|
HTML rendering created 2025-09-06 by Michael Kerrisk, author of The Linux Programming Interface. For details of in-depth Linux/UNIX system programming training courses that I teach, look here. Hosting by jambit GmbH. |
|
|
NAME | SYNOPSIS | DESCRIPTION | INSTALLATION | CONFIGURATION | METRICS | EXAMPLES | FILES | PCP ENVIRONMENT | DEBUGGING OPTIONS | SEE ALSO | COLOPHON |
|
|
|
PMDADM(1) General Commands Manual PMDADM(1)
pmdadm - Device Mapper PMDA
$PCP_PMDAS_DIR/dm/pmdadm [-D debug]
pmdadm is a Performance Metrics Domain Agent (PMDA) which exports
metric values for Device Mapper on the local system.
This PMDA collects its data through the dmsetup(8) utility and the
dmstats API and requires that the program is installed in order to
function. In addition, at least one statistics region must be
created using the dmstats(8) utility in order to get basic counter
values. See below for examples.
Note that device-mapper statistics collection is not enabled by
default and is not persistent across reboots (unless a system
administrator has configured something to run in rc.local or
equivalent). This is because there are overheads associated with
statistics collection. In addition, the pmdadm PMDA does not
automatically enable any statistics collection because it may not
be the only consumer of the data.
Install the DM PMDA by using the Install script as root:
# cd $PCP_PMDAS_DIR/dm
# ./Install
To uninstall, the following must be done as root:
# cd $PCP_PMDAS_DIR/dm
# ./Remove
pmdadm is launched by pmcd(1) and should never be executed
directly. The Install and Remove scripts notify pmcd when the
agent is installed or removed.
This PMDA uses the dmstats API (the userspace device-mapper
support library) to capture device-mapper performance data.
Using this PMDA, users need the userspace device-mapper support
library (libdevmapper) and the userspace command line tool
dmstats(8). Before device-mapper metrics will be available for
PCP, statistics regions have to be created. For example, to
enable basic statistics for all local device-mapper logical
devices, use the following command:
# dmstats create --alldevices
To create statistics regions with specified histogram boundaries,
use the following:
# dmstats create --alldevices --bounds histogram_boundaries
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 ns,
us, ms, or s may be given after each value to specify units of
nanoseconds, microseconds, miliseconds or seconds respectively.
For example,
# dmstats create --alldevices --bounds 10us,30us,50us,70us,90us
Further details and more complex examples can be found in
dmstats(8).
Basic Counters
Basic counters provide access to the raw counter data from the
kernel, allowing further processing to be carried out by
another program.
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.
dmstats.read
Count of reads completed this interval per device-mapper
device.
dmstats.reads_merged
Count of reads merged this interval per device-mapper
device.
dmstats.read_bytes
Count of kbytes read this interval per device-mapper
device.
dmstats.reads_time
Accumulated duration of all read requests per device-
mapper device.
dmstats.write
Count of writes completed this interval per device-mapper
device.
dmstats.writes_merged
Count of writes completed this interval per device-mapper
device.
dmstats.write_bytes
Count of kbytes write this interval per device-mapper
device.
dmstats.writes_time
Accumulated duration of all write requests per device-
mapper device.
dmstats.in_progress
Count of requests currently in progress per device-mapper
device.
dmstats.io_ticks
Nanoseconds spent servicing request per device-mapper
device.
dmstats.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
nanoseconds 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.
dmstats.read_ticks
Nanoseconds spent servicing reads per device-mapper
device.
dmstats.write_ticks
Nanoseconds spent servicing writes per device-mapper
device.
Histogram fields
Histograms measure the frequency distribution of user
specified I/O latency intervals. Histogram bin boundaries are
specified when a region is created.
Instance name represents devicename, region id and bin
boundaries.
dmstats.histogram.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.
dmstats.histogram.hist.bins
The number of latency histogram bins configured for the
area.
# dmstats create looptest0 --bounds 10us,30us,50us
looptest0: Created new region with 1 area(s) as region ID 0
# pminfo -f dmstats.read dmstats.histogram.hist_count
dmstats.read
inst [0 or "looptest0"] value 4099
dmstats.histogram.hist_count
inst [0 or "looptest0:0:0s"] value 1
inst [1 or "looptest0:0:10us"] value 3752
inst [2 or "looptest0:0:30us"] value 250
inst [3 or "looptest0:0:50us"] value 96
$PCP_PMDAS_DIR/dm/help
default help text file for the dm metrics
$PCP_PMDAS_DIR/dm/Install
installation script for the pmdadm agent
$PCP_PMDAS_DIR/dm/Remove
undo installation script for the pmdadm agent
$PCP_LOG_DIR/pmcd/dm.log
default log file for error messages from pmdadm
Environment variables with the prefix PCP_ are used to
parameterize the file and directory names used by PCP. On each
installation, the file /etc/pcp.conf contains the local values for
these variables. The $PCP_CONF variable may be used to specify an
alternative configuration file, as described in pcp.conf(5).
The -D or --debug option enables the output of additional
diagnostics on stderr to help triage problems, although the
information is sometimes cryptic and primarily intended to provide
guidance for developers rather end-users. debug is a comma
separated list of debugging options; use pmdbg(1) with the -l
option to obtain a list of the available debugging options and
their meaning.
PCPIntro(1), pmcd(1), pmstore(1), dmsetup(8) and dmstats(8).
This page is part of the PCP (Performance Co-Pilot) project.
Information about the project can be found at
⟨http://www.pcp.io/⟩. If you have a bug report for this manual
page, send it to pcp@groups.io. This page was obtained from the
project's upstream Git repository
⟨https://github.com/performancecopilot/pcp.git⟩ on 2025-08-11.
(At that time, the date of the most recent commit that was found
in the repository was 2025-08-11.) If you discover any rendering
problems in this HTML version of the page, or you believe there is
a better or more up-to-date source for the page, or you have
corrections or improvements to the information in this COLOPHON
(which is not part of the original manual page), send a mail to
man-pages@man7.org
Performance Co-Pilot PCP PMDADM(1)
Pages that refer to this page: pcp-dmcache(1)
|
HTML rendering created 2025-09-06 by Michael Kerrisk, author of The Linux Programming Interface. For details of in-depth Linux/UNIX system programming training courses that I teach, look here. Hosting by jambit GmbH. |
|