1
0
mirror of git://sourceware.org/git/lvm2.git synced 2024-12-22 17:35:59 +03:00
lvm2/man/dmstats.8_main

1245 lines
36 KiB
Plaintext
Raw Normal View History

2016-06-22 20:36:10 +03:00
.TH DMSTATS 8 "Jun 23 2016" "Linux" "MAINTENANCE COMMANDS"
.
.de OPT_PROGRAMS
. RB [ --allprograms | --programid
. IR id ]
..
.
.de OPT_REGIONS
. RB [ --allregions | --regionid
. IR id ]
..
.de OPT_OBJECTS
. RB [ --area ]
. RB [ --region ]
. RB [ --group ]
..
.de OPT_FOREGROUND
. RB [ --foreground ]
..
.
.\" Print units suffix, use with arg to print human
.\" man2html can't handle too many changes per command
.de UNITS
. BR b | B | s | S | k | K | m | M | \c
. BR g | G | t | T | p | P | e | E ]
..
.
.\" Print help text for units, use with arg to print human
.de HELP_UNITS
. RB ( b )ytes,
. RB ( s )ectors,
. RB ( k )ilobytes,
. RB ( m )egabytes,
. RB ( g )igabytes,
. RB ( t )erabytes,
. RB ( p )etabytes,
. RB ( e )xabytes.
. nop Capitalise to use multiples of 1000 (S.I.) instead of 1024.
..
.
.SH NAME
.
dmstats \(em device-mapper statistics management
.
.SH SYNOPSIS
.
.B dmsetup
.B stats
.I command
[OPTIONS]
.sp
.
.PD 0
.HP
.B dmstats
.de CMD_COMMAND
. ad l
. nh
. IR command
. IR device_name \ |
. BR --major
. IR major
. BR --minor
. IR minor \ |
. BR -u | --uuid
. IR uuid
. RB [ -v | --verbose ]
. hy
. ad b
..
.CMD_COMMAND
.
.HP
.B dmstats
.de CMD_CLEAR
. ad l
. nh
. BR clear
. IR device_name
. OPT_PROGRAMS
. OPT_REGIONS
. hy
. ad b
..
.CMD_CLEAR
.
.HP
.B dmstats
.de CMD_CREATE
. ad l
. nh
. BR create
. IR device_name ...| file_path ...| \fB--alldevices
. RB [ --areas
. IR nr_areas | \fB--areasize
. IR area_size ]
. RB [ --bounds
. IR histogram_boundaries ]
. RB [ --filemap ]
. RB [ --follow
. IR follow_mode ]
. OPT_FOREGROUND
. RB [ --nomonitor ]
. RB [ --nogroup ]
. RB [ --precise ]
. RB [ --start
. IR start_sector
. BR --length
. IR length | \fB--segments ]
. RB [ --userdata
. IR user_data ]
. RB [ --programid
. IR id ]
. hy
. ad b
..
.CMD_CREATE
.
.HP
.B dmstats
.de CMD_DELETE
. ad l
. nh
. BR delete
. IR device_name | \fB--alldevices
. OPT_PROGRAMS
. OPT_REGIONS
. hy
. ad b
..
.CMD_DELETE
.
.HP
.B dmstats
2016-06-22 20:36:10 +03:00
.de CMD_GROUP
. ad l
. nh
2016-06-22 20:36:10 +03:00
. BR group
. RI [ device_name | \fB--alldevices ]
. RB [ --alias
2016-06-22 20:36:10 +03:00
. IR name ]
. RB [ --regions
2016-06-22 20:36:10 +03:00
. IR regions ]
. hy
2016-06-22 20:36:10 +03:00
. ad b
..
.CMD_GROUP
.HP
.B dmstats
.de CMD_HELP
. ad l
. BR help
. RB [ -c | -C | --columns ]
. ad b
..
.CMD_HELP
.
.HP
.B dmstats
.de CMD_LIST
. ad l
. nh
. BR list
. RI [ device_name ]
. RB [ --histogram ]
. OPT_PROGRAMS
. RB [ --units
. IR units ]
. OPT_OBJECTS
. RB [ --nosuffix ]
. RB [ --notimesuffix ]
. RB [ -v | --verbose ]
. hy
. ad b
..
.CMD_LIST
.
.HP
.B dmstats
.de CMD_PRINT
. ad l
. nh
. BR print
. RI [ device_name ]
. RB [ --clear ]
. OPT_PROGRAMS
. OPT_REGIONS
. hy
. ad b
..
.CMD_PRINT
.
