Network UPS Tools Overview
==========================
// NOTE: No blank line here, document-header include processing should kick in!
//GH_MARKUP_1095//ifdef::top_srcdir[]
//GH_MARKUP_1095//include::{top_srcdir}docs/asciidoc-vars.conf[]
//GH_MARKUP_1095//endif::top_srcdir[]
//GH_MARKUP_1095//ifndef::top_srcdir[]
//GH_MARKUP_1095//include::docs/asciidoc-vars.conf[]
//GH_MARKUP_1095//endif::top_srcdir[]
//GH_MARKUP_1095_INCLUDE_BEGIN//7c5e90132 (2023-09-13) docs/asciidoc-vars.conf: fence against re-definitions of website-url and (top_)(src|build)dir attributes
ifndef::asciidoc-vars-nut-included[]
:asciidoc-vars-nut-included:	true
// NOTE: The big block of comments and definitions below comes from
// NUT::docs/asciidoc-vars.conf and is included into top-level document
// sources by maintenance recipes directly (`make maintainer-asciidocs`),
// due to current limitations of the GitHub Web UI asciidoc renderer.
// Hopefully it can be dropped in favor of compact include definitions
// (see README.adoc for anticipated example) after this issue is resolved
// on their side:
// * https://github.com/github/markup/issues/1095
//
// This file should be included into NUT documentation sources to consistently
// define certain expandable attributes, with contents defined based on the
// rendition target (e.g. GitHub Web UI, plain text, locally built HTML/PDF...)
// Note that currently GitHub Web UI references lead to nut-website (as of
// last built and published revision), not to neighboring documents in the
// source browser (which would make sense for branch revisions, etc.) due
// to certain complexity about referencing other-document sections with a
// partially functional rendering engine there. Exploration and fixes are
// welcome (actually working links like
// https://github.com/networkupstools/nut/tree/master#installing or
// https://github.com/networkupstools/nut/blob/master/UPGRADING.adoc#changes-from-274-to-280
// do seem promising)!
//
// Since the GitHub UI does not allow use of custom asciidoc configuration
// files, or generally does not process the `include:` requests at this time,
// clumsy expandable attributes had to be used (usually a set including a
// prefix with meaningful name, and one or more separators and/or a suffix
// with shortened names). For our classic documentation renditions, they
// should resolve to properly defined macros from `docs/asciidoc.conf`
// (usually named same as the variables defined here, for simplicity):
// * `linkdoc` allows to refer to a file under `docs/` directory (or
//   its nut-website rendition).
// * `xref` substitutes the asciidoc shorthand '<< >>' syntax with
//   attributes that conditionally expand to:
//   - links on GitHub (references can point at most to a section of
//     level docs/common.xsl's <chunk.section.depth>), or
//   - xref asciidoc macros when generating docs.
// * `linksingledoc` guarantees that, when chunked HTML is generated,
//   the link always points to a non-chunked file.
// * `linkman2` allows to support different names for the manpage and
//   the command shown. This is also needed to properly display links
//   to manpages in both GitHub and generated docs without defining an
//   attribute for each manpage.
//
// Optional attributes set by callers:
// * `website-url` (defaulted below) may be used for "historic website"
//   snapshot builds... hopefully
// * `website` is used as a boolean toggle in our recipes for nut-website
//   vs. offline documentation renditions
// * `env-github` is used as a boolean toggle, set by GitHub Web-UI renderer
