I’ve had similar trouble, but with a smaller network. I use a similar
method,
with the exception being that I use netmap’s ‘-F’ option as in:
on -n $tgtNode -r //$tgtNode/ netmap -F //$SERVER/etc/config/netmap
which allows the net nodes to read the file from the server, rather than
copy
the file locally. I also do somthing similar in the sysinit files.
This
allows for having only one netmap file, on the SERVER.
Also, the alive line in the script has the -fu option, which “forces”
attempts
to find nodes and returns a list of nodes that are alive. for faster
operation,
forgo to -f option. You’ll get less nodes in the list, but alive
returns much
quicker.
Thanks for the script, Richard.
“Brown, Richard” wrote:
Assuming your QNX nodes communicate using FLEET the following will work for
adding nodes. Designate one node as the QNX LAN server. Add the new MAC
address to the server and ensure the server can then see the new node. Then
run the script below and all nodes on the QNX LAN should be able to now see
the node.
#! /bin/sh
SERVER=1
tgtNodes=$(alive -fu)
for tgtNode in $tgtNodes
do
echo “Broadcasting netmap to node $tgtNode …”
cp //$SERVER/etc/config/netmap //$tgtNode/etc/config/netmap
on -n $tgtNode -r //$tgtNode/ netmap -f
done
“ian zagorskih” <> ianzag@novosoft-us.com> > wrote in message
news:9svfpl$i7c$> 1@inn.qnx.com> …
“Doug Rixmann” <> rixmannd@rdsdata.com> > wrote in message
news:9supki$5t2$> 1@inn.qnx.com> …
I want a sure fire method of adding a node to a network. Currently, we
edit the netmap file of one node, perform netmap -f and make sure
that the node can see all other nodes. Then copy the netmap file across
to the other nodes. Finally, we telnet to the other nodes and run
netmap -f
to refresh the netmap on the other nodes.
This is a tedious task and requires too much
user (helpdesk) intervention. We are hoping to build a utility that
we can run from the new node (or an exiting node) and have the
remainder of the network “automagically” reconfigure.
We also want to do the same thing for removing nodes.
It seems like a straight forward task but I thought I would ask the
experts in this pool. Thanks in advance.
when you start Net server it allows remote unknown nodes to be recognized
upon request. i.e. you don’t need to have a proper node number<->MAC
address
record inside your netmap file, remote node mapping will be temporarily
added at runtime but lost after Net’s termination. if you don’t need this
feature and want to force Net to talk only with listed nodes you should
start it with -A option. with this feature you can connect with Fleet from
your new node to remote node only knowing remote node’s number and MAC
address. just add remote node’s parameters into your node’s netmap file
and
“netmap -f” it. after that you can access remote node as usually so you
can
fix its netmap file etc.
also afair there was some utility that allows you to do very simular task.
afair it was written by Mario Charest but i’v never used it so no ideas.
ask
him > 
Doug
// wbr
\
–
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
/ \ | __ ) | ~Jeffrey Jordan jordanj@abc-naco.com
/ | \ | _ ( Phone: (610)630-2330x216 jljordan@wans.net
/ ++ \ | ) ( Fax: (610)630-2323
|| |||/_| 2550 Blvd. o/t Generals, Norristown PA 19403
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-