.HP
.B dmstats
.de CMD_REPORT
. ad l
. nh
. BR report
. RI [ device_name ]
. RB [ --interval
. IR seconds ]
. RB [ --count
. IR count ]
. RB [ --units
. IR units ]
. RB [ --histogram ]
. OPT_PROGRAMS
. OPT_REGIONS
. OPT_OBJECTS
. RB [ -O | --sort
. IR sort_fields ]
. RB [ -S | --select
. IR selection ]
. RB [ --units
. IR units ]
. RB [ --nosuffix ]
. RB [ --notimesuffix ]
. hy
. ad b
..
.CMD_REPORT
2016-06-22 20:36:10 +03:00
.HP
.B dmstats
.de CMD_UNGROUP
. ad l
. nh
2016-06-22 20:36:10 +03:00
. BR ungroup
. RI [ device_name | \fB--alldevices ]
. RB [ --groupid
2016-06-22 20:36:10 +03:00
. IR id ]
. hy
2016-06-22 20:36:10 +03:00
. ad b
..
.CMD_UNGROUP
.HP
.B dmstats
.de CMD_UPDATE_FILEMAP
. ad l
. nh
. BR update_filemap
. IR file_path
. RB [ --groupid
. IR id ]
. RB [ --follow
. IR follow_mode ]
. OPT_FOREGROUND
. hy
. ad b
..
.CMD_UPDATE_FILEMAP
.
.PD
.ad b
.
.SH DESCRIPTION
.
The dmstats program manages IO statistics regions for devices that use
the device-mapper driver. Statistics regions may be created, deleted,
listed and reported on using the tool.
.P
The first argument to dmstats is a \fIcommand\fP.
.P
The second argument is the \fIdevice name\fP,
\fIuuid\fP or \fImajor\fP and \fIminor\fP numbers.
.P
Further options permit the selection of regions, output format
control, and reporting behaviour.
.P
When no device argument is given dmstats will by default operate on all
device-mapper devices present. The \fBcreate\fP and \fBdelete\fP
commands require the use of \fB--alldevices\fP when used in this way.
.
.SH OPTIONS
.
.TP
.B --alias \fIname
2016-06-22 20:36:10 +03:00
Specify an alias name for a group.
.
.TP
.B --alldevices
If no device arguments are given allow operation on all devices when
creating or deleting regions.
.
.TP
.B --allprograms
Include regions from all program IDs for list and report operations.
.
.TP
.B --allregions
Include all present regions for commands that normally accept a single
region identifier.
.
.TP
.B --area
When performing a list or report, include objects of type area in the
results.
.
.TP
.B --areas \fInr_areas
Specify the number of statistics areas to create within a new region.
.
.TP
.B --areasize \fIarea_size\fR[\c
.UNITS
Specify the size of areas into which a new region should be divided. An
optional suffix selects units of:
.HELP_UNITS
.
.TP
.B --clear
When printing statistics counters, also atomically reset them to zero.
.
.TP
.B --count \fIcount
Specify the iteration count for repeating reports. If the count
argument is zero reports will continue to repeat until interrupted.
.
.TP
.B --group
When performing a list or report, include objects of type group in the
results.
.
.TP
.B --filemap
Instead of creating regions on a device as specified by command line
options, open the file found at each \fBfile_path\fP argument, and
create regions corresponding to the locations of the on-disk extents
allocated to the file(s).
.
.TP
.B --nomonitor
Disable the \fBdmfilemapd\fP daemon when creating new file mapped
groups. Normally the device-mapper filemap monitoring daemon,
\fBdmfilemapd\fP, is started for each file mapped group to update the
set of regions as the file changes on-disk: use of this option
disables this behaviour.
.sp
Regions in the group may still be updated with the
\fBupdate_filemap\fP command, or by starting the daemon manually.
.
.TP
.B --follow \fIfollow_mode
Specify the \fBdmfilemapd\fP file following mode. The file map
monitoring daemon can monitor files in two distinct ways: the mode
affects the behaviour of the daemon when a file under monitoring is
renamed or unlinked, and the conditions which cause the daemon to
terminate.
.sp
The \fBfollow_mode\fP argument is either "inode", for follow-inode
mode, or "path", for follow-path.
.sp
If follow-inode mode is used, the daemon will hold the file open, and
continue to update regions from the same file descriptor. This means
that the mapping will follow rename, move (within the same file
system), and unlink operations. This mode is useful if the file is
expected to be moved, renamed, or unlinked while it is being
monitored.
.sp
In follow-inode mode, the daemon will exit once it detects that the
file has been unlinked and it is the last holder of a reference to it.
.sp
If follow-path is used, the daemon will re-open the provided path on
each monitoring iteration. This means that the group will be updated
to reflect a new file being moved to the same path as the original
file. This mode is useful for files that are expected to be updated
via unlink and rename.
.sp
In follow-path mode, the daemon will exit if the file is removed and
not replaced within a brief tolerance interval.
.sp
In either mode, the daemon exits automatically if the monitored group
is removed.
.
.TP
.B --foreground
Specify that the \fBdmfilemapd\fP daemon should run in the foreground.
