lvm2/man/lvmraid.7.in

.TH "LVMRAID" "7" "LVM TOOLS #VERSION#" "Red Hat, Inc" "\""

.SH NAME
lvmraid \(em LVM RAID

.SH DESCRIPTION

LVM RAID is a way to create logical volumes (LVs) that use multiple physical
devices to improve performance or tolerate device failure.  How blocks of
data in an LV are placed onto physical devices is determined by the RAID
level.  RAID levels are commonly referred to by number, e.g. raid1, raid5.
Selecting a RAID level involves tradeoffs among physical device
requirements, fault tolerance, and performance.  A description of the RAID
levels can be found at
.br
www.snia.org/sites/default/files/SNIA_DDF_Technical_Position_v2.0.pdf

LVM RAID uses both Device Mapper (DM) and Multiple Device (MD) drivers
from the Linux kernel.  DM is used to create and manage visible LVM
devices, and MD is used to place data on physical devices.

.SH Create a RAID LV

To create a RAID LV, use lvcreate and specify an LV type.
The LV type corresponds to a RAID level.
The basic RAID levels that can be used are:
.B raid0, raid1, raid4, raid5, raid6, raid10.

.B lvcreate \-\-type
.I RaidLevel
[\fIOPTIONS\fP]
.B \-\-name
.I Name
.B \-\-size
.I Size
.I VG
[\fIPVs\fP]

To display the LV type of an existing LV, run:

.B lvs -o name,segtype
\fIVG\fP/\fILV\fP

(The LV type is also referred to as "segment type" or "segtype".)

LVs can be created with the following types:

.SS raid0

\&

Also called striping, raid0 spreads LV data across multiple devices in
units of stripe size.  This is used to increase performance.  LV data will
be lost if any of the devices fail.

.B lvcreate \-\-type raid0
[\fB\-\-stripes\fP \fINumber\fP \fB\-\-stripesize\fP \fISize\fP]
\fIVG\fP
[\fIPVs\fP]

.HP
.B \-\-stripes
specifies the number of devices to spread the LV across.

.HP
.B \-\-stripesize
specifies the size of each stripe in kilobytes.  This is the amount of
data that is written to one device before moving to the next.
.P

\fIPVs\fP specifies the devices to use.  If not specified, lvm will choose
\fINumber\fP devices, one for each stripe.

.SS raid1

\&

Also called mirroring, raid1 uses multiple devices to duplicate LV data.
The LV data remains available if all but one of the devices fail.
The minimum number of devices required is 2.

.B lvcreate \-\-type raid1
[\fB\-\-mirrors\fP \fINumber\fP]
\fIVG\fP
[\fIPVs\fP]

.HP
.B \-\-mirrors
specifies the number of mirror images in addition to the original LV
image, e.g. \-\-mirrors 1 means there are two images of the data, the
original and one mirror image.
.P

\fIPVs\fP specifies the devices to use.  If not specified, lvm will choose
\fINumber\fP devices, one for each image.

.SS raid4

\&

raid4 is a form of striping that uses an extra device dedicated to storing
parity blocks.  The LV data remains available if one device fails.  The
parity is used to recalculate data that is lost from a single device.  The
minimum number of devices required is 3.

.B lvcreate \-\-type raid4
[\fB\-\-stripes\fP \fINumber\fP \fB\-\-stripesize\fP \fISize\fP]
\fIVG\fP
[\fIPVs\fP]

.HP
.B \-\-stripes
specifies the number of devices to use for LV data.  This does not include
the extra device lvm adds for storing parity blocks.  \fINumber\fP stripes
requires \fINumber\fP+1 devices.  \fINumber\fP must be 2 or more.

.HP
.B \-\-stripesize
specifies the size of each stripe in kilobytes.  This is the amount of
data that is written to one device before moving to the next.
.P

\fIPVs\fP specifies the devices to use.  If not specified, lvm will choose
\fINumber\fP+1 separate devices.

raid4 is called non-rotating parity because the parity blocks are always
stored on the same device.

.SS raid5

\&

raid5 is a form of striping that uses an extra device for storing parity
blocks.  LV data and parity blocks are stored on each device.  The LV data
remains available if one device fails.  The parity is used to recalculate
data that is lost from a single device.  The minimum number of devices
required is 3.

.B lvcreate \-\-type raid5
[\fB\-\-stripes\fP \fINumber\fP \fB\-\-stripesize\fP \fISize\fP]
\fIVG\fP
[\fIPVs\fP]

.HP
.B \-\-stripes
specifies the number of devices to use for LV data.  This does not include
the extra device lvm adds for storing parity blocks.  \fINumber\fP stripes
requires \fINumber\fP+1 devices.  \fINumber\fP must be 2 or more.

.HP
.B \-\-stripesize
specifies the size of each stripe in kilobytes.  This is the amount of
data that is written to one device before moving to the next.
.P

\fIPVs\fP specifies the devices to use.  If not specified, lvm will choose
\fINumber\fP+1 separate devices.

raid5 is called rotating parity because the parity blocks are placed on
different devices in a round-robin sequence.  There are variations of
raid5 with different algorithms for placing the parity blocks.  The
default variant is raid5_ls (raid5 left symmetric, which is a rotating
parity 0 with data restart.)  See \fBRAID5 variants\fP below.