// * `(top_)srcdir` and `(top_)builddir` can be set by `Makefile.am`
//   calling the `a2x` tool, since some of the files with the asciidoc
//   mark-up are only generated or post-processed during build and
//   (due to `make dist` restrictions) being build products, they may
//   not reside in same directory as static source text files which
//   reference or include them. Note that the non-`top` paths would
//   normally differ based on location of the `Makefile` involved
//   (e.g. workspace root, or the `docs`, or `docs/man` directories).
//   These variables are expected to be absolute paths, or ones relative
//   to asciidoc-selected `:base_dir`, and to end with a relevant path
//   separator, or be empty -- so in all cases letting the resulting
//   string resolve meaningfully in the filesystem during docs build.
//
// Please keep the remaining comments and definitions as one big block
// so it does not become a series of empty paragraphs in the rendered
// documents!
//
ifndef::website-url[]
:website-url:	https://www.networkupstools.org/
endif::website-url[]
//
ifndef::srcdir[]
:srcdir:
endif::srcdir[]
//
ifndef::builddir[]
:builddir:
endif::builddir[]
//
ifndef::top_srcdir[]
:top_srcdir:
endif::top_srcdir[]
//
ifndef::top_builddir[]
:top_builddir:
endif::top_builddir[]
//
//
// Address links on GitHub vs docs
// (note: 'env-github' attribute is set on GitHub)
//
// - when generating docs:
ifndef::env-github[]
//   * xref -> xref
//     syntax: {xref}<id>{x-s}[<caption>]
//     -> xref:<id>[<caption>]
:xref:		xref:
:x-s:
//   * link to doc -> our macro
//     syntax: {linkdoc}<document>{ld-s}[<display title>]
//     -> linkdoc:<document>[<display title>]
:linkdoc:	linkdoc:
:ld-s:
//   * link to single doc -> our macro
//     syntax: {linksingledoc}<document>{lsd-s}[<display title>]
//     -> linksingledoc:<document>[<display title>]
:linksingledoc:	linksingledoc:
:lsd-s:
//   * link to manpage -> our macro
//     syntax: {linkman2}<command-page>{lm-s}<displayed-command>{lm-c}<manpage-section>{lm-e}
//     -> linkman2:<command-page>[<displayed-command>,<manpage-section>]
:linkman2:	linkman2:
:lm-s:		[
:lm-c:		,
:lm-e:		]
endif::env-github[]
//
// - on GitHub:
ifdef::env-github[]
//     In our normal builds, Makefile variables convey the needed paths
//     (used relatively below as `image:images/ci/...png` etc.)
:imagesdir:	docs
//   * xref -> link
//     syntax: {xref}<id>{x-s}[<caption>]
//     In order for it to work, <id> can reference at most a section of
//     level docs/common.xsl's <chunk.section.depth>
//     -> {website-url}docs/user-manual.chunked/<id>.html[<caption>]
:xref:		{website-url}docs/user-manual.chunked/
:x-s:		.html
//   * link to doc -> link
//     syntax: {linkdoc}<document>{ld-s}[<display title>]
//     -> {website-url}docs/<document>.chunked/index.html[<display title>]
:linkdoc:	{website-url}docs/
:ld-s:		.chunked/index.html
//   * link to single doc -> link
//     syntax: {linksingledoc}<document>{lsd-s}[<display title>]
//     -> {website-url}docs/<document>.html[<display title>]
:linksingledoc:	{website-url}docs/
:lsd-s:		.html
//   * link to manpage -> link
//     syntax: {linkman2}<command-page>{lm-s}<displayed-command>{lm-c}<manpage-section>{lm-e}
//     All the fields are mandatory.
//     -> {website-url}docs/man/<command-page>.html[<displayed-command>(<manpage-section>)]
:linkman2:	{website-url}docs/man/
:lm-s:		.html[
:lm-c:		(
:lm-e:		)]
endif::env-github[]
endif::asciidoc-vars-nut-included[]
//
//GH_MARKUP_1095_INCLUDE_END//