The daemon will not fork into the background, and will replace the
\fBdmstats\fP command that started it.
.
.TP
.B --groupid \fIid
2016-06-22 20:36:10 +03:00
Specify the group to operate on.
.
.TP
.B --bounds \fIhistogram_boundaries\c
.RB [ ns | us | ms | s ]
Specify the boundaries of a latency histogram to be tracked for the
region as a comma separated list of latency values. Latency values are
given in nanoseconds. An optional unit suffix of
.BR ns , us , ms ,
or \fBs\fP may be given after each value to specify units of
nanoseconds, microseconds, milliseconds or seconds respectively.
.
.TP
.B --histogram
When used with the \fBreport\fP and \fBlist\fP commands select default
fields that emphasize latency histogram data.
.
.TP
.B --interval \fIseconds
Specify the interval in seconds between successive iterations for
repeating reports. If \fB--interval\fP is specified but
\fB--count\fP is not,
reports will continue to repeat until interrupted.
.
.TP
.B --length \fIlength\fR[\c
.UNITS
Specify the length of a new statistics region in sectors. An optional
suffix selects units of:
.HELP_UNITS
.
.TP
.BR -j | --major " " \fImajor
Specify the major number.
.
.TP
.BR -m | --minor " " \fIminor
Specify the minor number.
.
.TP
.B --nogroup
When creating regions mapping the extents of a file in the file
system, do not create a group or set an alias.
.
.TP
.B --nosuffix
Suppress the suffix on output sizes. Use with \fB--units\fP
(except h and H) if processing the output.
.
.TP
.B --notimesuffix
Suppress the suffix on output time values. Histogram boundary values
will be reported in units of nanoseconds.
.
.TP
.BR -o | --options
Specify which report fields to display.
.
.TP
.BR -O | --sort " " \fIsort_fields
Sort output according to the list of fields given. Precede any
sort field with '\fB-\fP' for a reverse sort on that column.
.
.TP
.B --precise
Attempt to use nanosecond precision counters when creating new
statistics regions.
.
.TP
.B --programid \fIid
Specify a program ID string. When creating new statistics regions this
string is stored with the region. Subsequent operations may supply a
program ID in order to select only regions with a matching value. The
default program ID for dmstats-managed regions is "dmstats".
.
.TP
.B --region
When performing a list or report, include objects of type region in the
results.
.
.TP
.B --regionid \fIid
Specify the region to operate on.
.
.TP
.B --regions \fIregion_list
Specify a list of regions to group. The group list is a comma-separated
2016-06-22 20:36:10 +03:00
list of region identifiers. Continuous sequences of identifiers may be
expressed as a hyphen separated range, for example: '1-10'.
.
.TP
.B --relative
If displaying the histogram report show relative (percentage) values
instead of absolute counts.
.
.TP
.BR -S | --select " " \fIselection
Display only rows that match \fIselection\fP criteria. All rows with the
additional "selected" column (\fB-o selected\fP) showing 1 if the row matches
the \fIselection\fP and 0 otherwise. The selection criteria are defined by
specifying column names and their valid values while making use of
supported comparison operators.
.
.TP
.B --start \fIstart\fR[\c
.UNITS
Specify the start offset of a new statistics region in sectors. An
optional suffix selects units of:
.HELP_UNITS
.
.TP
.B --segments
When used with \fBcreate\fP, create a new statistics region for each
target contained in the given device(s). This causes a separate region
to be allocated for each segment of the device.
.sp
The newly created regions are automatically placed into a group unless
the \fB--nogroup\fP option is given. When grouping is enabled a group
alias may be specified using the \fB--alias\fP option.
.
.TP
.B --units \c
.RI [ units ] \c
.RB [ h | H | \c
.UNITS
Set the display units for report output.
All sizes are output in these units:
.RB ( h )uman-readable,
.HELP_UNITS
Can also specify custom units e.g. \fB--units\ 3M\fP.
.
.TP
.B --userdata \fIuser_data
Specify user data (a word) to be stored with a new region. The value
2020-10-03 14:52:37 +03:00
is added to any internal auxiliary data (for example, group
information), and stored with the region in the aux_data field provided
by the kernel. Whitespace is not permitted.
.
.TP
.BR -u | --uuid
Specify the uuid.
.
.TP
.BR -v | --verbose \ [ -v | --verbose ]
Produce additional output.
.
.SH COMMANDS
.
.HP
.CMD_CLEAR
.br
Instructs the kernel to clear statistics counters for the specified
regions (with the exception of in-flight IO counters).
.
.HP
.CMD_CREATE
.br
Creates one or more new statistics regions on the specified device(s).
.sp
The region will span the entire device unless \fB--start\fP and
\fB--length\fP or \fB--segments\fP are given. The \fB--start\fP an
\fB--length\fP options allow a region of arbitrary length to be placed
at an arbitrary offset into the device. The \fB--segments\fP option
causes a new region to be created for each target in the corresponding
device-mapper device's table.
