1 <!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML//EN">
4 <title>Hops, Network Isolation and other matters...</title>
5 <meta name="Keywords" content="DX Cluster, DXSpider, Spider, Packet Cluster, DXCluster, Pavillion Software, AK1A, AX25, AX.25, WWV, Packet Radio, Amateur Radio, Propagation, DX, DXing, G1TLH, GB7TLH, Dirk Koopman, Mailing list, Linux, RedHat, PERL">
6 <meta name="Description" content="Software and systems for realtime digital communications between amateur radio stations for the provision of information on propagation conditions and stations operating">
7 <meta name="Author" content="Dirk Koopman G1TLH">
10 <body TEXT="#000000" LINK="#0000ff" VLINK="#800080" BGCOLOR="#FFFFFF">
11 <FONT COLOR="#606060">
13 <h2>Hops, Network Isolation and other matters...</h2>
18 <address><a href="mailto:djk@tobit.co.uk">Dirk Koopman G1TLH</a></address>
20 <!-- Created: Sun Dec 13 20:25:14 GMT 1998 -->
22 Last modified: Mon Dec 21 11:57:54 GMT 1998
27 Starting with version 1.13 there is simple hop control available on a per
28 node basis. Also it is possible to isolate a network completely so that you
29 get all the benefits of being on that network, but can't pass on information
31 to any other networks you may be connected to (or vice versa).
33 <h4>Basic Hop Control</h4>
35 The number of hops that are set for all PC protocol messages (that require them)
36 are specified in <tt>/spider/perl/DXProtVars.pm</tt>.
38 <p>In versions prior to 1.13 you would move this file to
39 <tt>/spider/local/</tt> and modify the perl variables:
40 <tt>$def_hopcount</tt> and <tt>%hopcount</tt> to some reasonable
43 <p>From version 1.13 onwards a new mechanism has been introduced
44 which uses a file called <tt>/spider/data/hop_count.pl</tt>. The
45 prefered way of doing basic hop control is now to create this file
46 and modify it as you wish. Eventually this file will contain all
47 the hop control and related information. An example of the
48 <tt>hop_count.pl</tt> file can be found in the
49 <tt>/spider/examples</tt> directory.
51 <p>You can change this file at any time, including when the
52 cluster is running. If you do this then the changes only take
53 effect after you have run the <tt>load/hops</tt> command on a
54 client console with full sysop privileges.
56 <h4>Per Node Hop Control</h4>
58 From version 1.13 it is possible to control the number of hops to each
59 node. This is done by adding information to the <tt>%nodehops</tt> perl
60 variable in the <tt>hop_count.pl</tt> file (as described above). This
61 variable is a perl "hash of hashes", which means that you create an
62 entry for every callsign you wish to control and then one line for
63 every PC protocol message that you wish to alter.
65 <p>You can also have a entry called <tt>default</tt> for every callsign
66 so you can set the hops as a whole for all PC messages to just that
67 callsign. This is overridden by any specific hop counts you may have.
69 <h4>Example <tt>hop_count.pl</tt> File</h4>
75 # hop table construction
80 # default hopcount to use
83 # some variable hop counts based on message type
94 # the per node hop control thingy
118 <p>The figures chosen are not necessarily what I use. What I would say is that
119 until you are certain that you know what you are doing (and that the software
120 is working at least as well as advertised) you should keep the default hop
121 counts down to the sort of levels shown above.
123 <h4><a name="isolate"></a>Isolated Networks</h4>
125 It is possible to isolate networks from each other on a "gateway" node using
126 the <tt>set/isolate <node call></tt> command.
128 <p>The effect of this is to partition an isolated network
129 completely from another nodes connected to your node. Your node
130 will appear on and otherwise behave normally on every network to
131 which you are connected, but data from isolated network will not
132 cross onto any other network or vice versa.
134 <P>However all the spot, announce and WWV traffic and personal
135 messages will still be handled locally (because you are a real
136 node on all connected networks), that is locally connected users
137 will appear on all networks and will be able to access and receive
138 information from all networks transparently.
140 <p>All routed messages will be sent as normal, so if a user on one
141 network knows that you are a gateway for another network, he can still
142 still send a talk/announce etc message via your node and it will
145 <p>The only limitation currently is that non-private messages
146 cannot be passed down isolated links regardless of whether they
147 are generated locally. This will change when the bulletin routing
150 <!-- Standard Footer!! -->
153 <FONT COLOR="#606060"><hr></font>
154 <font color="#FF0000" size=-2>
155 Copyright © 1998 by Dirk Koopman G1TLH. All Rights Reserved<br>
157 <font color="#000000" size=-2>$Id$</font>