Configuring a remote-controlled node using NET.PE1CHL

                                             Updated: Aug 5, 1992


Because many nodes are now converting from NET/ROM or TheNET 
running on a TNC2 (or clone) to a PC or Atari ST, using an SCC 
(8530) board and running TCP/IP NET, some guidelines are collec-
ted in this document to help configuration of such a node.
This should not be considered a complete reference for the sysop, 
but rather a collection of remarks and examples in addition to 
the (non)existing program documentation!  You should at least 
also refer to the commands reference (netdoc.arc) and the SCC 
driver manual (scc.arc).
When you (the node sysop) have useful additions to these notes, 
like configuration tricks that you have used to overcome local 
difficulties, please send them to the author for addition to this 
document.

The document was written using 1stWord+ on an Atari ST, and is 
best printed using that program.  For users without an ST, a 
plaintext version of the document is available too.

Auto-start of program

The system should be configured in such a way, that it will 
automatically start the NET program when powered-up or RESET, and 
will re-start whenever the NET program exits.  This can be easily 
accomplished:

-    Atari ST systems:

     Create a folder (directory) on the startup disk, named AUTO.
     Copy the NET program into this folder, and name it "NET.PRG" 
     (not NET.TOS, as it is normally distributed).
     After that, copy a program into the AUTO folder that will 
     reboot the machine.  Such programs are available in the 
     public domain, and can be obtained from PE1CHL too.

     When you want to be able to upload new versions of the 
     program lateron, you should make sure the "reboot" program 
     is not in the folder immediately after NET, but there should 
     be some unused directory entries between them.  To accom-
     plish this, copy a few empty files to the AUTO folder, then 
     copy the reboot program, and then delete the empty files.
     Then, you can upload a new version to the AUTO folder (using 
     FTP), and delete the original program after you have 
     completed the upload.  A reset of the system will then start 
     the new version.

     A batch file that I use to prepare a floppy for the PI8NOS 
     node looks like this:

     copy tcnet.tos b:\auto\net.prg
     copy empty b:\auto\empty.1
     copy empty b:\auto\empty.2
     copy empty b:\auto\empty.3
     copy empty b:\auto\empty.4
     copy empty b:\auto\empty.5
     copy empty b:\auto\empty.6
     copy empty b:\auto\empty.7
     copy empty b:\auto\empty.8
     copy coldboot.prg b:\auto
     rm b:\auto\empty.*



-    MS-DOS systems:

     At the end of the AUTOEXEC.BAT file on the startup disk, put 
     the "NET" command to start "NET.EXE", and on the following 
     line put a command to re-boot the system.  Programs to do 
     that are available in the public domain, and are usually 
     called something like "BOOT.COM", "REBOOT.COM" etc.  You can 
     create your own "BOOT.COM" by creating a file with the 5-
     byte sequence "EA 00 00 FF FF", using DEBUG.

     Example of an AUTOEXEC.BAT file (for a 2-floppy system):

     @echo off
     prompt $P$G
     path a:\;a:\tools;b:\net;
     b:
     cd \net
     echo Starting NET...
     faster 2
     net
     echo EXIT from NET! >>b:\net\net.log
     boot


The effect of the above setup is the same for both machines: 
during the startup procedure the NET program is started, and it 
normally stays running (i.e. the startup procedure never termina-
tes).  Should the NET program ever exit, the next line in the 
procedure (or the next program in the AUTO folder) will reboot 
the machine, and the NET program will be re-started.


Watchdog

In the event of a program crash, it is very inconvenient to have 
to reset the machine manually.  Program crashes occur infre-
quently, and only when no operator is watching.
The program has a feature that allows the connection of a 
hardware watchdog circuit, that will RESET the machine when the 
program seems to hang.

The principle of such a circuit, is to generate a RESET pulse 
when no activity has been sensed on the input of the circuit 
during a certain interval of time.  The program can generate a 
pulse on some output pin (e.g. a DATA line of the parallel 
printer port) at each pass through the main loop.

A suitable watchdog reset circuit has been published before.  It 
uses only a few CMOS IC's, resistors and capacitors to generate a 
well-defined reset pulse after a while of inactivity on it's 
input.

For the Atari version of the program, the watchdog command also 
allows the definition of a software timeout that will cause a 
RESET of the machine after a certain period of inactivity, 
without requiring external hardware.  This method provides some 
protection against crashes, but is less reliable than the 
hardware-assisted approach.

The command to activate the watchdog pulses is:

-    Atari ST systems:

     watchdog 180 600 01 

     This enables 2 forms of watchdog:
     1.   A software timer that will reset the machine when no 
          activity has occurred for 180 seconds (or for 600 
          seconds when a "shell" command has been given).
     2.   An pulse output will be given on the lowest bit (D0) of 
          the parallel printer output at each pass through the 
          program loop.  Other outputs can be selected by speci-
          fying a different 3rd parameter.  A value of 00 for the 
          3rd parameter will disable the pulses on the parallel 
          output.

-    MS-DOS systems:

     watchdog 3bc 01 

     This sets up a trigger for a hardware watchdog (the MS-DOS 
     version of the program does not support the software timer).
     Each loop through the program, the value from port 3bc will 
     be read, OR-ed with the value 01, and sent out port 3bc.
     Then, the value is AND-ed with NOT 01, and this result is 
     sent out port 3bc.

     The watchdog circuit should then be connected to bit 0 of 
     the LPT1 printer port, i.e. port 3BC.  As there is normally 
     no use for a printer in an unattended node, and a printer 
     port is usually available on a PC, this is a convenient 
     point to connect the watchdog circuit.  Other lines, like 
     the DTR line of a COM port, can also be used when the 
     parameters of the watchdog command are adjusted.
     (in fact is is still possible to connect a printer to the 
     port, as the watchdog only drives a dataline, and does not 
     activate the strobe)


Watchdog circuit

"a simple watchdog for packet radio node controllers" by pe1chl

When operating a packet radio node on a hilltop site (or another 
site that is not easily accessable) it can be very inconvenient 
when the software crashes and leaves the node in an unusable 
state.
Of course, most node hardware already contains a transmitter 
(PTT) watchdog, that prevents the transmitter from being keyed 
indefinitely.  However, it would also be desirable to RESET the 
computer hardware whenever the software has crashed.
Assuming that one of the output signals of the node is changing 
between '0' and '1' whenever the node is operating correctly, the 
node can be reset whenever this signal stays at the same level 
for a certain time.