Description
-----------

Network UPS Tools is a collection of programs which provide a common
interface for monitoring and administering UPS, PDU and SCD hardware.
It uses a layered approach to connect all of the parts.

Drivers are provided for a wide assortment of equipment.  They
understand the specific language of each device and map it back to a
compatibility layer.  This means both an expensive high end UPS, a simple
"power strip" PDU, or any other power device can be handled transparently
with a uniform management interface.

This information is cached by the network server `upsd`, which then
answers queries from the clients.  upsd contains a number of access
control features to limit the abilities of the clients.  Only authorized
hosts may monitor or control your hardware if you wish.  Since the
notion of monitoring over the network is built into the software, you
can hang many systems off one large UPS, and they will all shut down
together. You can also use NUT to power on, off or cycle your data center
nodes, individually or globally through PDU outlets.

Clients such as `upsmon` check on the status of the hardware and do things
when necessary.  The most important task is shutting down the operating
system cleanly before the UPS runs out of power.  Other programs are
also provided to log information regularly, monitor status through your
web browser, and more.


NUT and the ecosystem
---------------------

NUT comes pre-packaged for many operating systems and embedded in storage,
automation or virtualization appliances, and is also often shipped as the
software companion by several UPS vendors. Of course, it is quite normal
and supported to build your own -- whether for an operating system which
lacks it yet, or for an older distribution which lacks the current NUT
version; whether to take advantage of new features or to troubleshoot a
new UPS deployment with a debugger in hand.

Given its core position at the heart of your systems' lifecycle, we make
it a point to have current NUT building and running anywhere, especially
where older releases did work before (including "abandonware" like the
servers and OSes from the turn of millennium): if those boxes are still
alive and in need of power protection, they should be able to get it.

[TIP]
=====
If you like how the NUT project helps protect your systems from power
outages, please consider sponsoring or at least "starring" it on GitHub at
https://github.com/networkupstools/nut/ - these stars are among metrics
which the larger potential sponsors consider when choosing how to help
FOSS projects. Keeping the lights shining in such a large non-regression
build matrix is a big undertaking!

See <<acknowledgements-ci-ops,acknowledgements of organizations which help
with NUT CI and other daily operations>> for an overview of the shared effort.
=====

As a FOSS project, for over a quarter of a century we welcome contributions
of both core code (drivers and other features), build recipes and other
integration elements to make it work on your favourite system, documentation
revisions to make it more accessible to newcomers, as well as hardware vendor
cooperation with first-hand driver and protocol submissions, and just about
anything else you can think of.


Installing
----------

If you are installing these programs for the first time, go read the
{xref}_installation_instructions{x-s}[installation instructions]
to find out how to do that.  This document contains more information
on what all of this stuff does.


Upgrading
---------

When upgrading from an older version, always check the
{xref}Upgrading_notes{x-s}[upgrading notes] to see what may have
changed.  Compatibility issues and other changes will be listed there
to ease the process.


Configuring and using
---------------------

Once NUT is installed, refer to the
{xref}Configuration_notes{x-s}[configuration notes] for directions.


Documentation
-------------

This is just an overview of the software.  You should read the man pages,
included example configuration files, and auxiliary documentation for the
parts that you intend to use.


Network Information
-------------------

These programs are designed to share information over the network.  In
the examples below, `localhost` is used as the hostname.  This can also
be an IP address or a fully qualified domain name.  You can specify a
port number if your upsd process runs on another port.

In the case of the program `upsc`, to view the variables on the UPS called
sparky on the `upsd` server running on the local machine, you'd do this:

	/usr/local/ups/bin/upsc sparky@localhost

The default port number is 3493.  You can change this with
"configure --with-port" at compile-time.  To make a client talk to upsd
on a specific port, add it after the hostname with a colon, like this:

	/usr/local/ups/bin/upsc sparky@localhost:1234

This is handy when you have a mixed environment and some of the systems
are on different ports.

The general form for UPS identifiers is this:

	<upsname>[@<hostname>[:<port>]]

Keep this in mind when viewing the examples below.


Manifest
--------

This package is broken down into several categories:

- *drivers*	- These programs talk directly to your UPS hardware.
- *server*	- upsd serves data from the drivers to the network.
- *clients*	- They talk to upsd and do things with the status data.
- *cgi-bin*	- Special class of clients that you can use with your web server.
- *scripts*	- Contains various scripts, like the Perl and Python binding,
integration bits and applications.