.SS raid6

\&

raid6 is a form of striping like raid5, but uses two extra devices for
parity blocks.  LV data and parity blocks are stored on each device.  The
LV data remains available if up to two devices fail.  The parity is used
to recalculate data that is lost from one or two devices.  The minimum
number of devices required is 5.

.B lvcreate \-\-type raid6
[\fB\-\-stripes\fP \fINumber\fP \fB\-\-stripesize\fP \fISize\fP]
\fIVG\fP
[\fIPVs\fP]

.HP
.B \-\-stripes
specifies the number of devices to use for LV data.  This does not include
the extra two devices lvm adds for storing parity blocks.  \fINumber\fP
stripes requires \fINumber\fP+2 devices.  \fINumber\fP must be 3 or more.

.HP
.B \-\-stripesize
specifies the size of each stripe in kilobytes.  This is the amount of
data that is written to one device before moving to the next.
.P

\fIPVs\fP specifies the devices to use.  If not specified, lvm will choose
\fINumber\fP+2 separate devices.

Like raid5, there are variations of raid6 with different algorithms for
placing the parity blocks.  The default variant is raid6_zr (raid6 zero
restart, aka left symmetric, which is a rotating parity 0 with data
restart.)  See \fBRAID6 variants\fP below.

.SS raid10

\&

raid10 is a combination of raid1 and raid0, striping data across mirrored
devices.  LV data remains available if one or more devices remains in each
mirror set.  The minimum number of devices required is 4.

.B lvcreate \-\-type raid10
.RS
[\fB\-\-mirrors\fP \fINumberMirrors\fP]
.br
[\fB\-\-stripes\fP \fINumberStripes\fP \fB\-\-stripesize\fP \fISize\fP]
.br
\fIVG\fP
[\fIPVs\fP]
.RE

.HP
.B \-\-mirrors
specifies the number of mirror images within each stripe.  e.g.
\-\-mirrors 1 means there are two images of the data, the original and one
mirror image.

.HP
.B \-\-stripes
specifies the total number of devices to use in all raid1 images (not the
number of raid1 devices to spread the LV across, even though that is the
effective result).  The number of devices in each raid1 mirror will be
NumberStripes/(NumberMirrors+1), e.g. mirrors 1 and stripes 4 will stripe
data across two raid1 mirrors, where each mirror is devices.

.HP
.B \-\-stripesize
specifies the size of each stripe in kilobytes.  This is the amount of
data that is written to one device before moving to the next.
.P

\fIPVs\fP specifies the devices to use.  If not specified, lvm will choose
the necessary devices.  Devices are used to create mirrors in the
order listed, e.g. for mirrors 1, stripes 2, listing PV1 PV2 PV3 PV4
results in mirrors PV1/PV2 and PV3/PV4.

RAID10 is not mirroring on top of stripes, which would be RAID01, which is
less tolerant of device failures.


.SH Synchronization

Synchronization makes all the devices in a RAID LV consistent with each
other.

In a RAID1 LV, all mirror images should have the same data.  When a new
mirror image is added, or a mirror image is missing data, then images need
to be synchronized.  Data blocks are copied from an existing image to a
new or outdated image to make them match.

In a RAID 4/5/6 LV, parity blocks and data blocks should match based on
the parity calculation.  When the devices in a RAID LV change, the data
and parity blocks can become inconsistent and need to be synchronized.
Correct blocks are read, parity is calculated, and recalculated blocks are
written.

The RAID implementation keeps track of which parts of a RAID LV are
synchronized.  This uses a bitmap saved in the RAID metadata.  The bitmap
can exclude large parts of the LV from synchronization to reduce the
amount of work.  Without this, the entire LV would need to be synchronized
every time it was activated.  When a RAID LV is first created and
activated the first synchronization is called initialization.

Automatic synchronization happens when a RAID LV is activated, but it is
usually partial because the bitmaps reduce the areas that are checked.
A full sync may become necessary when devices in the RAID LV are changed.

The synchronization status of a RAID LV is reported by the
following command, where "image synced" means sync is complete:

.B lvs -a -o name,sync_percent


.SS Scrubbing

Scrubbing is a full scan/synchronization of the RAID LV requested by a user.
Scrubbing can find problems that are missed by partial synchronization.

Scrubbing assumes that RAID metadata and bitmaps may be inaccurate, so it
verifies all RAID metadata, LV data, and parity blocks.  Scrubbing can
find inconsistencies caused by hardware errors or degradation.  These
kinds of problems may be undetected by automatic synchronization which
excludes areas outside of the RAID write-intent bitmap.

The command to scrub a RAID LV can operate in two different modes:

.B lvchange \-\-syncaction
.BR check | repair
.IR VG / LV

.HP
.B check
Check mode is read\-only and only detects inconsistent areas in the RAID
LV, it does not correct them.

.HP
.B repair
Repair mode checks and writes corrected blocks to synchronize any
inconsistent areas.