The following simple circuit performs this function, without 
requiring very long R/C time constants (and the associated big 
capacitors):

            11+-----------------------+
 ---| 2M7 |---|1                     |7     ------
 |          10|        4060         Q3|------|    |
 |--|560 K|---|0*                    |3     | &  |o-------o
 |           9|                    Q13|------|    |   |    ^
 -----||------|0      R      Q4      |      ------   | RESET*
     100nF    +-----------------------+               |   output
                       |12     |5                     |
    ~ 8 Hz             |       --| 2K2 -----          |
                       o                   |          |
                    -------                |          |
                    |     |                V->  LED   |
                    |  &  |               ---         |
                    -------                |          |
                      | |                 gnd         |
      +5 --| 10K |----- -----| 10K |-- +5             |
                    |     |                           |
               22n ===   === 22n                      |
                    |     |                           |
                    |     o                           |
                    |  -------                        |
                    |  |     |                        |
                    |  |  &  |     & gates: 4093      |
                    |  -------                        |
                    |    | |                          |
                    |----- ---------------------------|
                    o                                 |
                 -------                              |
                 |     |                              |
                 |  &  |                              |
                 -------                              |
 Pulse Input       | |                                |
  v                | ----------------------------------
  o---| 10K |-------                     


When the Pulse Input stays at a constant level (1 or 0) for more 
than a certain time limit, a RESET* pulse (active low) is 
generated at the output.  This output should be connected to the 
RESET button connector of the system. (assuming that the RESET 
switch is a make-contact switch connected between ground and this 
connector)

The time limit is equal to the oscillator period, divided by 
8192.  For the given R and C values this is about 1000 seconds.  
The 1000-second time limit allows for some time during startup of 
the system and/or when running a SHELL or EDITOR, that does not 
provide the watchdog pulses.

The LED in the circuit is flashing slowly when there is no 
activity at the Pulse Input.  When regular pulses are applied it 
stays OFF.  This can be used to verify the operation of the 
watchdog pulse output, and to alert the operator that the 
watchdog is ticking while in an editor.



Regular resetting of the system

Instead of the watchdog method shown above, some sysops have used 
a timer to reset the system on a periodic basis, e.g. each night 
at a fixed time.
This has been done to prevent problems that develop gradually, 
e.g. the system running out of memory.

The current version of the software has been shown to be able to 
run for several weeks without these problems, and it is therefore 
not a good idea to implement such a method.  It causes the loss 
of all connections that were present at the time of the reset, 
and that requires a re-transmission of data (e.g. a BBS message 
or a file retrieved using FTP) that can cause a lot of overhead 
to the network.

When you don't trust the watchdog alone to recover the system 
when it should hang, it is better to implement a "remote reset" 
feature, e.g. using a double-tone decoder on one of the link 
receivers.  Transmission of the proper double-tone would hard-
reset the PC.


HOSTS.NET file

For performance reasons, both during the startup of the system 
and when it is handling SMTP mail, it is best when the hosts.net 
file is located on a RAMdisk.  This is especially true when the 
system is floppy-based (no harddisk available).
It is easy to copy the hosts.net file from floppy to RAMdisk 
during the startup of the program.  See the example autoexec.net 
files for details.

Because the HOSTS.NET file could become corrupted, or could 
become unaccessible when it is located in a RAMdisk and the 
RAMdisk driver is not loaded for some reason, it is best to hard-
code the node's own IP address and the route to the primary sysop 
using fixed IP adresses instead of hostnames.
Use the [44.xx.yy.zz] notation instead of the hostname for these 
two purposes (see the example files)
This allows the node to initialize and to be accessed by the 
sysop, in case the HOSTS.NET file is not available.  The problem 
can then be corrected without visiting the site.


Remote control

Use the "rcmd" server for remote control of the station.  This 
allows manual updating of routes, copying of new configuration 
files that have been FTP'ed to the node etc etc.
Remote commanding is done using a "telnet" session from the 
sysop's system, to a pre-defined port on the node.  Although no 
official (NIC-assigned) port number has been allocated for this 
purpose, it is customary to use port number 333.

When a "telnet pi8xxx 333" is set up, the node responds with a 
NET/ROM-like challenge, to which the sysop must respond using a 
5-character string from the password string, corresponding to the 
numbered positions given in one of the three challenges.
Because a monitoring station does not know which of the three 
challenges was used by the sysop, this method is more secure than 
the simpeler implementation in NET/ROM.

The newer versions (after 920701) of NET include an automatic 
logon feature for the "rcmd" server.  To use this feature, a file 
must be present with the systemname as filename, in the directory 
"\net\rcmd".  This file should contain the password of the system 
to login to. Example:

     rcmd pi8rni  (the password should be in file "\net\rcmd\pi8rni")

NET will wait for the login challenge (3 groups of 5 numbers), 
and will randomly select one of them.  It will then send the 
corresponding password characters as defined in the password file 
\net\rcmd\pi8rni.  When the password is correct, the remote 
system will respond with its prompt. The directory for the 
password files can be changed by setting the environment variable 
"NETRCMD" to a different value.

For maximum reliability it is best to define the route to the 
sysop via NET/ROM, because in the event of a failure of a single 
transmitter the route will auto-update to an alternative.
Hard-coded IP routes will fail when the interface they are routed 
to fails to transmit!  That can mean a trip to the site....


Selecting CALLSIGN and SSID

Contrary to what is required for NET/ROM and TheNET nodes, NET 
does NOT require a different callsign/ssid combination to be used 
on each of the interfaces.  In fact, ALL interfaces could operate 
the NET/ROM using the same callsign and SSID.

When this is what you require, use the following form of the 
"attach netrom" command:

attach netrom pi8xyz-9 

The call "pi8xyz-9" will be used for NET/ROM purposes on all 
interfaces, independent of the callsign that may be given in the 
attach command for that interface.
The node will send only a single ALIAS:CALL entry for itself.  
This saves space in the node tables of other nodes, and makes 
the structure of the node clearer for the users.

Unfortunately, things aren't always that simple.  Local regula-
tions may force you to use a different callsign on each band (as 
is the case in Holland, where the local access points on 70cm use 
a PI8 call, and the interlinks on 23cm have to use a PI1 call).
In such cases, it may still be possible to use the same 
callsign/ssid on more than one interlink, so that still some 
table space is saved.
Initialize the NET/ROM using:

attach netrom 

The node will use the callsign specified with the "attach" 
command used for each interface ("attach com" or "attach scc").

The alias to be sent for each interface is defined in the "netrom 
interface" command (which also sets the interface quality and 
link/lap mode, see below)
It is necessary to use THE SAME alias for each interface which is 
defined with THE SAME callsign/ssid combination.  In the case of 
operation with a single callsign, this means that all aliases 
should be the same.
When this is not done, the neighboring nodes will pickup a 
different alias each time a broadcast is received along a 
different route, and the alias of your node will toggle between 
all possible values in their nodelist!

Note that the alias string should consist only of characters in 
the ASCII range between "!" and "~" (21 hex to 7E hex).  Nodes 
with an alias violating this rule will NOT be put in the 
nodelist!