Drivers
-------

These programs provide support for specific UPS models.  They understand
the protocols and port specifications which define status information
and convert it to a form that upsd can understand.

To configure drivers, edit ups.conf.  For this example, we'll have a UPS
called "sparky" that uses the apcsmart driver and is connected to
`/dev/ttyS1`.  That's the second serial port on most Linux-based systems.
The entry in `ups.conf` looks like this:

	[sparky]
		driver = apcsmart
		port = /dev/ttyS1

To start and stop drivers, use upsdrvctl of upsdrvsvcctl (installed on
operating systems with a service management framework supported by NUT).
By default, it will start or stop every UPS in the config file:

	/usr/local/ups/sbin/upsdrvctl start
	/usr/local/ups/sbin/upsdrvctl stop

However, you can also just start or stop one by adding its name:

	/usr/local/ups/sbin/upsdrvctl start sparky
	/usr/local/ups/sbin/upsdrvctl stop sparky

On operating systems with a supported service management framework,
you might wrap your NUT drivers into individual services instances
with:

	/usr/local/ups/sbin/upsdrvsvcctl resync

and then manage those service instances with commands like:

	/usr/local/ups/sbin/upsdrvsvcctl start sparky
	/usr/local/ups/sbin/upsdrvsvcctl stop sparky

To find the driver name for your device, refer to the section below
called "HARDWARE SUPPORT TABLE".

Extra Settings
~~~~~~~~~~~~~~

Some drivers may require additional settings to properly communicate
with your hardware.  If it doesn't detect your UPS by default, check the
driver's man page or help (-h) to see which options are available.

For example, the usbhid-ups driver allows you to use USB serial numbers to
distinguish between units via the "serial" configuration option.  To use this
feature, just add another line to your ups.conf section for that UPS:

	[sparky]
		driver = usbhid-ups
		port = auto
		serial = 1234567890

Hardware Compatibility List
~~~~~~~~~~~~~~~~~~~~~~~~~~~

The {xref}HCL{x-s}[Hardware Compatibility List] is available in the source directory
('nut-X.Y.Z/data/driver.list'), and is generally distributed with packages.
For example, it is available on Debian systems as:

	/usr/share/nut/driver.list

This table is also available link:{website-url}stable-hcl.html[online].


If your driver has vanished, see the {linksingledoc}FAQ{lsd-s}[FAQ] and
{xref}Upgrading_notes{x-s}[Upgrading notes].

Generic Device Drivers
~~~~~~~~~~~~~~~~~~~~~~

NUT provides several generic drivers that support a variety of very similar
models.

- The `genericups` driver supports many serial models that use the same basic
principle to communicate with the computer.  This is known as "contact
closure", and basically involves raising or lowering signals to indicate
power status.
+
This type of UPS tends to be cheaper, and only provides the very simplest
data about power and battery status.  Advanced features like battery
charge readings and such require a "smart" UPS and a driver which
supports it.
+
See the {linkman2}genericups{lm-s}genericups{lm-c}8{lm-e} man page for more information.

- The `usbhid-ups` driver attempts to communicate with USB HID Power Device
Class (PDC) UPSes. These units generally implement the same basic protocol,
with minor variations in the exact set of supported attributes. This driver
also applies several correction factors when the UPS firmware reports values
with incorrect scale factors.
+
See the {linkman2}usbhid-ups{lm-s}usbhid-ups{lm-c}8{lm-e} man page for more information.

- The `nutdrv_qx` driver supports the Megatec / Q1 protocol that is used in
many brands (Blazer, Energy Sistem, Fenton Technologies, Mustek, Voltronic
Power and many others).
+
See the {linkman2}nutdrv_qx{lm-s}nutdrv_qx{lm-c}8{lm-e} man page for more information.

- The `snmp-ups` driver handles various SNMP enabled devices, from many
different manufacturers. In SNMP terms, `snmp-ups` is a manager, that
monitors SNMP agents.
+
See the {linkman2}snmp-ups{lm-s}snmp-ups{lm-c}8{lm-e} man page for more information.

- The `powerman-pdu` is a bridge to the PowerMan daemon, thus handling all
PowerMan supported devices. The PowerMan project supports several serial
and networked PDU, along with Blade and IPMI enabled servers.
+
See the {linkman2}powerman-pdu{lm-s}powerman-pdu{lm-c}8{lm-e} man page for more
information.

- The `apcupsd-ups` driver is a bridge to the Apcupsd daemon, thus handling
all Apcupsd supported devices. The Apcupsd project supports many serial,
USB and networked APC UPS.
+
See the {linkman2}apcupsd-ups{lm-s}apcupsd-ups{lm-c}8{lm-e} man page for more information.

UPS Shutdowns
~~~~~~~~~~~~~