.sp
If the \fB--precise\fP option is used the command will attempt to
create a region using nanosecond precision counters.
.sp
If \fB--bounds\fP is given a latency histogram will be tracked for
the new region. The boundaries of the histogram bins are given as a
comma separated list of latency values. There is an implicit lower bound
of zero on the first bin and an implicit upper bound of infinity (or the
configured interval duration) on the final bin.
.sp
Latencies are given in nanoseconds. An optional unit suffix of ns, us,
ms, or s may be given after each value to specify units of nanoseconds,
microseconds, milliseconds or seconds respectively, so for example, 10ms
is equivalent to 10000000. Latency values with a precision of less than
one millisecond can only be used when precise timestamps are enabled: if
\fB--precise\fP is not given and values less than one millisecond are
used it will be enabled automatically.
.sp
An optional \fBprogram_id\fP or \fBuser_data\fP string may be associated
with the region. A \fBprogram_id\fP may then be used to select regions
for subsequent list, print, and report operations. The \fBuser_data\fP
stores an arbitrary string and is not used by dmstats or the
device-mapper kernel statistics subsystem.
.sp
By default dmstats creates regions with a \fBprogram_id\fP of
"dmstats".
.sp
On success the \fBregion_id\fP of the newly created region is printed
to stdout.
.sp
If the \fB--filemap\fP option is given with a regular file, or list
of files, as the \fBfile_path\fP argument, instead of creating regions
with parameters specified on the command line, \fBdmstats\fP will open
the files located at \fBfile_path\fP and create regions corresponding to
the physical extents allocated to the file. This can be used to monitor
statistics for individual files in the file system, for example, virtual
machine images, swap areas, or large database files.
.sp
To work with the \fB--filemap\fP option, files must be located on a
local file system, backed by a device-mapper device, that supports
physical extent data using the FIEMAP ioctl (Ext4 and XFS for e.g.).
.sp
By default regions that map a file are placed into a group and the
group alias is set to the basename of the file. This behaviour can be
overridden with the \fB--alias\fP and \fB--nogroup\fP options.
.sp
Creating a group that maps a file automatically starts a daemon,
\fBdmfilemapd\fP to monitor the file and update the mapping as the
extents allocated to the file change. This behaviour can be disabled
using the \fB--nomonitor\fP option.
.sp
Use the \fB--group\fP option to only display information for groups
when listing and reporting.
.
.HP
.CMD_DELETE
.br
Delete the specified statistics region. All counters and resources used
by the region are released and the region will not appear in the output
of subsequent list, print, or report operations.
.sp
All regions registered on a device may be removed using
\fB--allregions\fP.
.sp
To remove all regions on all devices both \fB--allregions\fP and
\fB--alldevices\fP must be used.
.sp
If a \fB--groupid\fP is given instead of a \fB--regionid\fP the
2016-06-22 20:36:10 +03:00
command will attempt to delete the group and all regions that it
contains.
.sp
2016-06-22 20:36:10 +03:00
If a deleted region is the first member of a group of regions the group
will also be removed.
.
.HP
.CMD_GROUP
.br
Combine one or more statistics regions on the specified device into a
group.
.sp
The list of regions to be grouped is specified with \fB--regions\fP
and an optional alias may be assigned with \fB--alias\fP. The set of
regions is given as a comma-separated list of region identifiers. A
continuous range of identifiers spanning from \fBR1\fP to \fBR2\fP may
be expressed as '\fBR1\fP-\fBR2\fP'.
.sp
Regions that have a histogram configured can be grouped: in this case
the number of histogram bins and their bounds must match exactly.
.sp
2016-06-22 20:36:10 +03:00
On success the group list and newly created \fBgroup_id\fP are
printed to stdout.
.sp
The group metadata is stored with the first (lowest numbered)
\fBregion_id\fP in the group: deleting this region will also delete
the group and other group members will be returned to their prior
state.
.
.HP
.CMD_HELP
.br
Outputs a summary of the commands available, optionally including
the list of report fields.
.
.HP
.CMD_LIST
.br
2016-06-22 20:36:10 +03:00
List the statistics regions, areas, or groups registered on the device.
If the \fB--allprograms\fP switch is given all regions will be listed
regardless of region program ID values.
.sp
2016-06-22 20:36:10 +03:00
By default only regions and groups are included in list output. If
\fB-v\fP or \fB--verbose\fP is given the report will also include a
2016-06-22 20:36:10 +03:00
row of information for each configured group and for each area contained
in each region displayed.
.sp
Regions that contain a single area are by default omitted from the
verbose list since their properties are identical to the area that they
contain - to view all regions regardless of the number of areas present
use \fB--region\fP). To also view the areas contained within regions
use \fB--area\fP.