Neighbor node filtering

A wellknown problem of the classic NET/ROM and TheNET software is 
the capture of distant NODES broadcasts during band openings.  
This results in long lists of nodes that cannot be connected 
because the neighbor node from which they were received cannot be 
reliably connected.
NET provides different ways to avoid this problem:

-    neighbor node filters

-    selective node broadcasts

Neighbor node filtering is a mechanism to define the nodes from 
which NODES broadcasts are to be received.  This allows the sysop 
to define which nodes are to be directly connected.  Automatic 
updates are still used to construct the node table defining which 
nodes can be reached via those neighbor nodes.

Four different node filtering modes are available:

none      no filtering in effect (broadcasts from every node are 
          accepted)

accept    only broadcasts from nodes in the filter list are 
          accepted

reject    broadcasts from nodes in the filter list are rejected, 
          all others are accepted

exclusive like accept, but all NET/ROM traffic from nodes not in 
          the filterlist is rejected (not only nodelists but also 
          connections)

The filtering mode and filter list can be defined for each 
interface.  This allows a restrictive filtering to be used on 
interlinks, while a more liberal filtering can be used on the 
local access channel.  When a suitably low quality is defined for 
the local access channel, this allows experiments with private 
nodes (e.g. for TCP/IP or G8BPQ BBS nodes).

Example for a node with interlinks and local access point:

netrom nodefilter mode accept
netrom nodefilter mode reject lap
#
netrom nodefilter add pi1duy-6  ln.n
netrom nodefilter add pi1ehv    ln.s48
netrom nodefilter add pi1esa-3  ln.w
netrom nodefilter add pi1rni    ln.s48
netrom nodefilter add pi1vrz-9  ln.e
netrom nodefilter add pi1vrz-9  ln.e48
#
netrom nodefilter add pe1mzu    lap
netrom nodefilter add pe1nat    lap

This example sets "accept" mode for all interfaces, then 
overrides this setting with "reject" mode for the local access 
point interface (named "lap").
Then, the neighbor nodes to be accessed on the link interfaces 
are defined, and those nodes which can be sometimes (but not 
reliably) heard on the LAP channel are put in the filter table so 
that nodes will not be accepted from them.


The second technique that can be used, is to send node broadcast 
not to NODES, but to the neighbor(s) expected on each link.  This 
avoids reception of the nodelists by nodes which run NET/ROM or 
TheNET software, and did not put our node in their ROUTES table 
at 0 quality.  (e.g. nodes which are very far away and receive 
our broadcasts seldomly)

Example of such selective broadcasting:

netrom bcdest ln.s48 drop nodes
netrom bcdest ln.s48 add pi1ehv
netrom bcdest ln.s48 add pi1rni

This drops the default destination "NODES" for the indicated 
interface, and adds two new destinations.  Note that this adds 
extra overhead, as the nodelist is sent twice.  However, the 
software is clever enough not to send nodes to a destination 
which has sent the node to us.  This will shorten the nodelist to 
each destination.


As an extra feature to control the nodelist, the "netrom exclude" 
command is available.
This command allows a certain callsign to be excluded from the 
node table.  This does not mean the node cannot be connected, or 
cannot connect your node.  It will only suppress the placement in 
the nodetable when this node appears in a broadcast received from 
a neighbor.
Use "netrom exclude" to filter link nodes that are far away (and 
for which the LAP node already appears in the table).


Nodelist transmission intervals

The periodic transmission of nodelists tends to result in a burst 
of transmitter activity, and a peak in memory usage.  Therefore 
it has been made possible to transmit the node list at slightly 
different intervals on each interface, so that the transmissions 
will not occur all at the same time.
This also allows for more frequent updates to be sent on those
interfaces where the baudrate is high enough.

There is always a single timer that determines the "obsoletion" 
of routes for which no update was received for some time.  The 
default for this timer was 1 hour (3600 seconds) in the original 
NET/ROM, but it has been agreed to set this to 1800 seconds 
within the dutch packet network, to get a quicker update in case 
of link failures.

Example of timer settings:

netrom nodetimer 1795 ln.n
netrom nodetimer 1781 ln.e
netrom nodetimer 1763 ln.w
netrom nodetimer 1751 ln.e48
netrom nodetimer 1746 ln.s48
netrom nodetimer 3532 lap

(in this case, the transmission of a nodelist on the LAP 
interface is to be seen as a service to experimenters, and the 
interval of approximately 1 hour should be sufficient.  the LAP 
channel is 1200 baud and a frequent nodelist transmission 
presents a high load)


Other NET/ROM parameters

The NET/ROM parameters are set using some cryptic commands, e.g.:

netrom logmask 1a0
netrom obsotimer 1800
netrom param 3 60 6 5 20 3 20 3
netrom param 4 180 3 10 180 10 10 1800
netrom param 7 1

The "netrom logmask" setting determines the NET/ROM events that 
are to be logged in the logfile.  A setting of "ffff" will log 
all events, but the setting "1a0" will only log unsuccesful 
connection attempts.  The parameter is formed as the bitwise OR 
of a number of different values.  See the HISTORY file for more 
details.

The "netrom obsotimer" sets the interval between decrement of the 
obsoletion value in all routes, in this case to half an hour.

The "netrom param" commands set some parameters for NET/ROM, like 
the "p" command of the original NET/ROM.  The parameters have 
been separated into 3 classes:

3    - network level
4    - transport level (connections)
7    - user level

The meaning of the values is as follows:

network level:
     Autofloor:     The lowest quality of a node accepted in a 
                    NODES broadcast.  Nodes below this quality 
                    are not put in the nodelist.
     Obso_init:     Initial value for the obsolescence counter 
                    when a node is received in a broadcast.  This 
                    is the counter that is decremented on each 
                    "obsotimer" tick.
     Obso_minbc:    Minimal value of obsolescence counter for a 
                    node to be put in the NODES broadcast.
     TTL:           Time to live.  Maximal number of nodes that 
                    the packets sent by this node will be 
                    repeated.
     Maxroutes:     Number of different routes that will be kept 
                    for a single node.
     Maxqueue:      Maximum number of packets that are allowed 
                    to appear on the AX.25 queue to a neighbor 
                    node.  If more packets are to be added, they 
                    are dropped, and an alternative route to the 
                    destination is tried.
     Maxfail:       Number of times the AX.25 link to a neighbor 
                    is allowed to fail before it is deleted from 
                    the table.

transport level:
     Timeout:       Time between re-tries at the transport level, 
                    in seconds.
     Tries:         Maximum number of re-tries at the transport 
                    level.
     Ackdelay:      Delay time (in seconds) before an ACK is sent 
                    when no data is outstanding.  An ACK is sent 
                    immediately when the transport window fills.  
                    This delay is used to be able to acknowledge 
                    multiple packets with a single ACK.
     Busydelay:     Time to wait when a CHOKE packet (buffer 
                    overflow) is received from the destination 
                    node.
     Window:        Maximal transport window allowed during the 
                    negotiation at connection setup.  The actual 
                    window to be used will be the lowest value of 
                    the window parameter of the two nodes.
     Backlog:       Number of packets the node will buffer before 
                    sending an RNR or CHOKE, when the outgoing 
                    connection does not accept packets.
     Inactive:      Timeout (in seconds) before disconnection of 
                    an idle transport connection.

user level:
     Bothnodes:     Determines if both the callsign of the uplink 
                    node and that of the local node will be put 
                    in the digipeater string used when connecting 
                    a downlink user.  This information makes it 
                    easy to connect back to someone who connected 
                    via the network.
                    Value 0 will disable this feature.