upsdrvctl can also shut down (power down) all of your UPS hardware.

WARNING: if you play around with this command, expect your filesystems
to die.  Don't power off your computers unless they're ready for it:

	/usr/local/ups/sbin/upsdrvctl shutdown
	/usr/local/ups/sbin/upsdrvctl shutdown sparky

You should read the {xref}UPS_shutdown{x-s}[Configuring automatic UPS shutdowns]
chapter to learn more about when to use this feature.  If called at the wrong
time, you may cause data loss by turning off a system with a filesystem
mounted read-write.

Power distribution unit management
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

NUT also provides an advanced support for power distribution units.

You should read the
{xref}outlet_management{x-s}[NUT outlets management and PDU notes]
chapter to learn more about when to use this feature.

Network Server
--------------

`upsd` is responsible for passing data from the drivers to the client
programs via the network.  It should be run immediately after `upsdrvctl`
in your system's startup scripts.

`upsd` should be kept running whenever possible, as it is the only source
of status information for the monitoring clients like `upsmon`.


Monitoring client
-----------------

`upsmon` provides the essential feature that you expect to find in UPS
monitoring software: safe shutdowns when the power fails.

In the layered scheme of NUT software, it is a client.  It has this
separate section in the documentation since it is so important.

You configure it by telling it about UPSes that you want to monitor in
upsmon.conf.  Each UPS can be defined as one of two possible types:
a "primary" or "secondary".

Primary
~~~~~~~

The monitored UPS possibly supplies power to this system running `upsmon`,
but more importantly -- this system can manage the UPS (typically, this
instance of `upsmon` runs on the same system as the `upsd` and driver(s)):
it is capable and responsible for shutting it down when the battery is
depleted (or in another approach, lingering to deplete it or to tell the
UPS to reboot its load after too much time has elapsed and this system
is still alive -- meaning wall power returned at a  "wrong" moment).

The shutdown of this (primary) system itself, as well as eventually an
UPS shutdown, occurs after any secondary systems ordered to shut down
first have disconnected, or a critical urgency threshold was passed.

If your UPS is plugged directly into a system's serial or USB port, the
`upsmon` process on that system should define its relation to that UPS
as a primary. It may be more complicated for higher-end UPSes with a
shared network management capability (typically via SNMP) or several
serial/USB ports that can be used simultaneously, and depends on what
vendors and drivers implement. Setups with several competing primaries
(for redundancy) are technically possible, if each one runs its own
full stack of NUT, but results can be random (currently NUT does not
provide a way to coordinate several entities managing the same device).

For a typical home user, there's one computer connected to one UPS.
That means you would run on the same computer the whole NUT stack --
a suitable driver, `upsd`, and `upsmon` in primary mode.

Secondary
~~~~~~~~~

The monitored UPS may supply power to the system running `upsmon` (or
alternatively, it may be a monitoring station with zero PSUs fed by
that UPS), but more importantly, this system can't manage the UPS --
e.g. shut it down directly (through a locally running NUT driver).

Use this mode when you run multiple computers on the same UPS.
Obviously, only one can be connected to the serial or USB port
on a typical UPS, and that system is the primary. Everything
else is a secondary.