.P

Scrubbing can consume a lot of bandwidth and slow down application I/O on
the RAID LV.  To control the I/O rate used for scrubbing, use:

.HP
.B \-\-maxrecoveryrate
.BR \fIRate [ b | B | s | S | k | K | m | M | g | G ]
.br
Sets the maximum recovery rate for a RAID LV.  \fIRate\fP is specified as
an amount per second for each device in the array.  If no suffix is given,
then KiB/sec/device is assumed.  Setting the recovery rate to \fB0\fP
means it will be unbounded.

.HP
.BR \-\-minrecoveryrate
.BR \fIRate [ b | B | s | S | k | K | m | M | g | G ]
.br
Sets the minimum recovery rate for a RAID LV.  \fIRate\fP is specified as
an amount per second for each device in the array.  If no suffix is given,
then KiB/sec/device is assumed.  Setting the recovery rate to \fB0\fP
means it will be unbounded.

.P

To display the current scrubbing in progress on an LV, including
the syncaction mode and percent complete, run:

.B lvs -a -o name,raid_sync_action,sync_percent

After scrubbing is complete, to display the number of inconsistent blocks
found, run:

.B lvs -o name,raid_mismatch_count

Also, if mismatches were found, the lvs attr field will display the letter
"m" (mismatch) in the 9th position, e.g.

.nf
# lvs -o name,vgname,segtype,attr vg/lvol0
  LV    VG   Type  Attr
  lvol0 vg   raid1 Rwi-a-r-m- 
.fi


.SS Scrubbing Limitations

The \fBcheck\fP mode can only report the number of inconsistent blocks, it
cannot report which blocks are inconsistent.  This makes it impossible to
know which device has errors, or if the errors affect file system data,
metadata or nothing at all.

The \fBrepair\fP mode can make the RAID LV data consistent, but it does
not know which data is correct.  The result may be consistent but
incorrect data.  When two different blocks of data must be made
consistent, it chooses the block from the device that would be used during
RAID intialization.  However, if the PV holding corrupt data is known,
lvchange \-\-rebuild can be used to reconstruct the data on the bad
device.

Future developments might include:

Allowing a user to choose the correct version of data during repair.

Using a majority of devices to determine the correct version of data to
use in a three-way RAID1 or RAID6 LV.

Using a checksumming device to pin-point when and where an error occurs,
allowing it to be rewritten.


.SH SubLVs

An LV is often a combination of other hidden LVs called SubLVs.  The
SubLVs either use physical devices, or are built from other SubLVs
themselves.  SubLVs hold LV data blocks, RAID parity blocks, and RAID
metadata.  SubLVs are generally hidden, so the lvs \-a option is required
display them:

.B lvs -a -o name,segtype,devices

SubLV names begin with the visible LV name, and have an automatic suffix
indicating its role:

.IP \(bu 3
SubLVs holding LV data or parity blocks have the suffix _rimage_#.
These SubLVs are sometimes referred to as DataLVs.

.IP \(bu 3
SubLVs holding RAID metadata have the suffix _rmeta_#.  RAID metadata
includes superblock information, RAID type, bitmap, and device health
information.  These SubLVs are sometimes referred to as MetaLVs.

.P

SubLVs are an internal implementation detail of LVM.  The way they are
used, constructed and named may change.

The following examples show the SubLV arrangement for each of the basic
RAID LV types, using the fewest number of devices allowed for each.

.SS Examples

.B raid0
.br
Each rimage SubLV holds a portion of LV data.  No parity is used.
No RAID metadata is used.

.nf
lvcreate --type raid0 --stripes 2 --name lvr0 ...

lvs -a -o name,segtype,devices
  lvr0            raid0  lvr0_rimage_0(0),lvr0_rimage_1(0)
  [lvr0_rimage_0] linear /dev/sda(...)
  [lvr0_rimage_1] linear /dev/sdb(...)
.fi

.B raid1
.br
Each rimage SubLV holds a complete copy of LV data.  No parity is used.
Each rmeta SubLV holds RAID metadata.

.nf
lvcreate --type raid1 --mirrors 1 --name lvr1 ...

lvs -a -o name,segtype,devices
  lvr1            raid1  lvr1_rimage_0(0),lvr1_rimage_1(0)
  [lvr1_rimage_0] linear /dev/sda(...)
  [lvr1_rimage_1] linear /dev/sdb(...)
  [lvr1_rmeta_0]  linear /dev/sda(...)
  [lvr1_rmeta_1]  linear /dev/sdb(...)
.fi

.B raid4
.br
Two rimage SubLVs each hold a portion of LV data and one rimage SubLV
holds parity.  Each rmeta SubLV holds RAID metadata.

.nf
lvcreate --type raid4 --stripes 2 --name lvr4 ...

lvs -a -o name,segtype,devices
  lvr4            raid4  lvr4_rimage_0(0),\\
                         lvr4_rimage_1(0),\\
                         lvr4_rimage_2(0)
  [lvr4_rimage_0] linear /dev/sda(...)
  [lvr4_rimage_1] linear /dev/sdb(...)
  [lvr4_rimage_2] linear /dev/sdc(...)
  [lvr4_rmeta_0]  linear /dev/sda(...)
  [lvr4_rmeta_1]  linear /dev/sdb(...)
  [lvr4_rmeta_2]  linear /dev/sdc(...)
.fi

.B raid5
.br
Three rimage SubLVs each hold a portion of LV data and parity.
Each rmeta SubLV holds RAID metadata.

