freebsd-skq/share/man/man4/ccd.4

288 lines
8.2 KiB
Groff

.\" $NetBSD: ccd.4,v 1.5 1995/10/09 06:09:09 thorpej Exp $
.\"
.\" Copyright (c) 1994 Jason Downs.
.\" Copyright (c) 1994, 1995 Jason R. Thorpe.
.\" All rights reserved.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\" notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\" notice, this list of conditions and the following disclaimer in the
.\" documentation and/or other materials provided with the distribution.
.\" 3. All advertising materials mentioning features or use of this software
.\" must display the following acknowledgement:
.\" This product includes software developed for the NetBSD Project
.\" by Jason Downs and Jason R. Thorpe.
.\" 4. Neither the name of the author nor the names of its contributors
.\" may be used to endorse or promote products derived from this software
.\" without specific prior written permission.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
.\" IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
.\" OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
.\" IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
.\" INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
.\" BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
.\" LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED
.\" AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
.\" OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
.\" SUCH DAMAGE.
.\"
.\" $FreeBSD$
.\"
.Dd August 9, 1995
.Dt CCD 4
.Os
.Sh NAME
.Nm ccd
.Nd Concatenated Disk driver
.Sh SYNOPSIS
.Cd "device ccd"
.Sh DESCRIPTION
The
.Nm
driver provides the capability of combining one or more disks/partitions
into one virtual disk.
.Pp
This document assumes that you are familiar with how to generate kernels,
how to properly configure disks and devices in a kernel
configuration file, and how to partition disks.
.Pp
In order to compile in support for the
.Nm ,
you must add a line similar
to the following to your kernel configuration file:
.Pp
.Dl "device ccd # concatenated disk devices"
.Pp
As of the
.Fx 3.0
release, you do not need to
configure your kernel with
.Nm
but may instead use it as a kernel loadable
module.
Simply running
.Xr ccdconfig 8
will load the module into the kernel.
.Pp
A
.Nm
may be either serially concatenated or interleaved.
To serially
concatenate the partitions, specify the interleave factor of 0.
Note that mirroring may not be used with an interleave factor of 0.
.Pp
There is a run-time utility that is used for configuring
.Nm Ns s .
See
.Xr ccdconfig 8
for more information.
.Ss The Interleave Factor
If a
.Nm
is interleaved correctly, a
.Dq striping
effect is achieved, which can increase sequential read/write
performance.
The interleave factor is expressed in units of
.Dv DEV_BSIZE
(usually 512 bytes).
For large writes, the optimum interleave factor
is typically the size of a track, while for large reads, it is about a
quarter of a track.
(Note that this changes greatly depending on the
number and speed of disks.)
For instance, with eight 7,200 RPM drives
on two Fast-Wide SCSI buses, this translates to about 128 for writes
and 32 for reads.
A larger interleave tends to work better when the
disk is taking a multitasking load by localizing the file I/O from
any given process onto a single disk.
You lose sequential performance when
you do this, but sequential performance is not usually an issue with a
multitasking load.
.Pp
An interleave factor must be specified when using a mirroring configuration,
even when you have only two disks (i.e., the layout winds up being the same
no matter what the interleave factor).
The interleave factor will determine
how I/O is broken up, however, and a value 128 or greater is recommended.
.Pp
.Nm
has an option for a parity disk, but does not currently implement it.
.Pp
The best performance is achieved if all component disks have the same
geometry and size.
Optimum striping cannot occur with different
disk types.
.Pp
For random-access oriented workloads, such as news servers, a larger
interleave factor (e.g., 65,536) is more desirable.
Note that there
is not much
.Nm
can do to speed up applications that are seek-time limited.
Larger
interleave factors will at least reduce the chance of having to seek
two disk-heads to read one directory or a file.
.Ss Disk Mirroring
You can configure the
.Nm
to
.Dq mirror
any even number of disks.
See
.Xr ccdconfig 8
for how to specify the necessary flags.
For example, if you have a
.Nm
configuration specifying four disks, the first two disks will be mirrored with
the second two disks.
A write will be run to both sides of
the mirror.
A read will be run to either side of the mirror depending
on what the driver believes to be most optimal.
If the read fails,
the driver will automatically attempt to read the same sector from the
other side of the mirror.
Currently
.Nm
uses a dual seek zone model to optimize reads for a multi-tasking load
rather than a sequential load.
.Pp
In an event of a disk
failure, you can use
.Xr dd 1
to recover the failed disk.
.Pp
Note that a one-disk
.Nm
is not the same as the original partition.
In particular, this means
if you have a file system on a two-disk mirrored
.Nm
and one of the disks fail, you cannot mount and use the remaining
partition as itself; you have to configure it as a one-disk
.Nm .
You cannot replace a disk in a mirrored
.Nm
partition without first backing up the partition, then replacing the disk,
then restoring the partition.
.Ss Linux Compatibility
The
.Tn Linux
compatibility mode does not try to read the label that
.Tn Linux Ns '
.Xr md 4
driver leaves on the raw devices.
You will have to give the order
of devices and the interleave factor on your own.
When in
.Tn Linux
compatibility mode,
.Nm
will convert the interleave factor from
.Tn Linux
terminology.
That means you give the same interleave factor that you
gave as chunk size in
.Tn Linux .
.Pp
If you have a
.Tn Linux
.Xr md 4
device in
.Dq legacy
mode, do not use the
.Dv CCDF_LINUX
flag in
.Xr ccdconfig 8 .
Use the
.Dv CCDF_NO_OFFSET
flag instead.
In that case you have to convert
the interleave factor on your own, usually it is
.Tn Linux Ns '
chunk size multiplied by two.
.Pp
Using a
.Tn Linux
RAID this way is potentially dangerous and can destroy
the data in there.
Since
.Fx
does not read the label used by
.Tn Linux ,
changes in
.Tn Linux
might invalidate the compatibility layer.
.Pp
However, using this is reasonably safe if you test the compatibility
before mounting a RAID read-write for the first time.
Just using
.Xr ccdconfig 8
without mounting does not write anything to the
.Tn Linux
RAID.
Then you do a
.Nm fsck.ext2fs Pq Pa ports/sysutils/e2fsprogs
on the
.Nm
device using the
.Fl n
flag.
You can mount the file system read-only to check files in there.
If all this works, it is unlikely that there is a problem with
.Nm .
Keep in mind that even when the
.Tn Linux
compatibility mode in
.Nm
is working correctly, bugs in
.Fx Ap s
.Nm ex2fs
implementation would still destroy
your data.
.Sh WARNINGS
If just one (or more) of the disks in a
.Nm
fails, the entire
file system will be lost unless you are mirroring the disks.
.Pp
If one of the disks in a mirror is lost, you should still
be able to back up your data.
If a write error occurs, however, data
read from that sector may be non-deterministic.
It may return the data
prior to the write or it may return the data that was written.
When a
write error occurs, you should recover and regenerate the data as soon
as possible.
.Pp
Changing the interleave or other parameters for a
.Nm
disk usually destroys whatever data previously existed on that disk.
.Sh FILES
.Bl -tag -width ".Pa /dev/ccd*"
.It Pa /dev/ccd*
.Nm
device special files
.El
.Sh SEE ALSO
.Xr dd 1 ,
.Xr ccdconfig 8 ,
.Xr config 8 ,
.Xr disklabel 8 ,
.Xr fsck 8 ,
.Xr gvinum 8 ,
.Xr mount 8 ,
.Xr newfs 8
.Sh HISTORY
The concatenated disk driver was originally written at the University of
Utah.