system, mount it for users to access it, and add it to /etc/fstab so that it is automatically mounted at boot time.
Extending a File System
Extending a file system inside a logical volume
is a two-step task: extending the logical volume, then extending the
file system. The first step is described in “Extending a Logical Volume”. The second step, extending the
file system itself, depends on the following factors:
What type of file system
is involved? If it is HFS or VxFS? HFS requires the file system to
be unmounted to be extended.
Check the type
of file system. For example:
# /usr/sbin/fstyp /dev/vg01/lvol2
If the file system is
VxFS, do you have the base VxFS product or the OnlineJFS product?
If you have only the base VxFS product, you must unmount the file
system before extending it.
To see if the OnlineJFS product is
installed, enter the following command:
# swlist -l product | grep -i OnlineJFS
OnlineJFS B.11.31 Online features of the VxFS File System
Can you unmount the file
system? To unmount system directories such as/var and /usr, you must be in single-user mode.
Is the file system the
root file system (/)? If so, there are two complications:
The logical volume containing
the root file system is created with the contiguous allocation policy,
so it might not be possible to extend it in place.
The root file system cannot
ever be unmounted, even if you change to single-user state.
If you are using VxFS as your root file system
and have the OnlineJFS product, you can extend the original root file
system without unmounting, provided contiguous disk space is available.
Otherwise, to extend the current root file system,
you must create and mount another root disk which
enables you to work with the unmounted original root disk, extending
it if contiguous disk space is still available.
If the original disk does not have contiguous disk space available,
instead of expanding the original root disk, you can create a new
root file system on another larger disk.
|CAUTION: This procedure can fail on a VxFS file system already
at 100% capacity (Error 28). You must remove some files before you
attempt this operation.|
Once you have the answers to these questions,
follow these steos:
If the file system must be
unmounted, unmount it.
Be sure no one has files open
in any file system mounted to this logical volume and that it is no
user's current working directory. For example:
# fuser -cu /work/project5
If the logical volume is in use, confirm that
the underlying applications no longer need it. If necessary, stop
|NOTE: If the file system is exported using NFS
to other systems, verify that no one is using those other systems,
then unmount it on those systems.|
If you cannot stop the applications
using the logical volume, or it is a system directory such as /var or/usr, change to single-user
state as follows:
the file system as follows:
# /sbin/umount /dev/vg01/lvol2
the logical volume. For example:
# /sbin/lvextend -L 332 /dev/vg01/lvol2
This increases the size of this volume to 332 MB.
the file system size to the logical volume size. If the file system
is unmounted, use the extendfs command as follows:
# /sbin/extendfs /dev/vg01/rlvol2
If you did not have to unmount the file system,
use the fsadm command instead. The new size is
specified in terms of the block size of the file system. In this example,
the block size of the file system /work/project5 is 1 KB. To extend the file system to 332 MB, the number of blocks
is 339968 (332 times 1024). For example:
# fsadm -b 339968 /work/project5
unmounted the file system, mount it again.
If you had to change to single-user
state, reboot the system.
You can skip any additional steps to mount the
file system and export it, since the boot process mounts and exports
any file systems.
Remount the file system as
# /sbin/mount /dev/vg01/rlvol2 /mount_point
|NOTE: If the file system will continue to be used by
NFS clients, export it on the server (exportfs -a) and remount it on the clients (mount -a).|
Verify that the file system
reflects the expansion by entering bdf, df, or fsadm -E.
Reducing the Size of a File System
You might want to shrink a file system that has
been allocated more disk space than it needs, allowing that disk space
to be freed for other uses.
Reducing the size of a file system is more complicated
than extending it. Because of file system block allocation strategies,
data can be scattered throughout the logical volume. Reducing the
logical volume reclaims space at the end of the logical volume, requiring
file system drivers to coalesce and rearrange data blocks ahead of
time. Most types of file system are unable to do such coalescence,
so you must back up the data in the file system, reduce the logical
volume, create a new file system in the smaller logical volume, and
restore the data from your backup.