.nf
lvcreate --type raid5 --stripes 2 --name lvr5 ...

lvs -a -o name,segtype,devices
  lvr5            raid5  lvr5_rimage_0(0),\\
                         lvr5_rimage_1(0),\\
                         lvr5_rimage_2(0)
  [lvr5_rimage_0] linear /dev/sda(...)                                     
  [lvr5_rimage_1] linear /dev/sdb(...)                           
  [lvr5_rimage_2] linear /dev/sdc(...)                                      
  [lvr5_rmeta_0]  linear /dev/sda(...)                                     
  [lvr5_rmeta_1]  linear /dev/sdb(...)                           
  [lvr5_rmeta_2]  linear /dev/sdc(...)                                      
.fi

.B raid6
.br
Six rimage SubLVs each hold a portion of LV data and parity.
Each rmeta SubLV holds RAID metadata.

.nf
lvcreate --type raid6 --stripes 3 --name lvr6

lvs -a -o name,segtype,devices
  lvr6            raid6  lvr6_rimage_0(0),\\
                         lvr6_rimage_1(0),\\
                         lvr6_rimage_2(0),\\
                         lvr6_rimage_3(0),\\
                         lvr6_rimage_4(0),\\
                         lvr6_rimage_5(0)
  [lvr6_rimage_0] linear /dev/sda(...)
  [lvr6_rimage_1] linear /dev/sdb(...)
  [lvr6_rimage_2] linear /dev/sdc(...)
  [lvr6_rimage_3] linear /dev/sdd(...)
  [lvr6_rimage_4] linear /dev/sde(...)
  [lvr6_rimage_5] linear /dev/sdf(...)
  [lvr6_rmeta_0]  linear /dev/sda(...)
  [lvr6_rmeta_1]  linear /dev/sdb(...)
  [lvr6_rmeta_2]  linear /dev/sdc(...)
  [lvr6_rmeta_3]  linear /dev/sdd(...)
  [lvr6_rmeta_4]  linear /dev/sde(...)
  [lvr6_rmeta_5]  linear /dev/sdf(...)

.B raid10
.br
Four rimage SubLVs each hold a portion of LV data.  No parity is used.
Each rmeta SubLV holds RAID metadata.

.nf
lvcreate --type raid10 --stripes 2 --mirrors 1 --name lvr10

lvs -a -o name,segtype,devices
  lvr10            raid10 lvr10_rimage_0(0),\\
                          lvr10_rimage_1(0),\\
                          lvr10_rimage_2(0),\\
                          lvr10_rimage_3(0)
  [lvr10_rimage_0] linear /dev/sda(...)
  [lvr10_rimage_1] linear /dev/sdb(...)
  [lvr10_rimage_2] linear /dev/sdc(...)
  [lvr10_rimage_3] linear /dev/sdd(...)
  [lvr10_rmeta_0]  linear /dev/sda(...)
  [lvr10_rmeta_1]  linear /dev/sdb(...)
  [lvr10_rmeta_2]  linear /dev/sdc(...)
  [lvr10_rmeta_3]  linear /dev/sdd(...)
.fi


.SH Device Failure

Physical devices in a RAID LV can fail or be lost for multiple reasons.
A device could be disconnected, permanently failed, or temporarily
disconnected.  The purpose of RAID LVs (levels 1 and higher) is to
continue operating in a degraded mode, without losing LV data, even after
a device fails.  The number of devices that can fail without the loss of
LV data depends on the RAID level:

.IP \[bu] 3
RAID0 (striped) LVs cannot tolerate losing any devices.  LV data will be
lost if any devices fail.

.IP \[bu] 3
RAID1 LVs can tolerate losing all but one device without LV data loss.

.IP \[bu] 3
RAID4 and RAID5 LVs can tolerate losing one device without LV data loss.

.IP \[bu] 3
RAID6 LVs can tolerate losing two devices without LV data loss.

.IP \[bu] 3
RAID10 is variable, and depends on which devices are lost.  It can
tolerate losing all but one device in a single raid1 mirror without
LV data loss.

.P

If a RAID LV is missing devices, or has other device-related problems, lvs
reports this in the health_status (and attr) fields:

.B lvs -o name,lv_health_status

.B partial
.br
Devices are missing from the LV.  This is also indicated by the letter "p"
(partial) in the 9th position of the lvs attr field.

.B refresh needed
.br
A device was temporarily missing but has returned.  The LV needs to be
refreshed to use the device again (which will usually require
partial synchronization).  This is also indicated by the letter "r" (refresh
needed) in the 9th position of the lvs attr field.  See
\fBRefreshing an LV\fP.  This could also indicate a problem with the
device, in which case it should be be replaced, see
\fBReplacing Devices\fP.

.B mismatches exist
.br
See
.BR Scrubbing .

Most commands will also print a warning if a device is missing, e.g.
.br
.nf
WARNING: Device for PV uItL3Z-wBME-DQy0-... not found or rejected ...
.fi

This warning will go away if the device returns or is removed from the
VG (see \fBvgreduce \-\-removemissing\fP).


