ORBIT-USER: Questions about ORBIT grid experiments

[Max Ott]

It's not necessary to switch the nodes on separately. The nodehandler
takes care of that.

max at console:~$ nodehandler -k ~/scripts/hello5
Using config /etc/nodehandler/grid.cfg
/etc/nodehandler/grid.cfg:20: warning: Insecure world writable dir
/tmp, mode 040777
Using logfile /etc/nodehandler/nodehandler_log.xml
 INFO init: NodeHandler Version 3.6.4-1 (849)
 INFO init: Experiment ID: grid_2006_11_06_23_03_26
 INFO Experiment: load system:exp:stdlib
 INFO prop.resetDelay: resetDelay = 180:Fixnum
 INFO Experiment: load /home/max/scripts/hello5
/tmp/eee.153/lib/util/communication.rb:127: warning: Insecure world
writable dir /tmp, mode 040777
 INFO stdlib: 2 out of 2 node(s) still down n_5_5,n_5_6
 ...
 INFO n_5_5: Checked in as /ip/10.10.5.5 booting off baseline:1.0.9
 INFO stdlib: 1 out of 2 node(s) still down n_5_6
 INFO n_5_6: Checked in as /ip/10.10.5.6 booting off baseline:1.0.9
 INFO n_5_5: Device 'net/w0' reported 00:00:00:00:00:00
 INFO whenAll: _ALL_: 'apps/app/status[text()='INSTALLED.OK']' fires
 INFO n_5_6: Device 'net/w0' reported 00:60:B3:AC:2C:D2
 INFO n_5_5: Device 'net/w0' reported 00:60:B3:AC:2C:D2
 INFO OML: Started: {"port"=>"7000", "iface"=>"eth1", "addr"=>"224.0.0.6"}
 INFO Experiment: DONE!
 INFO run: Experiment grid_2006_11_06_23_03_26 finished after 3:1

2. Noise
------------

The noise generator seems to be down.

3. IP addresses of wireless interfaces
------------------------------------------------------

ALl teh interfaces available to experimenters, e0, w0, and w1 are by
default off and unassigned. Use the hodehandler's ...net.w0.ip command
to assign an IP address (this command also automatically loads the
right driver and brings up the interface).

See http://www.orbit-lab.org/wiki/Documentation/NodeHandler/Commands/nodes

Or inthe 'hello world' script:

allNodes.net.w0 { |w|
  w.type = 'b'
  w.essid = "helloworld"
  w.ip = "%192.168.%x.%y"
}

which not only sets the 802.11 related paramters (type, essid), but
also the IP address. I should add that '%x' will be replaced by the X
coordinate of the node (also note the leading '%' which identifies the
string as a template string).

-max





On 11/7/06, Yang  Chen <yangchen at cc.gatech.edu> wrote:
> ========================================
> 1. About the helloworld experiment
> ========================================
>
> I can finish the helloworld experiment with the command "nodehandler -k
> ~/grid_script/hello-world"
> When I try to change it a little bit to have the data exchange between
> node5-5/5-6 instead of 1-1 and 1-2
>
> a) with the following modifications to the example code at
> http://www.orbit-lab.org/wiki/Tutorial/HelloWorld
>
> "defNodes('sender', [5,6]) {|node|" => "defNodes('sender', [1,1]) {|node|"
> "    'destinationHost' => '192.168.5.5'," => "'destinationHost' =>
> '192.168.5.5',"
> "defNodes('receiver', [5,5]) {|node|" => "defNodes('receiver', [1,2])
> {|node|"
>
> b) power on the node5-5/5-6
> yangchen at console.grid:~$ wget -O - -q 'http://cmc:5012/cmc/on?x=5&y=5'
> yangchen at console.grid:~$ wget -O - -q 'http://cmc:5012/cmc/on?x=5&y=6'
> (I can verify that node 5-5/5-6 are ON by ssh as root)
>
> c) nodehandler -k ~/grid_script/hello-world
> the following messages are reported forever
>
>  INFO stdlib: 2 out of 2 node(s) still down n_5_6,n_5_5
>  INFO stdlib: 2 out of 2 node(s) still down n_5_6,n_5_5
>
> Sometimes, I can see the effort to reset both nodes
>
>  WARN stdlib: Resetting node n_5_6
>  INFO n_5_6: Resseting node
>  WARN stdlib: Resetting node n_5_5
>  INFO n_5_5: Resseting node
>
> But generally, the node_5_6/5_5 are always down.
>
> ========================================
> 2. About the noise-generation experiment
> ========================================
>
> I copied the script from http://www.orbit-lab.org/wiki/HowTo/UseNoise
> and name it tutorial-noise.rb
>
> Without any changes and with the command "nodehandler
> ~/grid_script/tutorial-noise"
> node1-1 and 1-2 keeps down and eventually the experiment fails with
> following messages
>
>  INFO stdlib: 2 out of 2 node(s) still down n_1_2,n_1_1
>  INFO stdlib: 2 out of 2 node(s) still down n_1_2,n_1_1
> FATAL run: Exception: Connection timed out - connect(2) (Errno::ETIMEDOUT)
>
> I tried the following,
> a) refresh the image on node1-1/1-2 by
> imageNodes [1,1] baseline.ndz
> imageNodes [1,2] baseline.ndz
>
> b) power on those nodes and verifications by ssh as root
>
> but same error happens
>
> I guess there is some issue with the antenna configuration/operation part,
> this experiment finished successfully after I comment out the following code
> block
> #antenna(4,4).signal {|s|
> #   s.bandwidth = 20
> #   s.channel = 36
> #   s.power = prop.noisePower
> #}
> and all the power adjustment part in the whenAllInstalled(){}.
>
> ================================================================================
> 3. What is the IP address for ath0 and ath1 on each radio node in the grid?
> ================================================================================
>
> Any suggestions/answers would be appreciated.
>
> thx a lot.
>
> -Yang
>
>


-- 
Dr. Max Ott
Research Program Leader - Network and Pervasive Computing, NICTA Australia
Founder & CTO, Semandex Networks
Research Professor, WINLAB, Rutgers University