.sp
If \fB--histogram\fP is given the report will include the bin count
and latency boundary values for any configured histograms.
.
.HP
.CMD_PRINT
.br
Print raw statistics counters for the specified region or for all
present regions.
.
.HP
.CMD_REPORT
.br
2016-06-22 20:36:10 +03:00
Start a report for the specified object or for all present objects. If
the count argument is specified, the report will repeat at a fixed
interval set by the \fB--interval\fP option. The default interval is
one second.
.sp
If the \fB--allprograms\fP switch is given, all regions will be
listed, regardless of region program ID values.
.sp
If the \fB--histogram\fP is given the report will include the histogram
values and latency boundaries.
.sp
If the \fB--relative\fP is used the default histogram field displays
bin values as a percentage of the total number of I/Os.
.sp
Object types (areas, regions and groups) to include in the report are
selected using the \fB--area\fP, \fB--region\fP, and \fB--group\fP
options.
.
2016-06-22 20:36:10 +03:00
.HP
.CMD_UNGROUP
.br
Remove an existing group and return all the group's regions to their
original state.
.sp
The group to be removed is specified using \fB--groupid\fP.
.
.HP
.CMD_UPDATE_FILEMAP
.br
Update a group of \fBdmstats\fP regions specified by \fBgroup_id\fP,
that were previously created with \fB--filemap\fP, either directly,
or by starting the monitoring daemon, \fBdmfilemapd\fP.
.sp
This will add and remove regions to reflect changes in the allocated
extents of the file on-disk, since the time that it was created or last
updated.
.sp
Use of this command is not normally needed since the \fBdmfilemapd\fP
daemon will automatically monitor filemap groups and perform these
updates when required.
.sp
If a filemapped group was created with \fB--nomonitor\fP, or the
daemon has been killed, the \fBupdate_filemap\fP can be used to
manually force an update or start a new daemon.
.sp
Use \fB--nomonitor\fP to force a direct update and disable starting
the monitoring daemon.
2016-06-22 20:36:10 +03:00
.
.SH REGIONS, AREAS, AND GROUPS
.
The device-mapper statistics facility allows separate performance
counters to be maintained for arbitrary regions of devices. A region may
span any range: from a single sector to the whole device. A region may
be further sub-divided into a number of distinct areas (one or more),
2016-06-22 20:36:10 +03:00
each with its own counter set. In this case a summary value for the
entire region is also available for use in reports.
.P
2016-06-22 20:36:10 +03:00
In addition, one or more regions on one device can be combined into
a statistics group. Groups allow several regions to be aggregated and
reported as a single entity; counters for all regions and areas are
summed and used to report totals for all group members. Groups also
permit the assignment of an optional alias, allowing meaningful names
to be associated with sets of regions.
.P
The group metadata is stored with the first (lowest numbered)
\fBregion_id\fP in the group: deleting this region will also delete
the group and other group members will be returned to their prior
state.
.P
By default new regions span the entire device. The \fB--start\fP and
\fB--length\fP options allows a region of any size to be placed at any
location on the device.
.P
2016-06-22 20:36:10 +03:00
Using offsets it is possible to create regions that map individual
objects within a block device (for example: partitions, files in a file
system, or stripes or other structures in a RAID volume). Groups allow
several non-contiguous regions to be assembled together for reporting
2016-06-22 20:36:10 +03:00
and data aggregation.
.P
A region may be either divided into the specified number of equal-sized
areas, or into areas of the given size by specifying one of
\fB--areas\fP or \fB--areasize\fP when creating a region with the
\fBcreate\fP command. Depending on the size of the areas and the device
region the final area within the region may be smaller than requested.
.
.SS Region identifiers
.
Each region is assigned an identifier when it is created that is used to
reference the region in subsequent operations. Region identifiers are
unique within a given device (including across different \fBprogram_id\fP
values).
.P
Depending on the sequence of create and delete operations, gaps may
exist in the sequence of \fBregion_id\fP values for a particular device.
.P
2016-06-22 20:36:10 +03:00
The \fBregion_id\fP should be treated as an opaque identifier used to
reference the region.
.
.SS Group identifiers
.
2016-06-22 20:36:10 +03:00
Groups are also assigned an integer identifier at creation time;
like region identifiers, group identifiers are unique within the
containing device.
.P
2016-06-22 20:36:10 +03:00
The \fBgroup_id\fP should be treated as an opaque identifier used to
reference the group.
.
.SH FILE MAPPING
.
Using \fB--filemap\fP, it is possible to create regions that
correspond to the extents of a file in the file system. This allows
IO statistics to be monitored on a per-file basis, for example to
observe large database files, virtual machine images, or other files
of interest.
