Ben Pfaff
9795d7d750
Move vlog option descriptions into a separate file lib/vlog.man that is substituted into manpages. Get rid of individual rules for substituting most files in favor of a single suffix rule. Unfortunately this loses the (Emacs-specific) read-only markings but it simplifies the makefiles.
125 lines
4.8 KiB
Plaintext
125 lines
4.8 KiB
Plaintext
.TH ofp\-discover 8 "May 2008" "OpenFlow" "OpenFlow Manual"
|
|
|
|
.SH NAME
|
|
ofp\-discover \- controller discovery utility
|
|
|
|
.SH SYNOPSIS
|
|
.B ofp\-discover
|
|
[\fIoptions\fR] \fInetdev\fR [\fInetdev\fR...]
|
|
|
|
.SH DESCRIPTION
|
|
The \fBofp\-discover\fR program attempts to discover the location of
|
|
an OpenFlow controller on one of the network devices listed on the
|
|
command line. It repeatedly broadcasts a DHCP request with vendor
|
|
class identifier \fBOpenFlow\fR on each network device until it
|
|
receives an acceptable DHCP response. It will accept any valid DHCP
|
|
reply that has the same vendor class identifier and includes a
|
|
vendor-specific option with code 1 whose contents are a string
|
|
specifying the location of the controller in the same format used on
|
|
the \fBsecchan\fR command line (e.g. \fBssl:192.168.0.1\fR).
|
|
|
|
When \fBofp\-discover\fR receives an acceptable response, it prints
|
|
the details of the response on \fBstdout\fR. Then, by default, it
|
|
configures the network device on which the response was received with
|
|
the received IP address, netmask, and default gateway, and detaches
|
|
itself to the background.
|
|
|
|
.SH OPTIONS
|
|
.TP
|
|
\fB--accept-vconn=\fIregex\fR
|
|
By default, \fBofp\-discover\fR accepts any controller location
|
|
advertised over DHCP. With this option, only controllers whose names
|
|
match POSIX extended regular expression \fIregex\fR will be accepted.
|
|
Specifying \fBssl:.*\fR for \fIregex\fR, for example, would cause only
|
|
SSL controller connections to be accepted.
|
|
|
|
The \fIregex\fR is implicitly anchored at the beginning of the
|
|
controller location string, as if it begins with \fB^\fR.
|
|
|
|
.TP
|
|
\fB--exit-without-bind\fR
|
|
By default, \fBofp\-discover\fR binds the network device that receives
|
|
the first acceptable response to the IP address received over DHCP.
|
|
With this option, the configuration of the network device is not
|
|
changed at all, except to bring it up if it is initially down, and
|
|
\fBofp\-discover\fR will exit immediately after it receives an
|
|
acceptable DHCP response.
|
|
|
|
This option is mutually exclusive with \fB--exit-after-bind\fR and
|
|
\fB--no-detach\fR.
|
|
|
|
.TP
|
|
\fB--exit-after-bind\fR
|
|
By default, after it receives an acceptable DHCP response,
|
|
\fBofp\-discover\fR detaches itself from the foreground session and
|
|
runs in the background maintaining the DHCP lease as necessary. With
|
|
this option, \fBofp\-discover\fR will exit immediately after it
|
|
receives an acceptable DHCP response and configures the network device
|
|
with the received IP address. The address obtained via DHCP could
|
|
therefore be used past the expiration of its lease.
|
|
|
|
This option is mutually exclusive with \fB--exit-without-bind\fR and
|
|
\fB--no-detach\fR.
|
|
|
|
.TP
|
|
\fB--no-detach\fR
|
|
By default, \fBofp\-discover\fR runs in the foreground until it obtains
|
|
an acceptable DHCP response, then it detaches itself from the
|
|
foreground session and run as a background process. This option
|
|
prevents \fBofp\-discover\fR from detaching, causing it to run in the
|
|
foreground even after it obtains a DHCP response.
|
|
|
|
This option is mutually exclusive with \fB--exit-without-bind\fR and
|
|
\fB--exit-after-bind\fR.
|
|
|
|
.TP
|
|
\fB-P\fR[\fIpidfile\fR], \fB--pidfile\fR[\fB=\fIpidfile\fR]
|
|
Causes a file (by default, \fBofp\-discover.pid\fR) to be created indicating
|
|
the PID of the running process. If \fIpidfile\fR is not specified, or
|
|
if it does not begin with \fB/\fR, then it is created in
|
|
\fB@RUNDIR@\fR.
|
|
|
|
The \fIpidfile\fR is created when \fBofp\-discover\fR detaches, so
|
|
this this option has no effect when one of \fB--exit-without-bind\fR,
|
|
\fB--exit-after-bind\fR, or \fB--no-detach\fR is also given.
|
|
|
|
.TP
|
|
\fB-f\fR, \fB--force\fR
|
|
By default, when \fB-P\fR or \fB--pidfile\fR is specified and the
|
|
specified pidfile already exists and is locked by a running process,
|
|
\fBcontroller\fR refuses to start. Specify \fB-f\fR or \fB--force\fR
|
|
to cause it to instead overwrite the pidfile.
|
|
|
|
When \fB-P\fR or \fB--pidfile\fR is not specified, this option has no
|
|
effect.
|
|
|
|
.TP
|
|
.BR \-h ", " \-\^\-help
|
|
Prints a brief help message to the console.
|
|
|
|
@VLOG_OPTIONS@
|
|
|
|
.TP
|
|
.BR \-V ", " \-\^\-version
|
|
Prints version information to the console.
|
|
|
|
.SH BUGS
|
|
|
|
If the network devices specified on the command line have been added
|
|
to an OpenFlow switch with \fBdpctl addif\fR, then controller
|
|
discovery will fail because \fBofp\-discover\fR will not be able to
|
|
see DHCP responses, even though tools such as \fBtcpdump\fR(8) and
|
|
\fBwireshark\fR(1) can see them on the wire. This is because of the
|
|
structure of the Linux kernel networking stack, which hands packets
|
|
first to programs that listen for all arriving packets, then to
|
|
OpenFlow, then to programs that listen for a specific kind of packet.
|
|
OpenFlow consumes all the packets handed to it, so tools like
|
|
\fBtcpdump\fR that look at all packets will see packets arriving on
|
|
OpenFlow interfaces, but \fRofp\-discover\fR, which listens only for
|
|
arriving IP packets, will not.
|
|
|
|
.SH "SEE ALSO"
|
|
|
|
.BR secchan (8),
|
|
.BR ofp-pki (8)
|