| Infodoc ID |
|
Synopsis |
|
Date |
| 14622 |
|
Advantages of Splitting a Drive into Multiple File Systems |
|
20 Oct 1999 |
Rather than using an entire disk drive for one file system, which may
lead to inefficiencies and other problems, you can
split a single drive into sections. The sections are called slices, as
each is a slice of the disk's capacity. Once the partition has been
allocated, it becomes the a logical disk drive.
A disk can be split into eight subdisks. The splitting of the disk is
often called partitioning or labeling of the disk drive. Below is an
example:
Current partition table (original):
Total disk cylinders available: 2036 + 2 (reserved cylinders)
Part Tag Flag Cylinders Size Blocks
0 root wm 0 - 1872 921.87MB (1873/0/0) 1887984
1 unassigned wm 0 0 (0/0/0) 0
2 backup wm 0 - 2035 1002.09MB (2036/0/0) 2052288
3 unassigned wm 1873 - 2035 80.23MB (163/0/0) 164304
4 unassigned wm 0 0 (0/0/0) 0
5 unassigned wm 0 0 (0/0/0) 0
6 unassigned wm 0 0 (0/0/0) 0
7 unassigned wm 0 0 (0/0/0) 0
partition>
Here are some of the reasons for multiple filesystems on one hard
drive.
1. Damage Control: If the system were to crash due to software error,
hardware failure, or power problems, some of the
disk blocks might still be in the file system
cache and not have been written to disk yet. This
can cause damage to the filesystem structure.
While the methods used try to reduce this damage,
and the FSCK utility can repair most of the
damage, spreading the files across multiple
filesystems minimizes the possibility of damage,
especially to those files that are needed during
boot-up. When the files are split up across
the disk slices, critical files end up on slices
that rarely change or are mounted read-only and
never change. The chances of them being damaged
and preventing you from recovering the remainder
of the system are greatly reduced.
2. Access Control: Only complete slices can be marked as read-only or
read-write.
If you desire to mount the shared Operating System
sections as read-only to prevent changes, they
have to be on their own slice.
3. Space Management: Files are used from a reserve of free space on a
per-file system basis.
If, for example, a user has allocated a large
amount of space, depleting the free space, and the
entire system disk were a single filesystem, there
would be no free space left for critical system
files. The entire system would freeze when it ran
out of space.
Using separate filesystems, especially for user
files, allows only that a single user, or group of
users, to be inconvenienced when filesystem
becomes full. The system will continue to operate,
allowing the System Administrator to handle the
problem. The exception to the above scenario is
the root filesystem.
4. Performance: The larger the filesystem, the larger the tables
that must be managed.
As the disk fragments and space become
scarce, the further apart the fragments of a file
might be placed on the disk. Using multiple
(smaller) partitions reduces the absolute distance
and keeps the sizes of the tables manageable.
Although the UFS file filesystem does not suffer
from table size an fragmentation problems as much
as System V file systems, this is still a
concern.
5. Backups: Many of the back-up utilities, such as "ufsdump"
work on a complete filesystem basis. If a
filesystem is large, it could take longer than you
want to allocate to back-up. Most importantly,
multiple smaller backups are easier to handle and
recover from.
Below is a listing of slices, some that are required, root and swap,
and the recommended additional slices such as usr, var, opt, home and
tmp.
1. The root slice: The root slice is mounted at the top of the
filesystem hierarchy. It is mounted automatically
as the system boots, and cannot be unmounted. All
other file systems are mounted below the root.
The root filesystem needs to be large enough to
hold the following:
* The boot information and the bootable kernel
(kernel/genunix), and a backup
of the kernel just in case the main one gets damaged.
* Any local system configuration files, which
typically reside in the /etc directory.
* Any stand-alone programs, such as diagnostics,
that may be run instead
of the OS.
The root partition typically runs on between 15 and
30mb. It is usually placed on the first slice of
the disk, or more commonly know as slice 0 or a.
2. The swap slice: The default rule is that there is twice as much swap
space as there is RAM installed on the system. For
example, if you have 16mb of ram, the swap space
would need to be 32mb. Although this is just a
preliminary template as to how much swap to use,
their are other factors to consider, an example
would be if a users system is running large
applications that use large amounts of data, such
as a CAD application. You can monitor the
amount of swap space used via the pstat or swap
commands. If you did not allow enough swap
space during the initial install you can add
additional swap with either the swapon or swap
commands.
3. The usr slice: The usr slice holds the remainder of the operating
system utilities. It needs to be large enough to
hold all the packages you chose to install when
installing the OS. If you are going to install
local applications or third-party applications in
this slice, it needs to be large enough to hold
them. It is generally better if the usr slice
contains the operating system and only symbolic
links to the applications. The filesystem is often
mounted read-only to prevent changes.
4. The var slice: The var slice holds the spool directories used to
queue printer files and mail, as well as log files
that my be unique to the system. It also holds the
/var/tmp directory, which is used for larger
temporary files. It is the read-write counterpart
to the usr slice. Every system, even a diskless
client, needs it's own var filesystem. It is not a
filesystem that can be shared with any other
system(s).
5. The opt slice: In the newer UNIX systems based on System V release
4 (Solaris 2.x) many sections are now optional and
no longer needed to be loaded on the /usr
filesystem. They are now installed onto the /opt
filesystem. Additional add on packages are also
installed in this filesystem.
6. The home slice: The home directory is where the user's login
directories are placed. Making home its own slice
prevents users from hurting anything else if they
run this filesystem out of space. A good starting
point for the size of this slice is 1mb per
application user plus 5mb per power user and 10mb
per developer you intend to support. These are
rough estimates and are to be only used as a
guideline, you configuration may need more or less
space per user. Usually this is /export/home. Don't
put things into /home, as this is a reserved mount
point for automounted NFS filesystems. It's fine to
use when automounter is turned off, but it is on by
default.
7. The tmp slice: Large temporary files are placed in the /var/tmp
but sufficient temporary files are placed in /tmp.
The files in the /tmp directory are very
short-lived and are cleared out during a reboot of
the system. If users run mostly application based
programs 5 to 10mb should be sufficient for this
slice. If developers are the primary users of the
system 10 to 20mb may be needed. Once again these
numbers or only a guideline, your needs may be
different.
Top
Sun Proprietary/Confidential: Internal Use Only
Feedback to SunSolve Team