.P
To be able to use file mapping, the file must be backed by a
device-mapper device, and in a file system that supports the FIEMAP
ioctl (and which returns data describing the physical location of
extents). This currently includes \fBxfs(5)\fP and \fBext4(5)\fP.
.P
By default the regions making up a file are placed together in a
group, and the group alias is set to the \fBbasename(3)\fP of the
file. This allows statistics to be reported for the file as a whole,
aggregating values for the regions making up the group. To see only
the whole file (group) when using the \fBlist\fP and \fBreport\fP
commands, use \fB--group\fP.
.P
Since it is possible for the file to change after the initial
group of regions is created, the \fBupdate_filemap\fP command, and
\fBdmfilemapd\fP daemon are provided to update file mapped groups
either manually or automatically.
.
.SS File follow modes
.
The file map monitoring daemon can monitor files in two distinct ways:
follow-inode mode, and follow-path mode.
.P
The mode affects the behaviour of the daemon when a file under
monitoring is renamed or unlinked, and the conditions which cause the
daemon to terminate.
.P
If follow-inode mode is used, the daemon will hold the file open, and
continue to update regions from the same file descriptor. This means
that the mapping will follow rename, move (within the same file
system), and unlink operations. This mode is useful if the file is
expected to be moved, renamed, or unlinked while it is being
monitored.
.P
In follow-inode mode, the daemon will exit once it detects that the
file has been unlinked and it is the last holder of a reference to it.
.P
If follow-path is used, the daemon will re-open the provided path on
each monitoring iteration. This means that the group will be updated
to reflect a new file being moved to the same path as the original
file. This mode is useful for files that are expected to be updated
via unlink and rename.
.P
In follow-path mode, the daemon will exit if the file is removed and
not replaced within a brief tolerance interval (one second).
.P
To stop the daemon, delete the group containing the mapped regions:
the daemon will automatically shut down.
.P
The daemon can also be safely killed at any time and the group kept:
if the file is still being allocated the mapping will become
progressively out-of-date as extents are added and removed (in this
case the daemon can be re-started or the group updated manually with
the \fBupdate_filemap\fP command).
.P
See the \fBcreate\fP command and \fB--filemap\fP, \fB--follow\fP,
and \fB--nomonitor\fP options for further information.
.
.SS Limitations
.
The daemon attempts to maintain good synchronization between the file
extents and the regions contained in the group, however, since it can
only react to new allocations once they have been written, there are
inevitably some IO events that cannot be counted when a file is
growing, particularly if the file is being extended by a single thread
writing beyond end-of-file (for example, the \fBdd\fP program).
.P
There is a further loss of events in that there is currently no way
to atomically resize a \fBdmstats\fP region and preserve its current
counter values. This affects files when they grow by extending the
final extent, rather than allocating a new extent: any events that
had accumulated in the region between any prior operation and the
resize are lost.
.P
File mapping is currently most effective in cases where the majority
of IO does not trigger extent allocation. Future updates may address
these limitations when kernel support is available.
.
.SH REPORT FIELDS
.
The dmstats report provides several types of field that may be added to
the default field set, or used to create custom reports.
.P
All performance counters and metrics are calculated per-area.
.
.SS Derived metrics
.
A number of metrics fields are included that provide high level
performance indicators. These are based on the fields provided by the
conventional Linux iostat program and are derived from the basic counter
values provided by the kernel for each area.
.TP
.B reads_merged_per_sec
Reads merged per second.
.TP
.B writes_merged_per_sec
Writes merged per second.
.TP
.B reads_per_sec
Reads completed per second.
.TP
.B writes_per_sec
Writes completed per second.
.TP
.B read_size_per_sec
Size of data read per second.
.TP
.B write_size_per_sec
Size of data written per second.
.TP
.B avg_request_size
Average request size.
.TP
.B queue_size
Average queue size.
.TP
.B await
The average wait time for read and write operations.
.TP
.B r_await
The average wait time for read operations.
.TP
.B w_await
The average wait time for write operations.
.TP
.B throughput
The device throughput in operations per second.
.TP
.B service_time
The average service time (in milliseconds) for operations issued
to the device.
.TP
.B util
Percentage of CPU time during which I/O requests were issued to the
device (bandwidth utilization for the device). Device saturation occurs
when this value is close to 100%.
.
2016-06-22 20:36:10 +03:00
.SS Group, region and area meta fields
.
2016-06-22 20:36:10 +03:00
Meta fields provide information about the groups, regions, or areas that
the statistics values relate to. This includes the region and area
identifier, start, length, and counts, as well as the program ID and
user data values.
.TP
.B region_id
Region identifier. This is a non-negative integer returned by the kernel
when a statistics region is created.
.TP
.B region_start
The region start location. Display units are selected by the
\fB--units\fP option.
.TP
.B region_len
The length of the region. Display units are selected by the
\fB--units\fP option.