.SS Activating an LV with missing devices

A RAID LV that is missing devices may be activated or not, depending on
the "activation mode" used in lvchange:

.B lvchange \-ay \-\-activationmode
.RB { complete | degraded | partial }
.IR VG / LV

.B complete
.br
The LV is only activated if all devices are present.

.B degraded
.br
The LV is activated with missing devices if the RAID level can
tolerate the number of missing devices without LV data loss.

.B partial
.br
The LV is always activated, even if portions of the LV data are missing
because of the missing device(s).  This should only be used to perform
recovery or repair operations.

.BR lvm.conf (5)
.B activation/activation_mode
.br
controls the activation mode when not specified by the command.

The default value is printed by:
.nf
lvmconfig --type default activation/activation_mode
.fi

.SS Replacing Devices

Devices in a RAID LV can be replaced with other devices in the VG.  When
replacing devices that are no longer visible on the system, use lvconvert
\-\-repair.  When replacing devices that are still visible, use lvconvert
\-\-replace.  The repair command will attempt to restore the same number
of data LVs that were previously in the LV.  The replace option can be
repeated to replace multiple PVs.  Replacement devices can be optionally
listed with either option.

.B lvconvert \-\-repair
.IR VG / LV
[\fINewPVs\fP]

.B lvconvert \-\-replace
\fIOldPV\fP
.IR VG / LV
[\fINewPV\fP]

.B lvconvert
.B \-\-replace
\fIOldPV1\fP
.B \-\-replace
\fIOldPV2\fP
...
.IR VG / LV
[\fINewPVs\fP]

New devices require synchronization with existing devices, see
.BR Synchronization .

.SS Refreshing an LV

Refreshing a RAID LV clears any transient device failures (device was
temporarily disconnected) and returns the LV to its fully redundant mode.
Restoring a device will usually require at least partial synchronization
(see \fBSynchronization\fP).  Failure to clear a transient failure results
in the RAID LV operating in degraded mode until it is reactivated.  Use
the lvchange command to refresh an LV:

.B lvchange \-\-refresh
.IR VG / LV

.nf
# lvs -o name,vgname,segtype,attr,size vg
  LV    VG   Type  Attr       LSize
  raid1 vg   raid1 Rwi-a-r-r- 100.00g

# lvchange --refresh vg/raid1

# lvs -o name,vgname,segtype,attr,size vg
  LV    VG   Type  Attr       LSize
  raid1 vg   raid1 Rwi-a-r--- 100.00g
.fi

.SS Automatic repair

If a device in a RAID LV fails, device-mapper in the kernel notifies the
.BR dmeventd (8)
monitoring process (see \fBMonitoring\fP).
dmeventd can be configured to automatically respond using:

.BR lvm.conf (5)
.B activation/raid_fault_policy

Possible settings are:

.B warn
.br
A warning is added to the system log indicating that a device has
failed in the RAID LV.  It is left to the user to repair the LV, e.g.
replace failed devices.

.B allocate
.br
dmeventd automatically attempts to repair the LV using spare devices
in the VG.  Note that even a transient failure is handled as a permanent
failure; a new device is allocated and full synchronization is started.

The specific command run by dmeventd to warn or repair is:
.br
.B lvconvert \-\-repair \-\-use\-policies
.IR VG / LV


.SS Corrupted Data

Data on a device can be corrupted due to hardware errors, without the
device ever being disconnected, and without any fault in the software.
This should be rare, and can be detected (see \fBScrubbing\fP).


.SS Rebuild specific PVs

If specific PVs in a RAID LV are known to have corrupt data, the data on
those PVs can be reconstructed with:

.B lvchange \-\-rebuild PV
.IR VG / LV

The rebuild option can be repeated with different PVs to replace the data
on multiple PVs.


.SH Monitoring

When a RAID LV is activated the \fBdmeventd\fP(8) process is started to
monitor the health of the LV.  Various events detected in the kernel can
cause a notification to be sent from device-mapper to the monitoring
process, including device failures and synchronization completion (e.g.
for initialization or scrubbing).

The LVM configuration file contains options that affect how the monitoring
process will respond to failure events (e.g. raid_fault_policy).  It is
possible to turn on and off monitoring with lvchange, but it is not
recommended to turn this off unless you have a thorough knowledge of the
consequences.


.SH Configuration Options

There are a number of options in the LVM configuration file that affect
the behavior of RAID LVs.  The tunable options are listed
below.  A detailed description of each can be found in the LVM
configuration file itself.
.br
        mirror_segtype_default
.br
        raid10_segtype_default
.br
        raid_region_size
.br
        raid_fault_policy
.br
        activation_mode


.SH RAID1 Tuning

A RAID1 LV can be tuned so that certain devices are avoided for reading
while all devices are still written to.

.B lvchange
.BR \-\- [ raid ] writemostly
.BR \fIPhysicalVolume [ : { y | n | t }]
.IR VG / LV

The specified device will be marked as "write mostly", which means that
reading from this device will be avoided, and other devices will be
preferred for reading (unless no other devices are available.)  This
minimizes the I/O to the specified device.

