'\" t
.\"     Title: nvme-discoverd
.\"    Author: [FIXME: author] [see http://www.docbook.org/tdg5/en/html/author]
.\" Generator: DocBook XSL Stylesheets vsnapshot <http://docbook.sf.net/>
.\"      Date: 10/30/2026
.\"    Manual: NVMe Manual
.\"    Source: NVMe
.\"  Language: English
.\"
.TH "NVME\-DISCOVERD" "8" "10/30/2026" "NVMe" "NVMe Manual"
.\" -----------------------------------------------------------------
.\" * Define some portability stuff
.\" -----------------------------------------------------------------
.\" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.\" http://bugs.debian.org/507673
.\" http://lists.gnu.org/archive/html/groff/2009-02/msg00013.html
.\" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.ie \n(.g .ds Aq \(aq
.el       .ds Aq '
.\" -----------------------------------------------------------------
.\" * set default formatting
.\" -----------------------------------------------------------------
.\" disable hyphenation
.nh
.\" disable justification (adjust text to left margin only)
.ad l
.\" -----------------------------------------------------------------
.\" * MAIN CONTENT STARTS HERE *
.\" -----------------------------------------------------------------
.SH "NAME"
nvme-discoverd \- NVMe\-oF discovery and connectivity daemon
.SH "SYNOPSIS"
.sp
.nf
\fInvme\-discoverd\fR [\-\-config=<FILE>] [\-\-nvme\-path=<PATH>] [\-\-debug] [\-\-help]
.fi
.SH "DESCRIPTION"
.sp
nvme\-discoverd is a persistent daemon that connects the host\(cqs 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\&.
.sp
At shutdown, a unit is stopped only after the file systems mounted with the \fI_netdev\fR option are unmounted\&. A file system on an NVMe\-oF namespace must use \fI_netdev\fR in fstab(5)\&. Without it, the controller can be disconnected while the file system is still mounted\&.
.sp
Every connection attempt is delegated to \fInvme connect\fR, 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\(cqs event loop\&. Each unit is named after a hash of the controller\(cqs transport identity\&.
.sp
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 \fIdc\-giveup\-timeout\fR (default 72hours) of unbroken failure and is dropped from tracking\&.
.sp
A TLS connection needs tlshd(8) from ktls\-utils\&. Enable it with \fIsystemctl enable \-\-now tlshd\fR\&. When both are enabled, nvme\-discoverd starts after tlshd\&.
.sp
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)\&.
.sp
A Discovery Controller that has no owner, for example one connected with \fInvme discover \-\-persistent\fR, is not tracked\&. When its discovery log changes, or when it reconnects, nvme\-discoverd runs \fInvme connect\-all\fR for it in a transient unit, as the legacy udev rules do\&. The connections it makes have no owner, and \fInvme disconnect\-all\fR removes them\&. Events from a Discovery Controller with another owner are ignored\&.
.sp
nvme\-discoverd is still experimental\&. It is meant to replace the older udev\-rule/systemd\-unit autoconnect mechanism\&. By default, the legacy mechanism is built only when nvme\-discoverd is not\&. Building with \fI\-Dnvmf\-autoconnect=enabled\fR installs both\&. The legacy udev rules are active as soon as they are installed\&. nvme\-discoverd runs only when the administrator enables it with \fIsystemctl enable \-\-now nvme\-discoverd\fR\&. While nvme\-discoverd is active, nvmf\-autoconnect\&.service and the FC kickstart skip their connect step\&.
.SH "OPTIONS"
.PP
\-c <FILE>, \-\-config=<FILE>
.RS 4
Path to the daemon\(cqs own configuration file\&. Default: /usr/local/etc/nvme/nvme\-discoverd\&.conf; see CONFIGURATION below\&. A missing file is not an error \(em every knob keeps its default\&.
.RE
.PP
\-\-nvme\-path=<PATH>
.RS 4
Absolute or relative path to the
\fInvme\fR
binary that transient connection units exec\&. Default: /usr/local/sbin/nvme\&. Mainly useful for running nvme\-discoverd against a locally built
\fInvme\fR
instead of the installed one\&.