The qualities of the interfaces should be selected depending on 
the reliability and speed (baudrate) of the link involved.
Assign a "hidden" alias (starting with a '#') to each interface 
only used for interlinks.  The keyword "lap" should be added only 
for the interface that is to be the default for downlink 
connects.  Other interfaces can have the keyword "uplink" added 
to allow AX.25 connects, but this should not be done for 
interlink interfaces.  AX.25 parameters for interlink-only 
interfaces are automatically tuned for better performance on 
these links.

Example:

netrom interface ln.n48 "#Bilt" 230
netrom interface lap.48 "DeBilt" 192 uplink
netrom interface lap "DeBilt" 100 lap






FASTER.COM

On MS-DOS (PC) systems, all timing for the SCC driver is derived 
from the system timer, which unfortunately runs at a rather low 
speed.  This reduces the resolution of all timing (txdelay, p-
persistence etc.) to 55ms.
This is not a problem on the Atari ST, where the system clock has 
a 5ms resolution which has been divided in the SCC driver to the 
familiar 10ms tick rate also found in most TNCs.

For the PC, a TSR program has been developed to fix this problem:

"FASTER.COM" is a small TSR that re-programs the PC's timer to 
interrupt the CPU at a faster rate than the default 18 ticks per 
second (55ms/tick).  The resident portion will divide-down the 
interrupt ticks to call the original handler at the standard 18 
ticks per second. Software that installs itself on the timer 
interrupt vector, like the SCC driver in NET, will now be called 
at a higher frequency, allowing better control over the radio 
timing (txdelay, hold after packet).

It must be installed once and only once after startup of the PC, 
and before NET is loaded.
It takes a parameter to specify the speedup factor.  The timer 
tick will be speeded by 2^n where n is the parameter of "faster".
Example:

        faster 2

Clock will be running at 4 times (2^2) the original rate, which 
has proven to be adequate for the baudrates currently in use.

Installing "faster" a second time will set the tick handler to 
the frequency you specify at the second invocation, but the 
resident portion will be installed 2 times, causing the real time 
clock to run slow at the rate specified at the first invocation!
Be careful, the program does not check for this condition!
It is best to install "faster" from AUTOEXEC.BAT, so that it will 
not be installed a second time.

Some software that uses timer interrupts or directly accesses the 
registers of the timer can have trouble when "faster" is 
installed!

(Thanks to Alan Vlieger for writing this program)

The SCC driver will measure the tickrate when it is initialized, 
and will print it.  The default parameter values for txdelay, p-
persistence slot and txhold times are automatically adjusted.  
Make sure you provide the proper values when you override these 
parameters from the configuration files!  (SCC AX.25 channel 
parameters 1, 3, 4 and 7)


Example configuration files

Below, the configuration files in use at two nodes in the dutch 
packet radio network are listed.
These are real-life configuration files, that have been in use 
and have been maintained for quite some time.  Therefore, many 
tricks that can be used are found in these files.

Two configurations are given: one for PI8NOS, a complex node with 
4 links to neighboring nodes and a local access point.  The node 
provides SMTP forwarding and FTP file access services.
The second configuration is for PI8RNI, which has only a single 
link (to PI8NOS), and does not provide any SMTP or FTP service.



PI8NOS

The hardware in use at PI8NOS is an Atari Mega-ST2, with an 8-
channel SCC interface and a 1200baud remote control connection 
connected to the AUX (RS232) port.  This is connected to a MODEM 
via the controller of an on-site 70cm repeater, which switches 
through the modem signals after verification of a password.  This 
allows an extra method to perform remote control, which is almost 
never used as the remote control via telnet proves to be 
perfectly adequate.

Some of the commands are specific for the Atari version of the 
program, but most of them apply to the PC version as well.

The configuration assumes the presence of a RAMdisk (provided by 
a utility program), and will copy some frequently accessed files 
to it.  This is for use on floppy-disk-only systems, and can be 
removed for harddisk-based systems.

The use of environment variables to define the different disks 
(floppy disk containing NET, RAMdisk) is shown in the example.  
This allows easy change of the drive letters when this is 
required.


----------------------------------------------------------------
AUTOEXEC.NET for PI8NOS
----------------------------------------------------------------