If the PV name has no suffix, the write mostly attribute is set.  If the
PV name has the suffix \fB:n\fP, the write mostly attribute is cleared,
and the suffix \fB:t\fP toggles the current setting.

The write mostly option can be repeated on the command line to change
multiple devices at once.

To report the current write mostly setting, the lvs attr field will show
the letter "w" in the 9th position when write mostly is set:

.B lvs -a -o name,attr

When a device is marked write mostly, the maximum number of outstanding
writes to that device can be configured.  Once the maximum is reached,
further writes become synchronous.  When synchronous, a write to the LV
will not complete until writes to all the mirror images are complete.

.B lvchange
.BR \-\- [ raid ] writebehind
.IR IOCount
.IR VG / LV

To report the current write behind setting, run:

.B lvs -o name,raid_write_behind

When write behind is not configured, or set to 0, all LV writes are
synchronous.


.SH RAID Takeover

RAID takeover is converting a RAID LV from one RAID level to another, e.g.
raid5 to raid6.  Changing the RAID level is usually done to increase or
decrease resilience to device failures.  This is done using lvconvert and
specifying the new RAID level as the LV type:

.B lvconvert --type
.I RaidLevel
\fIVG\fP/\fILV\fP
[\fIPVs\fP]

The most common and recommended RAID takeover conversions are:

.HP
\fBlinear\fP to \fBraid1\fP
.br
Linear is a single image of LV data, and
converting it to raid1 adds a mirror image which is a direct copy of the
original linear image.

.HP
\fBstriped\fP/\fBraid0\fP to \fBraid4/5/6\fP
.br
Adding parity devices to a
striped volume results in raid4/5/6.

.P

Unnatural conversions that are not recommended include converting between
striped and non-striped types.  This is because file systems often
optimize I/O patterns based on device striping values.  If those values
change, it can decrease performance.

Converting to a higher RAID level requires allocating new SubLVs to hold
RAID metadata, and new SubLVs to hold parity blocks for LV data.
Converting to a lower RAID level removes the SubLVs that are no longer
needed.

Conversion often requires full synchronization of the RAID LV (see
\fBSynchronization\fP).  Converting to RAID1 requires copying all LV data
blocks to a new image on a new device.  Converting to a parity RAID level
requires reading all LV data blocks, calculating parity, and writing the
new parity blocks.  Synchronization can take a long time and degrade
performance (rate controls also apply to conversion, see
\fB\-\-maxrecoveryrate\fP.)

.P

The following takeover conversions are currently possible:
.br
.IP \(bu 3
between linear and raid1.
.IP \(bu 3
between striped and raid4.

.SS Examples

1. Converting an LV from \fBlinear\fP to \fBraid1\fP.

.nf
# lvs -a -o name,segtype,size vg
  LV   Type   LSize
  lv   linear 300.00g

# lvconvert --type raid1 --mirrors 1 vg/lv

# lvs -a -o name,segtype,size vg
  LV            Type   LSize
  lv            raid1  300.00g
  [lv_rimage_0] linear 300.00g
  [lv_rimage_1] linear 300.00g
  [lv_rmeta_0]  linear   3.00m
  [lv_rmeta_1]  linear   3.00m
.fi

2. Converting an LV from \fBmirror\fP to \fBraid1\fP.

.nf
# lvs -a -o name,segtype,size vg
  LV            Type   LSize
  lv            mirror 100.00g
  [lv_mimage_0] linear 100.00g
  [lv_mimage_1] linear 100.00g
  [lv_mlog]     linear   3.00m

# lvconvert --type raid1 vg/lv

# lvs -a -o name,segtype,size vg
  LV            Type   LSize
  lv            raid1  100.00g
  [lv_rimage_0] linear 100.00g
  [lv_rimage_1] linear 100.00g
  [lv_rmeta_0]  linear   3.00m
  [lv_rmeta_1]  linear   3.00m
.fi

3. Converting an LV from \fBlinear\fP to \fBraid1\fP (with 3 images).

.nf
Start with a linear LV:

# lvcreate -L1G -n my_lv vg

Convert the linear LV to raid1 with three images
(original linear image plus 2 mirror images):

# lvconvert --type raid1 --mirrors 2 vg/my_lv
.fi

.ig
4. Converting an LV from \fBstriped\fP (with 4 stripes) to \fBraid6_nc\fP.

.nf
Start with a striped LV:

# lvcreate --stripes 4 -L64M -n my_lv vg

Convert the striped LV to raid6_nc:

# lvconvert --type raid6_nc vg/my_lv

# lvs -a -o lv_name,segtype,sync_percent,data_copies
  LV               Type      Cpy%Sync #Cpy
  my_lv            raid6_n_6 100.00      3
  [my_lv_rimage_0] linear
  [my_lv_rimage_1] linear
  [my_lv_rimage_2] linear
  [my_lv_rimage_3] linear
  [my_lv_rimage_4] linear
  [my_lv_rimage_5] linear
  [my_lv_rmeta_0]  linear
  [my_lv_rmeta_1]  linear
  [my_lv_rmeta_2]  linear
  [my_lv_rmeta_3]  linear
  [my_lv_rmeta_4]  linear
  [my_lv_rmeta_5]  linear
.fi

This convert begins by allocating MetaLVs (rmeta_#) for each of the
existing stripe devices.  It then creates 2 additional MetaLV/DataLV pairs
(rmeta_#/rimage_#) for dedicated raid6 parity.