.TP
.B area_id
Area identifier. Area identifiers are assigned by the device-mapper
statistics library and uniquely identify each area within a region. Each
ID corresponds to a distinct set of performance counters for that area
of the statistics region. Area identifiers are always monotonically
increasing within a region so that higher ID values correspond to
greater sector addresses within the area and no gaps in the sequence of
identifiers exist.
.TP
.B area_start
The area start location. Display units are selected by the
\fB--units\fP option.
.TP
.B area_len
The length of the area. Display units are selected by the
\fB--units\fP option.
.TP
.B area_count
The number of areas in this region.
.TP
.B program_id
The program ID value associated with this region.
.TP
.B user_data
The user data value associated with this region.
.TP
2016-06-22 20:36:10 +03:00
.B group_id
Group identifier. This is a non-negative integer returned by the dmstats
2016-06-22 20:36:10 +03:00
\fBgroup\fP command when a statistics group is created.
.TP
.B interval_ns
The estimated interval over which the current counter values have
2020-10-03 14:52:37 +03:00
accumulated. The value is reported as an integer expressed in units
of nanoseconds.
.TP
.B interval
The estimated interval over which the current counter values have
accumulated. The value is reported as a real number in units of
seconds.
.
.SS Basic counters
.
Basic counters provide access to the raw counter data from the kernel,
allowing further processing to be carried out by another program.
.P
The kernel provides thirteen separate counters for each statistics
area. The first eleven of these match the counters provided in
/proc/diskstats or /sys/block/*/*/stat. The final pair provide separate
counters for read and write time.
.TP
.B read_count
Count of reads completed this interval.
.TP
.B reads_merged_count
Count of reads merged this interval.
.TP
.B read_sector_count
Count of 512 byte sectors read this interval.
.TP
.B read_time
Accumulated duration of all read requests (ns).
.TP
.B write_count
Count of writes completed this interval.
.TP
.B writes_merged_count
Count of writes merged this interval.
.TP
.B write_sector_count
Count of 512 byte sectors written this interval.
.TP
.B write_time
Accumulated duration of all write requests (ns).
.TP
.B in_progress_count
Count of requests currently in progress.
.TP
.B io_ticks
Nanoseconds spent servicing requests.
.TP
.B queue_ticks
This field is incremented at each I/O start, I/O completion, I/O merge,
or read of these stats by the number of I/Os in progress multiplied by
the number of milliseconds spent doing I/O since the last update of this
field. This can provide an easy measure of both I/O completion time and
the backlog that may be accumulating.
.TP
.B read_ticks
Nanoseconds spent servicing reads.
.TP
.B write_ticks
Nanoseconds spent servicing writes.
.
.SS Histogram fields
.
Histograms measure the frequency distribution of user specified I/O
latency intervals. Histogram bin boundaries are specified when a region
is created.
.P
A brief representation of the histogram values and latency intervals can
be included in the report using these fields.
.TP
.B hist_count
A list of the histogram counts for the current statistics area in order
of ascending latency value. Each value represents the number of I/Os
with latency times falling into that bin's time range during the sample
period.
.TP
.B hist_count_bounds
A list of the histogram counts for the current statistics area in order
of ascending latency value including bin boundaries: each count is
prefixed by the lower bound of the corresponding histogram bin.
.TP
.B hist_count_ranges
A list of the histogram counts for the current statistics area in order
of ascending latency value including bin boundaries: each count is
prefixed by both the lower and upper bounds of the corresponding
histogram bin.
.TP
.B hist_percent
A list of the relative histogram values for the current statistics area
in order of ascending latency value, expressed as a percentage. Each
value represents the proportion of I/Os with latency times falling into
that bin's time range during the sample period.
.TP
.B hist_percent_bounds
A list of the relative histogram values for the current statistics area
in order of ascending latency value, expressed as a percentage and
including bin boundaries. Each value represents the proportion of I/Os
with latency times falling into that bin's time range during the sample
period and is prefixed with the corresponding bin's lower bound.
.TP
.B hist_percent_ranges
A list of the relative histogram values for the current statistics area
in order of ascending latency value, expressed as a percentage and
including bin boundaries. Each value represents the proportion of I/Os
with latency times falling into that bin's time range during the sample
period and is prefixed with the corresponding bin's lower and upper
bounds.
.TP
.B hist_bounds
A list of the histogram boundary values for the current statistics area
in order of ascending latency value. The values are expressed in whole
units of seconds, milliseconds, microseconds or nanoseconds with a suffix
indicating the unit.
.TP
.B hist_ranges
A list of the histogram bin ranges for the current statistics area in
order of ascending latency value. The values are expressed as
"LOWER-UPPER" in whole units of seconds, milliseconds, microseconds or
nanoseconds with a suffix indicating the unit.
.TP
.B hist_bins
The number of latency histogram bins configured for the area.
.
.SH EXAMPLES
.