For a typical home user, there's one computer connected to one UPS.
That means you run a driver, `upsd`, and `upsmon` in primary mode.

Additional Information
~~~~~~~~~~~~~~~~~~~~~~

More information on configuring upsmon can be found in these places:

- The {linkman2}upsmon{lm-s}upsmon{lm-c}8{lm-e} man page
- {xref}BigServers{x-s}[Typical setups for big servers]
- {xref}UPS_shutdown{x-s}[Configuring automatic UPS shutdowns] chapter
- The stock `upsmon.conf` that comes with the package


Clients
-------

Clients talk to upsd over the network and do useful things with the data
from the drivers.  There are tools for command line access, and a few
special clients which can be run through your web server as CGI
programs.

For more details on specific programs, refer to their man pages.

upsc
~~~~

`upsc` is a simple client that will display the values of variables known
to `upsd` and your UPS drivers.  It will list every variable by default,
or just one if you specify an additional argument.  This can be useful
in shell scripts for monitoring something without writing your own
network code.

`upsc` is a quick way to find out if your driver(s) and upsd are working
together properly.  Just run `upsc <ups>` to see what's going on, i.e.:

	morbo:~$ upsc sparky@localhost
	ambient.humidity: 035.6
	ambient.humidity.alarm.maximum: NO,NO
	ambient.humidity.alarm.minimum: NO,NO
	ambient.temperature: 25.14
	...

If you are interested in writing a simple client that monitors `upsd`,
the source code for `upsc` is a good way to learn about using the
upsclient functions.

See the {linkman2}upsc{lm-s}upsc{lm-c}8{lm-e} man page and
{xref}nut-names{x-s}[NUT command and variable naming scheme] for more information.

upslog
~~~~~~

`upslog` will write status information from `upsd` to a file at set
intervals.  You can use this to generate graphs or reports with other
programs such as `gnuplot`.

upsrw
~~~~~

`upsrw` allows you to display and change the read/write variables in your
UPS hardware.  Not all devices or drivers implement this, so this may
not have any effect on your system.

A driver that supports read/write variables will give results like this:

	$ upsrw sparky@localhost

	( many skipped )

	[ups.test.interval]
	Interval between self tests
	Type: ENUM
	Option: "1209600"
	Option: "604800" SELECTED
	Option: "0"

	( more skipped )

On the other hand, one that doesn't support them won't print anything:

	$ upsrw fenton@gearbox

	( nothing )

`upsrw` requires administrator powers to change settings in the hardware.
Refer to {linkman2}upsd.users{lm-s}upsd.users{lm-c}5{lm-e} for information on defining
users in `upsd`.

upscmd
~~~~~~

Some UPS hardware and drivers support the notion of an instant command -
a feature such as starting a battery test, or powering off the load.
You can use upscmd to list or invoke instant commands if your
hardware/drivers support them.

Use the -l command to list them, like this:

	$ upscmd -l sparky@localhost
	Instant commands supported on UPS [sparky@localhost]:

	load.on - Turn on the load immediately
	test.panel.start - Start testing the UPS panel
	calibrate.start - Start run time calibration
	calibrate.stop - Stop run time calibration
	...

`upscmd` requires administrator powers to start instant commands.
To define users and passwords in `upsd`, see
{linkman2}upsd.users{lm-s}upsd.users{lm-c}5{lm-e}.


CGI Programs
------------

The CGI programs are clients that run through your web server.  They
allow you to see UPS status and perform certain administrative commands
from any web browser.  Javascript and cookies are not required.

These programs are not installed or compiled by default.  To compile
and install them, first run `configure --with-cgi`, then do `make` and
`make install`.  If you receive errors about "gd" during configure, go
get it and install it before continuing.

You can get the source here:

- http://www.libgd.org/

In the event that you need libpng or zlib in order to compile gd,
they can be found at these URLs:

- http://www.libpng.org/pub/png/pngcode.html
- http://www.zlib.net/


Access Restrictions
~~~~~~~~~~~~~~~~~~~

The CGI programs use hosts.conf to see if they are allowed to talk to a
host.  This keeps malicious visitors from creating queries from your web
server to random hosts on the Internet.