If rotating data/parity is required, such as with raid6_nr, it must be
done by reshaping (see below).
..


.SH RAID Reshaping

RAID reshaping is changing attributes of a RAID LV while keeping the same
RAID level, i.e. changes that do not involve changing the number of
devices.  This includes changing RAID layout, stripe size, or number of
stripes.

When changing the RAID layout or stripe size, no new SubLVs (MetaLVs or
DataLVs) need to be allocated, but DataLVs are extended by a small amount
(typically 1 extent).  The extra space allows blocks in a stripe to be
updated safely, and not corrupted in case of a crash.  If a crash occurs,
reshaping can just be restarted.

(If blocks in a stripe were updated in place, a crash could leave them
partially updated and corrupted.  Instead, an existing stripe is quiesced,
read, changed in layout, and the new stripe written to free space.  Once
that is done, the new stripe is unquiesced and used.)

(The reshaping features are planned for a future release.)

.ig
.SS Examples

1. Converting raid6_n_6 to raid6_nr with rotating data/parity.

This conversion naturally follows a previous conversion from striped to
raid6_n_6 (shown above).  It completes the transition to a more
traditional RAID6.

.nf
# lvs -o lv_name,segtype,sync_percent,data_copies
  LV               Type      Cpy%Sync #Cpy
  my_lv            raid6_n_6 100.00      3
  [my_lv_rimage_0] linear
  [my_lv_rimage_1] linear
  [my_lv_rimage_2] linear
  [my_lv_rimage_3] linear
  [my_lv_rimage_4] linear
  [my_lv_rimage_5] linear
  [my_lv_rmeta_0]  linear
  [my_lv_rmeta_1]  linear
  [my_lv_rmeta_2]  linear
  [my_lv_rmeta_3]  linear
  [my_lv_rmeta_4]  linear
  [my_lv_rmeta_5]  linear

# lvconvert --type raid6_nr vg/my_lv

# lvs -a -o lv_name,segtype,sync_percent,data_copies
  LV               Type     Cpy%Sync #Cpy
  my_lv            raid6_nr 100.00      3
  [my_lv_rimage_0] linear
  [my_lv_rimage_0] linear
  [my_lv_rimage_1] linear
  [my_lv_rimage_1] linear
  [my_lv_rimage_2] linear
  [my_lv_rimage_2] linear
  [my_lv_rimage_3] linear
  [my_lv_rimage_3] linear
  [my_lv_rimage_4] linear
  [my_lv_rimage_5] linear
  [my_lv_rmeta_0]  linear
  [my_lv_rmeta_1]  linear
  [my_lv_rmeta_2]  linear
  [my_lv_rmeta_3]  linear
  [my_lv_rmeta_4]  linear
  [my_lv_rmeta_5]  linear
.fi

The DataLVs are larger (additional segment in each) which provides space
for out-of-place reshaping.  The result is:

FIXME: did the lv name change from my_lv to r?
.br
FIXME: should we change device names in the example to sda,sdb,sdc?
.br
FIXME: include -o devices or seg_pe_ranges above also?

.nf
# lvs -a -o lv_name,segtype,seg_pe_ranges,dataoffset
  LV           Type     PE Ranges          data
  r            raid6_nr r_rimage_0:0-32 \\
                        r_rimage_1:0-32 \\
                        r_rimage_2:0-32 \\
                        r_rimage_3:0-32
  [r_rimage_0] linear   /dev/sda:0-31      2048
  [r_rimage_0] linear   /dev/sda:33-33
  [r_rimage_1] linear   /dev/sdaa:0-31     2048
  [r_rimage_1] linear   /dev/sdaa:33-33
  [r_rimage_2] linear   /dev/sdab:1-33     2048
  [r_rimage_3] linear   /dev/sdac:1-33     2048
  [r_rmeta_0]  linear   /dev/sda:32-32
  [r_rmeta_1]  linear   /dev/sdaa:32-32
  [r_rmeta_2]  linear   /dev/sdab:0-0
  [r_rmeta_3]  linear   /dev/sdac:0-0
.fi

All segments with PE ranges '33-33' provide the out-of-place reshape space.
The dataoffset column shows that the data was moved from initial offset 0 to
2048 sectors on each component DataLV.
..

.SH RAID5 Variants

raid5_ls
.br
\[bu]
RAID5 left symmetric
.br
\[bu]
Rotating parity N with data restart

raid5_la
.br
\[bu]
RAID5 left symmetric
.br
\[bu]
Rotating parity N with data continuation

raid5_rs
.br
\[bu]
RAID5 right symmetric
.br
\[bu]
Rotating parity 0 with data restart

raid5_ra
.br
\[bu]
RAID5 right asymmetric
.br
\[bu]
Rotating parity 0 with data continuation

.ig
raid5_n
.br
\[bu]
RAID5 striping
.br
\[bu]
Same layout as raid4 with a dedicated parity N with striped data.
.br
\[bu]
Used for
.B RAID Takeover
..