# NET configuration file for PI1NOS/PI8NOS network station in Hilversum
# uses special SCC interface board with 4 Z8530s installed
# AUX (RS232) is used as a connection to an async modem for remote control
# MIDI is not used
#
# Drive A: is a 720K floppy, drive P: is a 1MB RAMDisk
setenv NETDISK a:
setenv RAMDISK p:
#
date tzset MDT-2
#
# setup some directories and files. RAMDisk is faster so transfer commonly
# used files like HOSTS.NET to it. FTP public area is also on RAMDisk, and
# is initialized with some of our configuration files.
#
cd ${NETDISK}\net
mkdir -f ${RAMDISK}\tmp ${RAMDISK}\net ${RAMDISK}\net\mqueue ${RAMDISK}\net\finger ${RAMDISK}\public
#
setenv TMPDIR ${RAMDISK}\tmp
setenv TMP $TMPDIR
del -f $TMPDIR\*.*
#
copy $NETHOSTS ${RAMDISK}\net\hosts.net
setenv NETHOSTS ${RAMDISK}\net\hosts.net
copy $NETALIAS ${RAMDISK}\net\alias
setenv NETALIAS ${RAMDISK}\net\alias
copy ${NETDISK}\net\nr_info.txt ${RAMDISK}\net\nr_info.txt
copy ${NETDISK}\net\finger\*.txt ${RAMDISK}\net\finger
setenv NETFINGER ${RAMDISK}\net\finger\
#
copy ${NETMAILQ}*.* ${RAMDISK}\net\mqueue
del -f ${NETMAILQ}*.txt ${NETMAILQ}*.wrk ${NETMAILQ}*.lck
setenv NETMAILQ ${RAMDISK}\net\mqueue\
del -f ${NETMAILQ}*.lck
#
copy $NETALIAS routes.net finger\*.txt ${RAMDISK}\public
setenv ENVSTYLE
shell =${NETDISK}\arc.ttp
#
hostname pi8nos.ampr.org
ip address [44.137.36.16]
#
# define buffers for use by SCC driver
buffers 64 2 128
#
# attach commands for SCC driver.  first init, then attach the channels.
attach scc 4 init fffd00 8 3 7 -2 fffd3f 3 p4915200
#
attach scc 7 ax25 ln.n48 256 d4800 pi1nos
attach scc 6 ax25 ln.n   256 d1200 pi1nos
attach scc 5 ax25 ln.s48 256 d4800 pi1nos
attach scc 4 ax25 ln.e48 256 d4800 pi1nos
attach scc 3 ax25 ln.w48 256 d4800 pi1nos
attach scc 2 ax25 ln.w   256 d1200 pi1nos
attach scc 1 ax25 lap48  256 d4800 pi8nos
attach scc 0 ax25 lap    256 d1200 pi8nos
#
# attach the NET/ROM interface (no call, default interface calls are used)
#
attach netrom
#
# remainder of configuration, routing, misc
#
source config.net
source routes.net
#
# explicit NET/ROM route to sysop for emergency cases
# (NET/ROM routes can be re-routed when an interface fails)
#
route add [44.137.40.1] netrom
arp   add [44.137.40.1] netrom  pe1chl-9
#
log ${NETDISK}\net\net.log startup
log proc
watchdog 180 600 1
#
# NET/ROM configuration. Filtering and routes are in NETROM.NET
#
netrom interface lap    "HVS"  100 lap
netrom interface lap48  "HVS"  192 uplink
netrom interface ln.e48 "#HVS" 230
netrom interface ln.n   "#HVS" 200
netrom interface ln.n48 "#HVS" 210
netrom interface ln.s48 "#HVS" 230
netrom interface ln.w   "#HVS" 220
netrom interface ln.w48 "#HVS" 230
netrom tcpip link "NOCALL"
source netrom.net
netrom info ${RAMDISK}\net\nr_info.txt
#netrom route save ${RAMDISK}\net\netrom.rts
#source ${RAMDISK}\net\netrom.rts
#
# filtering of unwanted AX.25 stations
#
ax25 exclude nocall
ax25 exclude pi1nos
#
# MHEARD lists
#
mheard lap    23
mheard lap48  23
mheard ln.e48 10
mheard ln.n   10
mheard ln.n48 10
mheard ln.s48 10
mheard ln.w   10
mheard ln.w48 10
#
# AX25 ports.  1=TNC 2=NetDigi 3=MHEARD 4=Bridge 5=TNC2 6=MBOX
#
ax25 port 1 conn pi8nos-15
ax25 port 4 conn pi8nos-6
ax25 port 6 conn pi8nos-1
#
# SCC driver timer tick is 10ms
#
# initialize SCC driver on lap
#
param lap 1 20
param lap 8 10
ax25 persist lap 128 25 64 60 900
#
# initialize SCC driver on lap48
#
param lap48 1 10
ax25 persist lap48 128 50 128 60 900
#
# initialize SCC driver on ln.e48
#
param ln.e48 1 12
param ln.e48 2 255
param ln.e48 7 1
#
# initialize SCC driver on ln.n
#
param ln.n 1 20
param ln.n 2 255
param ln.n 7 1
param ln.n 8 15
#
# initialize SCC driver on ln.n48
#
param ln.n48 1 12
param ln.n48 2 255
param ln.n48 7 1
#
# initialize SCC driver on ln.s48
#
param ln.s48 1 12
param ln.s48 2 255
param ln.s48 7 1
#
# initialize SCC driver on ln.w
#
param ln.w 1 20
param ln.w 2 255
param ln.w 7 1
param ln.w 8 15
#
# initialize SCC driver on ln.w48
#
param ln.w48 1 12
param ln.w48 2 255
param ln.w48 7 1
#
# send the netrom nodes on startup (links only)
#
netrom bcnodes ln.e48
netrom bcnodes ln.n
netrom bcnodes ln.n48
netrom bcnodes ln.s48
netrom bcnodes ln.w
netrom bcnodes ln.w48
#
# now we can safely start all servers
#
start discard
start echo
start finger
start ftp
start smtp
start tp
source remote.net
ax25 start bridge
ax25 start mbox ${NETFINGER}network.txt
#
# commands to be run at specified time-of-day
#
at 00:00 source ${NETDISK}\net\at0000.net
at 06:00 source ${NETDISK}\net\at0600.net
at 12:00 source ${NETDISK}\net\at1200.net
at 18:00 source ${NETDISK}\net\at1800.net
at 67/12:00 source ${NETDISK}\net\at1200we.net
#
# allow control from AUX port 1200 baud / 8 bits / no flow control
# (modem currently not connected so let's disable this for now)
#auxcon 11 1200 8 noflow
#
hostname
ip address


----------------------------------------------------------------
CONFIG.NET for PI8NOS
----------------------------------------------------------------

# remainder of configuration
# SMTP options
#
smtp expiry 7
#smtp gateway ...
smtp maxclients 10
smtp maxsessions 1
smtp mode forward
smtp timer 4000 300
#
# TCP parameters
#
tcp irtt 30000
tcp mss 512
tcp timeout 7200
tcp ubound 3600
tcp window 648
#
# IP parameters
#
ip ttl 16
#
# ICMP parameters
#
ping length 0
#
# AX25 parameters
#
ax25 digipeat off
ax25 maxframe 4 16
ax25 paclen 256
ax25 pthresh 64
ax25 retry 10
ax25 t1 10000 5000
ax25 t2 2000 500
ax25 t3 1800000
ax25 t4 900000
ax25 window 2048
#
# NET/ROM parameters
#
netrom logmask 1a0
netrom obsotimer 1800
netrom param 3 60 6 5 20 3 20 3
netrom param 4 120 3 10 180 10 10 1800
netrom param 7 1 0 0
#
# Misc settings
#
escape F10
flow on
screen direct


----------------------------------------------------------------
ROUTES.NET for PI8NOS
----------------------------------------------------------------

