*/etc/rrqnet/ifupdown.sh* is a utility script for managing +rrqnet+
virtual cables via +/etc/network/interfaces+ declarations. In normal
-use the script is set up as "pre-up" and "post-down" scripts by means
-of links from +/etc/network/if-pre-up.d/rrqnet+ and from
-+/etc/network/if-post-down.d/rrqnet+. It will then be invoked by
-+ifup+ and +ifdown+ for handling the +rrqnet+ declarations and bring
-up or tear down +rrqnet+ virtual cable plugs over +tap+ interfaces.
+use the script is set up as hook script by means of links from all
++/etc/network/if-*.d/rrqnet+ for automatic invocation with the
+commands +ifup+ and +ifdown+.
-An +rrqnet+ virtual cable uses a +tap+ interface at each cable end
+A +rrqnet+ virtual cable uses a +tap+ interface at each cable end
host, and a service process (aptly named +rrqnet+) to tunnel the +tap+
network traffic over UDP. Each +rrqnet+ process acts as a networking
switch that facilitates level 2 connectivity among all its end points
addresses.
This script handles the special purpose "option codes" for the IFACE
-stanzas that are used for declaring the +rrqnet+ daemon configuration
+blocks that are used for declaring the +rrqnet+ daemon configuration
to for the +tap+ interface for its virtual cabling. I.e., the IFACE
stanza is made for the tap interface and it includes one or more of
these +rrqnet+ options.
duly handled by +rrqnet-ifupdown.sh+.
The +rrqnet_options+ in the example tells the +rrqnet+ process to use
-an ipv4-only socket, 10 packet buffers and a single delivery thread.
-If left out, the default is to use an ipv6 socket, 10 bufffers and 5
-threads.
+an ipv4-only socket, 10 packet buffers and a single delivery thread,
+in addition to the socket reader threads. The default is to use an
+ipv6 socket, 10 bufffers and 5 delivery threads.
NOTES
-----
collection of fully connected hosts, although the more common is a
"star' formation. See the EXAMPLES section for inspiration.
-*rrqnet* includes logic aiming to protect against broadcast cycles.
-Howewer it does not have the more advanced spanning tree logic that is
-offered by bridge interfaces. In general it's probably best to avoid
-cabling cycles and perhaps rather run several *rrqnet* on a host with
-their taps connected with a bridge interface. Though, multiple virtual
-cabling paths between hosts might increase connection reliability.
+*rrqnet* includes logic to protect against broadcast cycles. Whilst
+acting as a switch for all its remotes it does not have spanning tree
+logic as is offered by bridge interfaces. Generally it's probably best
+to avoid cabling cycles, but *rrqnet* also handles full mesh linking
+between multiple hosts without additional configuration. Such a setup
+is also dynamically adapting to prefer the fastest link between hosts.
By default *rrqnet* opens an +ipv6+ socket on the given port. This
mode handles both +ipv6+ and +ipv4+ remotes with +ipv4+ remotes
SEE ALSO
--------
*rrqnet-cron(8)* - Management script to uphold a *rrqnet* plug.
+*rrqnet-ifupdown(8)* - Management script to uphold a *rrqnet* plug.
*vde_plug(1)* - Virtual Distributed Ethernet plug.