Create a whole-device region with one area on vg00/lvol1
.br
#
.B dmstats create vg00/lvol1
.br
2015-08-16 21:30:16 +03:00
vg00/lvol1: Created new region with 1 area(s) as region ID 0
.P
Create a 32M region 1G into device d0
.br
#
.B dmstats create --start 1G --length 32M d0
.br
2015-08-16 21:30:16 +03:00
d0: Created new region with 1 area(s) as region ID 0
.P
Create a whole-device region with 8 areas on every device
.br
.br
#
.B dmstats create --areas 8
.br
vg00-lvol1: Created new region with 8 area(s) as region ID 0
.br
vg00-lvol2: Created new region with 8 area(s) as region ID 0
.br
vg00-lvol3: Created new region with 8 area(s) as region ID 0
.br
vg01-lvol0: Created new region with 8 area(s) as region ID 2
.br
vg01-lvol1: Created new region with 8 area(s) as region ID 0
.br
vg00-lvol2: Created new region with 8 area(s) as region ID 1
.P
Delete all regions on all devices
.br
.br
#
.B dmstats delete --alldevices --allregions
.P
Create a whole-device region with areas 10 GiB in size on vg00/lvol1
using dmsetup
.br
.br
#
.B dmsetup stats create --areasize 10G vg00/lvol1
.br
vg00-lvol1: Created new region with 5 area(s) as region ID 1
.P
Create a 1 GiB region with 16 areas at the start of vg00/lvol1
.br
#
.B dmstats create --start 0 --len 1G --areas=16 vg00/lvol1
.br
vg00-lvol1: Created new region with 16 area(s) as region ID 0
.P
List the statistics regions registered on vg00/lvol1
.br
#
.B dmstats list vg00/lvol1
.br
2015-08-16 21:30:16 +03:00
Name RgID RStart RSize #Areas ASize ProgID
.br
2015-08-16 21:30:16 +03:00
vg00-lvol1 0 0 61.00g 1 61.00g dmstats
.br
2015-08-16 21:30:16 +03:00
vg00-lvol1 1 61.00g 19.20g 1 19.20g dmstats
.br
2015-08-16 21:30:16 +03:00
vg00-lvol1 2 80.20g 2.14g 1 2.14g dmstats
.P
Display five statistics reports for vg00/lvol1 at an interval of one second
.br
.br
#
.B dmstats report --interval 1 --count 5 vg00/lvol1
.br
#
.B dmstats report
.br
2015-08-16 21:30:16 +03:00
Name RgID ArID AStart ASize RRqM/s WRqM/s R/s W/s RSz/s WSz/s AvRqSz QSize Util% AWait RdAWa WrAWa
.br
vg_hex-lv_home 0 0 0 61.00g 0.00 0.00 0.00 218.00 0 1.04m 4.50k 2.97 81.70 13.62 0.00 13.62
.br
vg_hex-lv_home 1 0 61.00g 19.20g 0.00 0.00 0.00 5.00 0 548.00k 109.50k 0.14 11.00 27.40 0.00 27.40
.br
vg_hex-lv_home 2 0 80.20g 2.14g 0.00 0.00 0.00 14.00 0 1.15m 84.00k 0.39 18.70 27.71 0.00 27.71
.P
Create one region for reach target contained in device vg00/lvol1
.br
.br
#
.B dmstats create --segments vg00/lvol1
.br
vg00-lvol1: Created new region with 1 area(s) as region ID 0
.br
vg00-lvol1: Created new region with 1 area(s) as region ID 1
.br
vg00-lvol1: Created new region with 1 area(s) as region ID 2
.P
Create regions mapping each file in the directory images/ and place
them into separate groups, each named after the corresponding file
.br
#
.B dmstats create --filemap images/*
.br
images/vm1.qcow2: Created new group with 87 region(s) as group ID 0.
.br
images/vm1-1.qcow2: Created new group with 8 region(s) as group ID 87.
.br
images/vm2.qcow2: Created new group with 11 region(s) as group ID 95.
.br
images/vm2-1.qcow2: Created new group with 1454 region(s) as group ID 106.
.br
images/vm3.img: Created new group with 2 region(s) as group ID 1560.
.P
Print raw counters for region 4 on device d0
.br
#
.B dmstats print --regionid 4 d0
.br
2097152+65536 0 0 0 0 29 0 264 701 0 41 701 0 41
.
.SH AUTHORS
.
Bryn M. Reeves <bmr@redhat.com>
.
.SH SEE ALSO
.
.BR dmsetup (8)
.P
LVM2 resource page:
.UR https://www.sourceware.org/lvm2
.UE
.br
Device-mapper resource page:
.UR http://sources.redhat.com/dm
.UE
.P
Device-mapper statistics kernel documentation
.br
.I Documentation/device-mapper/statistics.txt