# IP routing table for PI1NOS/PI8NOS
#
# ARP statements for AX.25 stations
#
arp add pa0hzp          ax25:lap48 pa0hzp-4
arp add pi8ehv          ax25:ln.s48 pi1ehv
arp add pi8rni          ax25:ln.s48 pi1rni
#
# ARP statements and ROUTES for NET/ROM stations
#
route add dl1bke        netrom
arp   add dl1bke        netrom  dl1bke-9
route add pa0mjr        netrom
arp   add pa0mjr        netrom  pa0mjr-9
route add pa3ayh        netrom
arp   add pa3ayh        netrom  pa3ayh-9
route add pa3cay        netrom
arp   add pa3cay        netrom  pa3cay-9
route add pa3ebv        netrom
arp   add pa3ebv        netrom  pa3ebv-2
route add pe1brj        netrom
arp   add pe1brj        netrom  pe1brj-9
route add pe1dzq        netrom
arp   add pe1dzq        netrom  pe1dzq-9
route add pe1ljf        netrom
arp   add pe1ljf        netrom  pe1ljf-9
route add pi8daz        netrom
arp   add pi8daz        netrom  pi8daz-7
route add pi8gro        netrom
arp   add pi8gro        netrom  pi8gro
route add pi8hvh        netrom
arp   add pi8hvh        netrom  pi8hvh
route add pi8yrc        netrom
arp   add pi8yrc        netrom  pi8yrc
#
# set direct route for nearby networks (no default, to get unreachable mesgs)
#
route add Holland/16    lap
#
# Exceptions to gateway routings
#
route add pa0hzp        lap48
route add pe1drr        unreach
#
# stations reachable via (their own private) gateways
#
route add on1btj        via     pi8hvh
route add pa3ayb        ln.s48  pi8rni
route add pa3ecp        via     pi8yrc
route add pa3eko        ln.s48  pi8ehv
route add pa3fmc        via     pi8yrc
route add pe1dcy        ln.s48  pi8rni
route add pe1mzu        via     pi8yrc
route add pe1nnh        ln.s48  pi8rni
route add sun4c.pe1hzg/24 ln.s48 pi8ehv
#
# some networks reachable via gateways
#
route add Belgium/16    ln.s48  pi8ehv
route add Germany/16    ln.s48  pi8ehv
route add Switzerland/16 ln.s48 pi8ehv
route add PA_DRnet/22   via     pi8gro
route add PA_FRnet/22   via     pi8gro
route add PA_GLnet/22   via     pe1brj
route add pa3ayh/24     via     pa3ayh
route add PA_GRnet/22   via     pi8gro
route add PA_LBnet/22   ln.s48  pi8ehv
route add PA_NBOnet/22  ln.s48  pi8ehv
route add PA_NBWnet/22  ln.s48  pi8ehv
route add PA_NH1net/22  via     pi8yrc
route add PA_OVnet1/24  via     pi8daz
route add PA_OVnet2/24  via     pa3ebv
route add PA_UTnet/22   ln.s48  pi8rni
route add PA_ZH1net/22  via     pa3cay
route add PA_ZH2net/22  via     pi8hvh
route add PA_ZLnet/22   via     pi8hvh


----------------------------------------------------------------
NETROM.NET for PI8NOS
----------------------------------------------------------------

# NET/ROM routes and filters for PI8NOS
# only the nodes in the list below are accepted for automatic route table
# updates (NODES broadcasts) on the interlink.  this is controlled by the
# "mode accept".  on LAP48 anybody can send broadcasts, except when an
# explicit filter entry is added to disallow it. ("mode reject")
# on LAP, only stations in the list can use the NET/ROM ("mode exclusive"),
# and nobody is in the list.  so, on LAP no NET/ROM links are possible.
#
netrom nodefilter mode accept
netrom nodefilter mode exclusive lap
netrom nodefilter mode reject lap48
#
netrom nodefilter add pi1ehv    ln.s48
netrom nodefilter add pi1esa-3  ln.w
netrom nodefilter add pi1esa-3  ln.w48
netrom nodefilter add pi1jyl-1  ln.n
netrom nodefilter add pi1jyl-1  ln.n48
netrom nodefilter add pi1rni    ln.s48
netrom nodefilter add pi1vrz-9  ln.e48
#
netrom nodefilter add pe1chl-9  lap
#
# explicitly addressed broadcasts on links (avoids reception by others)
#
netrom bcdest ln.e48 drop nodes
netrom bcdest ln.e48 add pi1vrz-9
#
netrom bcdest ln.n drop nodes
netrom bcdest ln.n add pi1jyl-1
netrom bcdest ln.n48 drop nodes
netrom bcdest ln.n48 add pi1jyl-1
#
netrom bcdest ln.s48 drop nodes
netrom bcdest ln.s48 add pi1ehv
netrom bcdest ln.s48 add pi1rni
#
netrom bcdest ln.w drop nodes
netrom bcdest ln.w add pi1esa-3
netrom bcdest ln.w48 drop nodes
netrom bcdest ln.w48 add pi1esa-3
#
# extended sequence numbers can be used with NET neighbors
#
ax25 extended pi1ehv ln.s48
ax25 extended pi1rni ln.s48
#
# uneven nodetimers reduce the need for parallel transmission and memory
#
netrom nodetimer 0    lap
netrom nodetimer 1771 lap48
netrom nodetimer 1747 ln.e48
netrom nodetimer 1763 ln.n
netrom nodetimer 1752 ln.n48
netrom nodetimer 1734 ln.s48
netrom nodetimer 1728 ln.w
netrom nodetimer 1712 ln.w48
#
# some excludes to prevent connect attempts when link has failed.
#
ax25 exclude pi1ehv lap
ax25 exclude pi1esa lap
ax25 exclude pi1jyl lap
ax25 exclude pi1rni lap
ax25 exclude pi1vrz lap
ax25 exclude pi1yrc lap
ax25 exclude pi8ehv lap
ax25 exclude pi8yrc lap
#
# some nodes excluded because they are non-adjacent link-nodes
# or are unreliable (not QRV 24h/day)
#
netrom exclude pi1awt
netrom exclude pi1cxc
netrom exclude pi1dre
netrom exclude pi1eae
netrom exclude pi1fwd
netrom exclude pi1gro
netrom exclude pi1gwo
netrom exclude pi1hvh
netrom exclude pi1hwb
netrom exclude pi1lim
netrom exclude pi1mpi
netrom exclude pi1nym
netrom exclude pi1pac
netrom exclude pi1rmd
netrom exclude pi1vab
netrom exclude pi1yrc
netrom exclude pi1zld
netrom exclude pi8awt


----------------------------------------------------------------
REMOTE.NET for PI8NOS
----------------------------------------------------------------

# commands to start remote-control servers
# put in a separate file so that we don't have to send it every time
# autoexec.net is updated (somebody could be monitoring us)
#
start rcmd 333 "dsfjkaewrqbefndscuwerfjasfvaaaaarughrueggdfv"
start remote 3534

[of course this is not the real file used at PI8NOS!!!]

----------------------------------------------------------------
ONEXIT.NET for PI8NOS
----------------------------------------------------------------

# this file will be executed on "exit" from NET
#
copy ${NETMAILQ}*.* ${NETDISK}\net\mqueue
del ${NETMAILQ}*.*
netrom route save

----------------------------------------------------------------
AT0000.NET for PI8NOS
----------------------------------------------------------------

# commands executed each day at 00:00
#
netrom filter mode n
copy ${RAMDISK}\net\mqueue\sequence.seq ${NETDISK}\net\mqueue

----------------------------------------------------------------
AT0600.NET for PI8NOS
----------------------------------------------------------------

# commands executed each day at 06:00
#

----------------------------------------------------------------
AT1200.NET for PI8NOS
----------------------------------------------------------------

# commands executed each day at 12:00
#
copy ${RAMDISK}\net\mqueue\sequence.seq ${NETDISK}\net\mqueue

----------------------------------------------------------------
AT1800.NET for PI8NOS
----------------------------------------------------------------

# commands executed each day at 18:00
#
netrom filter mode y

----------------------------------------------------------------
NR_INFO.TXT for PI8NOS
----------------------------------------------------------------