If you get error messages that say "Access to that host is not
authorized", you're probably missing an entry in your hosts.conf.

upsstats
~~~~~~~~

`upsstats` generates web pages from HTML templates, and plugs in status
information in the right places.  It looks like a distant relative of
APC's old Powerchute interface.  You can use it to monitor several
systems or just focus on one.

It also can generate IMG references to `upsimage`.

upsimage
~~~~~~~~

This is usually called by upsstats via IMG SRC tags to draw either the
utility or outgoing voltage, battery charge percent, or load percent.

upsset
~~~~~~

`upsset` provides several useful administration functions through a web
interface.  You can use `upsset` to kick off instant commands on your UPS
hardware like running a battery test.  You can also use it to change
variables in your UPS that accept user-specified values.

Essentially, `upsset` provides the functions of `upsrw` and `upscmd`, but
with a happy pointy-clicky interface.

`upsset` will not run until you convince it that you have secured your
system.  You *must* secure your CGI path so that random interlopers
can't run this program remotely.  See the `upsset.conf` file.  Once you
have secured the directory, you can enable this program in that
configuration file.  It is not active by default.


Version Numbering
-----------------

The version numbers work like this: if the middle number is odd, it's a
development tree, otherwise it is the stable tree.

The past stable trees were 1.0, 1.2, 1.4, 2.0, 2.2 and 2.4, with the
latest such stable tree designated 2.6.  The development trees were 1.1,
1.3, 1.5, 2.1 and 2.3.  Since the 2.4 release, there is no real separate
development branch anymore since the code is available through a revision
control system (namely, Git -- or actually Subversion back then) and its
snapshots become published releases.

Since 2.7 line of releases, sources are tracked in Git revision control
system, with the project ecosystem being hosted on GitHub, and any code
improvements or other contributions merged through common pull request
approach and custom NUT CI testing on multiple platforms.

Major release jumps are mostly due to large changes to the features
list.  There have also been a number of architectural changes which
may not be noticeable to most users, but which can impact developers.

Backwards and Forwards Compatibility
------------------------------------

The network protocol for the current version of NUT should be
backwards-compatible all the way back to version 1.4. A newer client should
fail gracefully when querying an older server.

If you need more details about cross-compatibility of older NUT releases
(1.x vs. 2.x), please see the {xref}Project_History{x-s}[Project history] chapter.

Support / Help / etc.
---------------------

If you are in need of help, refer to the
{xref}Support_Request{x-s}[Support instructions] in the user manual.


Hacking / Development Info
--------------------------

Additional documentation can be found in:

- the {linkdoc}developer-guide{ld-s}[Developer Guide],
- the {linkdoc}packager-guide{ld-s}[Packager Guide].


Acknowledgements / Contributions
--------------------------------

The many people who have participated in creating and improving NUT are
listed in the user manual {xref}Acknowledgements{x-s}[acknowledgements appendix].

[[acknowledgements-ci-ops]]

We would like to highlight some organizations which provide continuous
support to the NUT project (and many other FOSS projects) on technological
and organizational sides, such as helping keep the donations transparent,
NUT CI farm afloat, and public resources visible. Thanks for keeping the
clocks ticking, day and night:

////////////
FIXME: Use different (better-resolution) images for PDF rendering?

FIXME: PDF cells seem to align weirdly, like setting the bottom of the first
line of text to be on the same level as bottom of the image, or similar to that.

NOTE: GitHub renderer (or CSS stack?) ignores style settings and squashes the