The only current file system type able to do online
coalescence and size reduction is OnlineJFS, and it can fail in some
Reducing a File System Created with OnlineJFS
Using the fsadm command shrinks
the file system, provided the blocks it attempts to deallocate are
not currently in use; otherwise, it fails. If sufficient free space
is currently unavailable, file system defragmentation of both directories
and extents might consolidate free space toward the end of the file
system, allowing the contraction process to succeed when retried.
For example, suppose your VxFS file system is currently
6 GB. However, you decide you only need 2 GB with an additional 1
GB for reserve space. Thus, you want to resize the file system to
3 GB. Use fsadm with the -b option
to specify the new size of the file system in sectors, then reduce
the size of the logical volume to match. Assuming the file system
sector size is 1K, use the following commands:
# fsadm -b 3145728 /home
# lvreduce -L 3072 /dev/vg01/lvol5
Reducing a File System Created with HFS
To reduce the size of a file system created with
HFS or VxFS, follow these steps:
sure no one has files open in any file system on the logical volume
and that the logical volume is no user's current working directory.
# fuser -cu /dev/vg01/lvol5
|NOTE: If the file system is exported using NFS to other
systems, verify that no one is using those other systems, then unmount
the file system on those systems before unmounting it on the server.|
up the data in the logical volume.
to back up /work/project5 to the system default
tape device, enter the following command:
the data in the file system the logical volume is mounted to, as follows:
Because /work/project5 is
a mount point, rm -r does not remove the directory
the file system to which the logical volume is mounted as follows:
the size of the logical volume as follows:
# lvreduce -L 500 /dev/vg01/lvol5
This command reduces the logical volume/dev/vg01/lvol5 to 500 MB.
Create a new file system in
the reduced logical volume, using the character device file. For example:
# newfs -f fstype /dev/vg01/rlvol5
the logical volume as follows:
# mount /dev/vg01/lvol5 /work/project5
the data from the backup. For example:
This recovers all the contents of a tape in the
system default drive.
If /work/project5 continues to be used by NFS clients, re-export
it on the server (exportfs -a) and remount it on
the clients (mount -a).
Backing Up a VxFS Snapshot File System
|NOTE: Creating and backing up a VxFS snapshot file system
requires that you have the optional HP OnlineJFS product installed
on your system. For more information, see HP-UX System
Administrator's Guide: Configuration Management.|
you to perform backups without taking the file system offline by making
a snapshot of the file system, a read-only image of the file system
at a moment in time. The primary file system remains online and continues
to change. After you create the snapshot, you can back it up with
any backup utility except the dump command.
To create and back up a VxFS snapshot file system,
follow these steps:
how large the snapshot file system must be, and create a logical volume
to contain it. Use bdf to assess the primary file
system size and consider the following:
Block size of the file system
(1,024 bytes per block by default)
How much the data in this
file system is likely to change (HP recommends 15 to 20% of total
file system size)
For example, to determine how large to make a snapshot
of lvol4, mounted on the /home directory, examine its bdf output:
# bdf /home
filesystem kbytes used avail %used mounted on
/dev/vg00/lvol4 40960 38121 2400 94% /home
Allowing for 20% change to this 40 MB file system,
create a logical volume of eight blocks (8 MB).
Use lvcreate to create a logical volume to contain the snapshot file system.
# lvcreate -L 8 -n snap /dev/vg02
This creates an 8 MB logical volume called /dev/vg02/snap, which can contain a snapshot file system
Create a directory for the
mount point of the snapshot file system. For example:
Create and mount the snapshot
file system. For example:
# mount -f vxfs -o snapof=/dev/vg00/lvol4 /dev/vg02/snap /tmp/house
In this example, a snapshot is taken of logical
volume /dev/vg00/lvol4, contained in logical
volume /dev/vg02/snap, and mounted on /tmp/house.
Back up the snapshot file system
with any backup utility except dump.
For example, use tar to archive the
snapshot file system /tmp/house, ensuring that
the files on the tape have relative path names, as follows:
# cd tmp; tar cf /dev/rtape/tape0BEST house
Alternatively, the following vxdump command backs up a snapshot file system /tmp/house, which has extent attributes:
# vxdump -0 -f /dev/rtape/tape0BEST /tmp/house