PI1NOS/PI8NOS - NET/ROM and TCP/IP node in Hilversum

Local Access Point is 430.750 MHz, links are on 23cm

Commands are:

CONNECT nodecall              Connect another node (call or alias)
CONNECT ax25call [VIA digis]  Connect an AX.25 user on the LAP
INFO                          Show this information
MHEARD                        Show MHEARD list of the LAP
MHEARD interface              Show MHEARD list of any interface
NODES                         Show local access NODES
NODES *                       Show all NODES
NODES * *                     Show complete NODES info
NODES nodecall                Show info about a single NODE (call or alias)
ROUTES                        Show routes to neighbor nodes
USERS                         Show connected users

Local services:

PI8NOS-1        simple mailbox that can be used to send messages to the SYSOP
PI8NOS-6        conference bridge

Het bouwen en instandhouden van deze node kost veel geld, dat uit eigen zak
betaald wordt door enkele van uw mede-amateurs.  Wilt U ook een bijdrage
leveren, stort deze dan op Postbank rek.nr. 3408886 t.n.v. penningmeester Scoop
Repeaterfonds Jan Rozema, Kortenhoef.  Zie ook Electron Mei 1991, blz. 247.



----------------------------------------------------------------


This concludes the PI8NOS configuration files.


PI8RNI

The second example node, PI8RNI, is running on a PC (12MHz 80286) 
with 2 floppy drives and 1M RAM.  It uses a PA0HZP OptoPcSCC 
interface card for connection to the modems.  This is a PC-card 
version of the interface that is also used in PI8NOS.

The system board supports Expanded Memory, which can be 
configured as a RAMdisk using the config.sys file:

----------------------------------------------------------------
CONFIG.SYS for PI8RNI
----------------------------------------------------------------

files=20
buffers=20
stacks=0,0
device=dems.sys
device=emmdisk.sys 320


The node is started from the A: floppy, which contains MS-DOS, 
the EMS and RAMdisk device drivers, and some utility programs 
(editor, ARChive program) for maintenance purposes.  The program 
is started from the AUTOEXEC.BAT file like this:

----------------------------------------------------------------
AUTOEXEC.BAT for PI8RNI
----------------------------------------------------------------

@echo off
prompt $P$G
path a:\;a:\tools;b:\net;
b:
cd \net
boottime >boottime
echo Starting BM...
bm -s "Auto boot log" egon@pa3edn marcel@pa3azk rob@pe1chl <boottime
echo Starting NET...
faster 2
net
echo EXIT from NET! >>b:\net\net.log
boot

NET and its configuration files are located on the B: floppy, so 
the .BAT file changes to that drive and the \net directory.  The 
"boottime" program simply prints a message with the current date 
and time, so that this can be typed later.  A mail is sent to 
some people notifying of the startup.
The "faster" program is used to speedup the clock interrupt by a 
factor of 4 (2^2) to get a better resolution on the SCC driver 
timing.

----------------------------------------------------------------
AUTOEXEC.NET for PI8RNI
----------------------------------------------------------------

# NET configuration file for PI1RNI/PI8RNI network station in Bilthoven
# uses PA0HZP SCC interface board with 2 Z8530s installed
#
# A: is DOS floppy with utilities (360K)
# B: is NET floppy with config files (720K)
# C: is RAMdisk (300K)
#
setenv DOSDISK A:
setenv NETDISK B:
setenv RAMDISK C:
#
mkdir -f $RAMDISK\tmp $RAMDISK\net $RAMDISK\public
copy $NETHOSTS $RAMDISK\net
setenv NETHOSTS $RAMDISK\net\hosts.net
setenv TMP $RAMDISK\tmp
setenv TZ MET0
#
cd \net
del -f $TMP\*.*
#
# define buffers for use by SCC driver
buffers 64 2 64
#
# attach commands for SCC driver.  first init, then attach the channels.
attach scc 2 init 150 4 2 0 1 168 3 p4915200
#
attach scc 0 ax25 ln.n48 256 d4800 pi1rni
attach scc 1 ax25 lap.48 256 d4800 pi8rni
attach scc 2 ax25 lap    256 d1200 pi8rni
#
# attach the NET/ROM interface (no call, default interface calls are used)
#
attach netrom
#
# filtering of unwanted AX.25 stations
#
ax25 exclude nocall
#
# remainder of configuration
#
source config.net
#
# Addresses, IP routing table and ARP commands, Misc
#
hostname pi8rni.ampr.org
ip address [44.137.40.49]
source routes.net
log ${NETDISK}\net\net.log startup
#
# lpt1 watchdog 20min
watchdog 3bc 1 
#
# NET/ROM configuration. Filtering and routes are in NETROM.NET
#
netrom interface ln.n48 "#Bilt" 230
netrom interface lap.48 "DeBilt" 192 uplink
netrom interface lap "DeBilt" 100 lap
netrom tcpip broadcast "#TCPIP"
source netrom.net
netrom info ${NETDISK}\net\nr_info.txt
#
# KISS and AX25 stuff for each TNC connected
#
mheard ln.n48 5
mheard lap.48 23
mheard lap 23
ax25 digipeat ln.n48 off
ax25 digipeat lap.48 off
ax25 digipeat lap off
#
# AX25 ports.  1=TNC 2=NetDigi 3=MHEARD 4=Bridge 5=TNC2 6=MBOX
#
ax25 port 4 conn pi8rni-6
#
# SCC driver interface groups
#
param lap group 301
param lap.48 group 201
#
# initialize SCC driver on lap
#
param lap 2 64
ax25 persist lap 256 25 64 30 900
#
# initialize SCC driver on lap.48
#
param lap.48 1 11
param lap.48 2 128
param lap.48 7 1
ax25 persist lap.48 256 25 128 60 900
#
# initialize SCC driver on ln.n48
#
param ln.n48 1 11
param ln.n48 2 64
param ln.n48 7 1
#
# send the netrom nodes on startup (link only)
#
netrom bcnodes ln.n48
#
# now we can safely start all servers
#
start discard
start echo
start finger
source remote.net
ax25 start bridge
#
# commands to be run at specified time-of-day
#
at 00:00 source ${NETDISK}\net\at0000.net
at 06:00 source ${NETDISK}\net\at0600.net
at 12:00 source ${NETDISK}\net\at1200.net
at 18:00 source ${NETDISK}\net\at1800.net
#
hostname
ip address
smtp kick


----------------------------------------------------------------
CONFIG.NET for PI8RNI
----------------------------------------------------------------