logo column into a fixed-width monster with either our specified heights, or
with teeny-tiny thumbnail magnitude images, so it is prettier to leave it as
a "single-column table" by default. Grid/Frame settings are also ignored, but
we can try our best anyway.

NOTE: The classic asciidoc/a2x renderer seems to not support link/url options,
but at least does not complain about them either.
////////////

ifndef::env-github[]
[frame="none",grid="none",cols="^.<1,<.<2"]
endif::env-github[]
ifdef::env-github[]
[frame="none",grid="none",cols="<1*"]
endif::env-github[]
|===
| image:images/ci/GitHub-Mark-140pxW.png[alt="GitHub logo",width="140",height="140",link="https://github.com/"]
| The link:https://github.com/networkupstools/["NetworkUPSTools" organization
  on GitHub] arranges a lot of things, including source code hosting for NUT
  itself and several related projects, team management, projects, issue and
  pull request discussions, sponsorship, nut-website rendering and hosting,
  some automated actions, and more...

| image:images/ci/jenkins-nut-transparent-bg-140pxW.png[alt="Jenkins and NUT logo",width="139",height="104",link="https://www.jenkins.io/"]
| The link:https://www.jenkins.io/[Jenkins CI] project and its huge plugin
  ecosystem provides the technological foundation for the largest island of
  the link:https://ci.networkupstools.org/[self-hosted NUT CI farm].
  There is a fair amount of cross-pollination between the upstream project
  and community, and the development done originally for the NUT CI farm.

  See more at link:https://stories.jenkins.io/user-story/jenkins-is-the-way-for-networkupstools/[Jenkins
  is the way to build multi-platform NUT] article.

| image:images/ci/fosshost_org_Host_Light_38px.png[alt="Fosshost logo",width="112",height="38"]
| Fosshost provided virtual machines where the multi-platform NUT CI farm with
  a link:https://github.com/networkupstools/jenkins-dynamatrix/[jenkins-dynamatrix]
  link:https://github.com/networkupstools/nut/blob/master/Jenkinsfile-dynamatrix[setup]
  runs to arrange builds in numerous operating environments and a lot of toolkit
  versions and implementations. Some workers running on NUT community members'
  machines can also dial in to provide an example of their favourite platforms.
  Literally hundreds of NUT builds run for each iteration, to make sure NUT can
  always build and work everywhere.

  This allows us to ensure that NUT remains portable across two decades' worth
  of operating systems, compilers, script interpreters, tools and third-party
  dependencies.

| image:images/ci/CircleCI_vertical_black_logo.png[alt="CircleCI logo",width="130",height="107",link="https://circleci.com/"]
| The
  link:https://app.circleci.com/pipelines/github/networkupstools/nut/[CircleCI
  NUT pipeline] allows us to test NUT CI builds on MacOS.

| image:images/ci/AppVeyor_logo-ar21.png[alt="AppVeyor logo",width="120",height="60",link="https://www.appveyor.com/"]
| The link:https://ci.appveyor.com/project/nut-travis/nut/[AppVeyor
  NUT pipeline] allows us to test NUT CI builds on Windows (and publish
  preview tarballs with binaries).

| image:images/ci/DO_Powered_by_Badge_blue_140pxW.png[alt="DigitalOcean logo",width="140",height="29",link="https://www.digitalocean.com/?refcode=d2fbf2b9e082&utm_campaign=Referral_Invite&utm_medium=Referral_Program&utm_source=badge"]
| The link:https://www.digitalocean.com/?refcode=d2fbf2b9e082&utm_campaign=Referral_Invite&utm_medium=Referral_Program&utm_source=badge[DigitalOcean]
  droplets allow us to host NUT CI farm Jenkins controller and the build agents
  for multiple operating systems.

| image:images/ci/gandi-ar21.png[alt="Gandi.Net logo",width="120",height="60",link="https://www.gandi.net/"]
| link:https://www.gandi.net/[Gandi.Net] took up the costs of NUT DNS hosting.

| image:images/ci/OC_logo_merged_140x26.png[alt="Open Collective logo",width="140",height="26",link="https://opencollective.com/"]
| https://opencollective.com/networkupstools allows us to arrange monetary
  donations and spending, with public transparency of everything that happens.
|===