.RE
.PP
\-d, \-\-debug
.RS 4
Enable debug\-level logging for both nvme\-discoverd and the libnvme calls it makes, overriding the configured debug\-level\&.
.RE
.PP
\-h, \-\-help
.RS 4
Print usage and exit\&.
.RE
.SH "CONFIGURATION"
.sp
nvme\-discoverd\&.conf holds the daemon\(cqs own knobs, in a \fI[Global]\fR section and a \fI[Discovery]\fR section:
.sp
.if n \{\
.RS 4
.\}
.nf
[Global]
nbft = true
debug\-level = info
.fi
.if n \{\
.RE
.\}
.sp
.if n \{\
.RS 4
.\}
.nf
[Discovery]
epcsd\-poll\-interval\-minutes = 15
fc\-kickstart\-interval\-minutes = 0
dc\-giveup\-timeout = 72hours
zeroconf = false
.fi
.if n \{\
.RE
.\}
.sp
\fInbft\fR (boolean, default true) adopts and connects controllers listed in the NBFT ACPI table\&. \fIdebug\-level\fR (err, warn, info, or debug; default info) sets the log threshold; \fI\-\-debug\fR on the command line overrides it\&.
.sp
\fIepcsd\-poll\-interval\-minutes\fR (default 15, minimum 1) sets how often, in minutes, to reconnect and re\-check a DC whose effective EPCSD (Explicit Persistent Connection Support for Discovery) is 0\&. Such a DC is disconnected between checks\&. \fIfc\-kickstart\-interval\-minutes\fR (default 0) re\-issues FC kickstart every N minutes; 0 = only at startup and on every FC controller drop\&. \fIdc\-giveup\-timeout\fR (default 72hours) is the give\-up horizon described above\&. It takes a systemd time span ("72hours", "3 days 5 hours") or a unit\-less value in seconds; "infinity" = never give up, 0 = give up on the first failure\&.
.sp
A DC connects with a keep\-alive timeout of 30 seconds, or with its configured \fIkeep\-alive\-tmo\fR (see nvme\-fabrics\&.conf(5))\&. A DC configured with \fIpersistent = no\fR connects with none\&.
.sp
\fIzeroconf\fR (boolean, default false) connects to and manages Discovery Controllers found through mDNS (TP8009)\&. It requires systemd\-resolved v258 or later with mDNS enabled (\fIMulticastDNS=yes\fR), and an nvme\-discoverd built with mDNS support\&. It is re\-read on SIGHUP\&.
.sp
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)\&.
.SH "SIGNALS"
.PP
SIGHUP
.RS 4
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\&.
.RE
.PP
SIGTERM, SIGINT
.RS 4
Shut down gracefully\&.
.RE
.SH "FILES"
.PP
/usr/local/etc/nvme/nvme\-discoverd\&.conf
.RS 4
The daemon\(cqs own configuration; see CONFIGURATION\&.
.RE
.PP
/usr/local/etc/nvme/nvme\-discoverd\&.conf\&.sample
.RS 4
A commented example of nvme\-discoverd\&.conf with every knob at its default\&. It is not read; copy it to nvme\-discoverd\&.conf to change a setting\&.
.RE
.PP
/usr/local/etc/nvme/nvme\-fabrics\&.conf, /usr/local/etc/nvme/nvme\-fabrics\&.conf\&.d/*\&.conf
.RS 4
The connections nvme\-discoverd manages; see nvme\-fabrics\&.conf(5)\&.
.RE
.PP
/run/nvme/discoverd/
.RS 4
Runtime state linking each connected controller to the systemd transient unit that owns it\&. Does not survive a reboot\&.
.RE
.SH "SEE ALSO"
.sp
nvme\-fabrics\&.conf(5) tlshd(8) nvme\-connect(1) nvme\-registry\-list(1) nvme\-exclusion\-add(1)
.SH "NVME"
.sp
Part of the nvme\-user suite