# remainder of configuration
# SMTP options
#
#smtp gateway ...
smtp maxclients 10
smtp mode forward
smtp timer 4000 0
#
# TCP parameters
#
tcp irtt 30000
tcp mss 512
tcp timeout 7200
tcp ubound 3600
tcp window 648
#
# IP parameters
#
ip ttl 16
#
# ICMP parameters
#
ping length 0
#
# AX25 parameters
#
ax25 digipeat off
ax25 maxframe 4
ax25 paclen 256
ax25 pthresh 64
ax25 retry 10
ax25 t1 10000
ax25 t2 2000
ax25 t3 1800000
ax25 t4 900000
ax25 window 2048
#
# NET/ROM parameters
#
netrom logmask 1a0
netrom obsotimer 1800
netrom param 3 60 6 5 20 3 20 3
netrom param 4 180 3 10 180 10 10 1800
netrom param 7 1
#
# Misc settings
#
escape F10
flow on


----------------------------------------------------------------
ROUTES.NET for PI8RNI
----------------------------------------------------------------

# IP routing table for PI1RNI/PI8RNI
#
# ARP statements for AX.25 stations
#
arp add pa3edn          ax25:lap.48 pa3edn-4
arp add pe1chl          ax25:lap.48 pe1chl-4
arp add pi8nos          ax25:ln.n48 pi1nos
#
# ARP statements and ROUTES for NET/ROM stations
#
route add [44.137.40.1] netrom
arp   add [44.137.40.1] netrom  pe1chl-9
#
# set direct route for nearby networks (no default, to get unreachable mesgs)
#
route add PA_UTnet/22   lap
#
# Exceptions to gateway routings
#
route add pa3ayb        lap
route add pa3edn        lap.48
route add pa3ein        lap.48
route add pa3eqb        lap.48
route add pb0aaq        lap.48
route add pe1chl        lap.48
route add pe1nnh        lap.48
route add pi8nos        ln.n48
#
# stations reachable via (their own private) gateways
#
route add pa3cay        via     pa3edn
route add sun4c.pe1hzg  via     pi8nos
route add sys2.pe1chl   via     pe1chl
route add sys2.pa3azk   via     pa3azk
route add sys2.pa3edn   via     pa3edn
#
# some networks reachable via gateways
#
route add Holland/16    via     pi8nos
route add Belgium/16    via     pi8nos
route add Germany/16    via     pi8nos
route add Switzerland/16 via    pi8nos
route add USA/9         via     pi8nos
#
# explicit NET/ROM route to sysop for emergency cases
# (NET/ROM routes can be re-routed when an interface fails)
#
route add [44.137.40.1] netrom


----------------------------------------------------------------
NETROM.NET for PI8RNI
----------------------------------------------------------------

# NET/ROM routes and filters for PI8RNI
# only the nodes in the list below are accepted for automatic route table
# updates (NODES broadcasts) on the interlink.  this is controlled by the
# "mode accept".  on the LAP anybody can send broadcasts, except when an
# explicit filter entry is added to disallow it. ("mode reject")
#
netrom nodefilter mode accept
netrom nodefilter mode reject lap
netrom nodefilter mode reject lap.48
#
netrom nodefilter add pi1nos    ln.n48
#
netrom nodefilter add dd9qp-9   lap
netrom nodefilter add dh9kae    lap
netrom nodefilter add dl3eam-9  lap
netrom nodefilter add dl7oq-9   lap
netrom nodefilter add dl9kas    lap
netrom nodefilter add dl9kaw    lap
netrom nodefilter add pa3dfr-9  lap
netrom nodefilter add pa3epp-9  lap
netrom nodefilter add pe1bmm    lap
netrom nodefilter add pe1bmm-9  lap
netrom nodefilter add pi8awt    lap
netrom nodefilter add pi8duy-7  lap
netrom nodefilter add pi8gro-1  lap
netrom nodefilter add pi8gro-7  lap
netrom nodefilter add pi8mqp-7  lap
netrom nodefilter add pi8rni    lap
netrom nodefilter add pi8rni    lap.48
#
# explicitly addressed broadcasts on links (avoids reception by others)
#
netrom bcdest ln.n48 drop nodes
netrom bcdest ln.n48 add pi1nos
#
# uneven nodetimers reduce the need for parallel transmission and memory
#
netrom nodetimer 1790 ln.n48
netrom nodetimer 1740 lap.48
netrom nodetimer 3500 lap
#
# some excludes to prevent connect attempts when link has failed.
#
ax25 exclude pi1nos lap
ax25 exclude pi8ehv lap
ax25 exclude pi8esa lap
ax25 exclude pi8nos lap
ax25 exclude pi8vrz lap
ax25 exclude pi8yrc lap
#
# some nodes excluded because they are non-adjacent link-nodes
# or are unreliable (not QRV 24h/day)
#
#
# hard-wired downlink destinations
#
netrom downlink pa3edn-1 lap.48
netrom downlink pe1chl-1 lap.48
netrom downlink pi8utr lap.48


----------------------------------------------------------------
REMOTE.NET for PI8RNI
----------------------------------------------------------------

# commands to start remote-control servers
# put in a separate file so that we don't have to send it every time
# autoexec.net is updated (somebody could be monitoring us)
#
start rcmd 333 "dsfjkaewrqbefndscuwerfjasfvaaaaarughrueggdfv"
start remote 3534

[of course this is not the real file used at PI8RNI!!!]

----------------------------------------------------------------
ONEXIT.NET for PI8RNI
----------------------------------------------------------------

# this file will be executed on "exit" from NET
#

----------------------------------------------------------------
AT0000.NET for PI8RNI
----------------------------------------------------------------

# commands executed each day at 00:00
#

----------------------------------------------------------------
AT0600.NET for PI8RNI
----------------------------------------------------------------

# commands executed each day at 06:00
#

----------------------------------------------------------------
AT1200.NET for PI8RNI
----------------------------------------------------------------

# commands executed each day at 12:00
#

----------------------------------------------------------------
AT1800.NET for PI8RNI
----------------------------------------------------------------

# commands executed each day at 18:00
#

----------------------------------------------------------------
NR_INFO.TXT for PI8RNI
----------------------------------------------------------------

PI1RNI/PI8RNI - NET/ROM and TCP/IP node in Bilthoven (De Bilt)

Local Access Point is 430.725 MHz, link is on 23cm

Commands are:

CONNECT nodecall              Connect another node (call or alias)
CONNECT ax25call [VIA digis]  Connect an AX.25 user on the LAP
INFO                          Show this information
MHEARD                        Show MHEARD list of the LAP
MHEARD interface              Show MHEARD list of any interface
NODES                         Show local access NODES
NODES *                       Show all NODES
NODES * *                     Show complete NODES info
NODES nodecall                Show info about a single NODE (call or alias)
PORTS                         Show available interfaces
PORTS downlink_interface      Set the downlink interface
ROUTES                        Show routes to neighbor nodes
USERS                         Show connected users

Please don't connect PI1RNI, this call is for interlink use only.  Users should
connect only to PI8RNI.

Send messages for sysops to "pe1chl@pi8utr" or "pa3azk@pi8utr" (BBS) or
to sysop@pi8nos (SMTP).

Node cmd?

----------------------------------------------------------------





Rob PE1CHL

