nvme-discoverd(8)
=================

NAME
----
nvme-discoverd - NVMe-oF discovery and connectivity daemon

SYNOPSIS
--------
[verse]
'nvme-discoverd' [--config=<FILE>] [--nvme-path=<PATH>] [--debug] [--help]

DESCRIPTION
-----------
nvme-discoverd is a persistent daemon that connects the host's desired
NVMe-oF controllers and keeps them connected. It reads three sources to
build its set of desired connections: the NBFT ACPI table (boot-time
controllers), the shared NVMe-oF fabrics configuration
(nvme-fabrics.conf(5)), and the Discovery Log Pages of any Discovery
Controllers it connects to (including FC targets found via FC
kickstart). It never disconnects a live controller in response to a
discovery change; the only disconnects it causes are a side effect of
stopping its own systemd units, at shutdown or on request.

Every connection attempt is delegated to 'nvme connect', which runs
inside a systemd transient unit instead of writing directly to
/dev/nvme-fabrics. As a result, a stuck connect attempt never blocks
the daemon's event loop. Each unit is named after a hash of the
controller's transport identity.

Failed connections retry with exponential backoff. Discovery
Controllers from NBFT or the fabrics config represent explicit intent,
so they always retry forever. A Discovery Controller found only
through a referral or FC kickstart is different: it gives up after
'dc-giveup-hours' (default 72) hours of unbroken failure and is
dropped from tracking.

Before connecting or reconnecting anything, nvme-discoverd checks the
system-wide exclusion list. It also registers ownership of every
controller it connects (--owner discoverd, or --owner nbft for an
NBFT-sourced controller) in the NVMe-oF ownership registry. See
nvme-registry-list(1) and nvme-exclusion-add(1).

nvme-discoverd is still experimental. It is meant to replace the
older udev-rule/systemd-unit autoconnect mechanism (built when the
'nvmf-autoconnect' meson option is enabled). For now, nvme-discoverd
can run alongside the legacy udev rules -- the goal is to eventually
eliminate them completely. Switching a host over is a deliberate
administrator choice ('systemctl enable --now nvme-discoverd',
alongside disabling the legacy units), not something either
mechanism decides automatically.

OPTIONS
-------
-c <FILE>::
--config=<FILE>::
	Path to the daemon's own configuration file. Default:
	@SYSCONFDIR@/nvme/nvme-discoverd.conf; see CONFIGURATION below.
	A missing file is not an error -- every knob keeps its default.

--nvme-path=<PATH>::
	Absolute or relative path to the 'nvme' binary that transient
	connection units exec. Default: @SBINDIR@/nvme. Mainly useful for
	running nvme-discoverd against a locally built 'nvme' instead of
	the installed one.

-d::
--debug::
	Enable debug-level logging for both nvme-discoverd and the
	libnvme calls it makes, overriding the configured debug-level.

-h::
--help::
	Print usage and exit.

CONFIGURATION
-------------
nvme-discoverd.conf holds only the daemon's own knobs, in a single
'[Global]' section:

	[Global]
	nbft = true
	debug-level = info
	fc-kickstart-interval-minutes = 0
	dc-giveup-hours = 72

'nbft' (boolean, default true) controls whether NBFT-listed controllers
are adopted and reconnected. 'debug-level' (err, warn, info, or debug;
default info) sets the log threshold; '--debug' on the command line
overrides it. 'fc-kickstart-interval-minutes' (default 0, disabled)
re-probes the FC fabric on a timer. This is independent of the
kickstart nvme-discoverd already issues at startup and on every FC
controller drop -- use it to detect FC equipment replacement sooner
than the next connection event. 'dc-giveup-hours' (default 72) is the
give-up horizon described above; 0 makes a dynamically-discovered DC
retry forever, same as a static/NBFT-sourced one.

The connections nvme-discoverd manages come entirely from the shared
fabrics configuration, not from this file: which Discovery Controllers
and subsystems to connect, host identity, per-connection parameters.
See nvme-fabrics.conf(5).

SIGNALS
-------
SIGHUP;;
	Reload both nvme-discoverd.conf and the shared fabrics configuration.
	Newly desired connections are started; nothing already connected
	is ever disconnected by a reload. If the fabrics configuration
	fails to reload (e.g. a syntax error), the previous one is kept
	and the failure is logged.

SIGTERM, SIGINT;;
	Shut down gracefully.

FILES
-----
@SYSCONFDIR@/nvme/nvme-discoverd.conf::
	The daemon's own configuration; see CONFIGURATION.

@SYSCONFDIR@/nvme/nvme-fabrics.conf, @SYSCONFDIR@/nvme/nvme-fabrics.conf.d/*.conf::
	The connections nvme-discoverd manages; see nvme-fabrics.conf(5).

@RUNDIR@/nvme/discoverd/::
	Runtime state linking each connected controller to the systemd
	transient unit that owns it. Does not survive a reboot.

SEE ALSO
--------
nvme-fabrics.conf(5)
nvme-connect(1)
nvme-registry-list(1)
nvme-exclusion-add(1)

NVME
----
Part of the nvme-user suite