.SH RAID6 Variants

raid6
.br
\[bu]
RAID6 zero restart (aka left symmetric)
.br
\[bu]
Rotating parity 0 with data restart
.br
\[bu]
Same as raid6_zr

raid6_zr
.br
\[bu]
RAID6 zero restart (aka left symmetric)
.br
\[bu]
Rotating parity 0 with data restart

raid6_nr
.br
\[bu]
RAID6 N restart (aka right symmetric)
.br
\[bu]
Rotating parity N with data restart

raid6_nc
.br
\[bu]
RAID6 N continue
.br
\[bu]
Rotating parity N with data continuation

.ig
raid6_n_6
.br
\[bu]
RAID6 N continue
.br
\[bu]
Fixed P-Syndrome N-1 and Q-Syndrome N with striped data
.br
\[bu]
Used for
.B RAID Takeover

raid6_ls_6
.br
\[bu]
RAID6 N continue
.br
\[bu]
Same as raid5_ls for N-1 disks with fixed Q-Syndrome N
.br
\[bu]
Used for
.B RAID Takeover

raid6_la_6
.br
\[bu]
RAID6 N continue
.br
\[bu]
Same as raid5_la for N-1 disks with fixed Q-Syndrome N
.br
\[bu]
Used for
.B RAID Takeover

raid6_rs_6
.br
\[bu]
RAID6 N continue
.br
\[bu]
Same as raid5_rs for N-1 disks with fixed Q-Syndrome N
.br
\[bu]
Used for
.B RAID Takeover

raid6_ra_6
.br
\[bu]
RAID6 N continue
.br
\[bu]
Same as raid5_ra for N-1 disks with fixed Q-Syndrome N
.br
\[bu]
Used for
.B RAID Takeover
..


.ig
.SH RAID Duplication

RAID LV conversion (takeover or reshaping) can be done out\-of\-place by
copying the LV data onto new devices while changing the RAID properties.
Copying avoids modifying the original LV but requires additional devices.
Once the LV data has been copied/converted onto the new devices, there are
multiple options:

1. The RAID LV can be switched over to run from just the new devices, and
the original copy of the data removed.  The converted LV then has the new
RAID properties, and exists on new devices.  The old devices holding the
original data can be removed or reused.

2. The new copy of the data can be dropped, leaving the original RAID LV
unchanged and using its original devices.

3. The new copy of the data can be separated and used as a new independent
LV, leaving the original RAID LV unchanged on its original devices.

The command to start duplication is:

.B lvconvert \-\-type
.I RaidLevel
[\fB\-\-stripes\fP \fINumber\fP \fB\-\-stripesize\fP \fISize\fP]
.RS
.B \-\-duplicate
.IR VG / LV
[\fIPVs\fP]
.RE

.HP
.B \-\-duplicate
.br
Specifies that the LV conversion should be done out\-of\-place, copying
LV data to new devices while converting. 

.HP
.BR \-\-type , \-\-stripes , \-\-stripesize
.br
Specifies the RAID properties to use when creating the copy.

.P
\fIPVs\fP specifies the new devices to use.

The steps in the duplication process:

.IP \(bu 3
LVM creates a new LV on new devices using the specified RAID properties
(type, stripes, etc) and optionally specified devices.

.IP \(bu 3
LVM changes the visible RAID LV to type raid1, making the original LV the
first raid1 image (SubLV 0), and the new LV the second raid1 image
(SubLV 1).

.IP \(bu 3
The RAID1 synchronization process copies data from the original LV
image (SubLV 0) to the new LV image (SubLV 1).

.IP \(bu 3
When synchronization is complete, the original and new LVs are
mirror images of each other and can be separated.

.P

The duplication process retains both the original and new LVs (both
SubLVs) until an explicit unduplicate command is run to separate them.  The
unduplicate command specifies if the original LV should use the old
devices (SubLV 0) or the new devices (SubLV 1).

To make the RAID LV use the data on the old devices, and drop the copy on
the new devices, specify the name of SubLV 0 (suffix _dup_0):

.B lvconvert \-\-unduplicate
.BI \-\-name
.IB LV _dup_0
.IR VG / LV

To make the RAID LV use the data copy on the new devices, and drop the old
devices, specify the name of SubLV 1 (suffix _dup_1):

.B lvconvert \-\-unduplicate
.BI \-\-name
.IB LV _dup_1
.IR VG / LV

FIXME: To make the LV use the data on the original devices, but keep the
data copy as a new LV, ...

FIXME: include how splitmirrors can be used.


.SH RAID1E

TODO
..

.SH History

The 2.6.38-rc1 version of the Linux kernel introduced a device-mapper
target to interface with the software RAID (MD) personalities.  This
provided device-mapper with RAID 4/5/6 capabilities and a larger
development community.  Later, support for RAID1, RAID10, and RAID1E (RAID
10 variants) were added.  Support for these new kernel RAID targets was
added to LVM version 2.02.87.  The capabilities of the LVM \fBraid1\fP
type have surpassed the old \fBmirror\fP type.  raid1 is now recommended
instead of mirror.  raid1 became the default for mirroring in LVM version
2.02.100.