[orbit-user] [ath 0] woes

Wed Feb 13 15:26:28 EST 2008

Although I am not sure what the solution to the problem is, the problem is
with the query to the inventory database and not any specific node.

A look at the logs and a few trial commands to the cmc fail to give
interface information for any selected node.

In short ,

wget -v http://cmc:5022/inventory/getMacAddress?x=4
<http://cmc:5022/inventory/getMacAddress?x=4&ifname=ath0&domain=grid&y=1>
&ifname=ath0&domain=grid&y=1

always returns saying that no info is available.

Hope someone else can pick it up from here.

Sanjit

From: orbit-user-bounces at orbit-lab.org
[mailto:orbit-user-bounces at orbit-lab.org] On Behalf Of Shrinivas Mohan
Sent: Wednesday, February 13, 2008 2:37 PM
To: orbit-user at orbit-lab.org
Subject: [orbit-user] [ath 0] woes

Hi Kishore,

Thanks for the reply. I have been getting this problem for different nodes
each time. The tutorials I was referring to, pick nodes at random from the
nodes that are active. The problem does not seem to be node-specific.As I
said, it was of the form..different runs brought up the same problem for
different nodes.

Thanks
Shrini

Message: 2
Date: Wed, 13 Feb 2008 10:19:17 -0500
From: "Kishore Ramachandran" <kishore at winlab.rutgers.edu>
Subject: Re: [orbit-user] [ath 0] woes
To: "Orbit user discussion mailing list" <orbit-user at orbit-lab.org>
Message-ID:
       <604f6e750802130719n1905f7a3t6587cf995ab09a8 at mail.gmail.com>
Content-Type: text/plain; charset="iso-8859-1"

Hi Shrini:

You need not use "orbit tell on all" before imaging - in fact, you would
rather not do that since imageNodes has to perform a bootup cycle on all
nodes before the imaging process can begin.

As for the other issue regarding the "no information on [ath0]", can you
check whether that particular node has an Atheros interface? Have you tried
the tutorial using a node other than [19,12]? That would be the first
debugging step I would pursue. I am guessing the software (in this case
nodehandler) does not know how to deal with the first issue - that of
missing information in the inventory for a particular interface - and hence
fails to exit gracefully. Can you please try using a different node in your
script and report back on whether the problem goes away and is
node-specific?

I request the Nodehandler v4 maintainers to look at this problem and fix the
related bug.

regards,
Kishore

On Feb 13, 2008 2:08 AM, Shrinivas Mohan <shrini at ucla.edu> wrote:

> Hi Everyone,
>
> I am repeating a post from last week. Can someone please help.
> Once I login to the testbed, I do
> 1. imageNodes
> 2. orbit exec tut_topo_2
>
> Should I use orbit tell on all ??? Before/after imageNodes ?? The
> documentation didn't clarify on this.
>
> I tried running tut_topo_2.rb (setting up a multi-hop topology)
>
>
>  INFO init: NodeHandler Version 4.2.0 (1272)
>  INFO init: Experiment ID: grid_2008_02_13_01_57_10
>  INFO Experiment: load system:exp:stdlib
>  INFO prop.resetDelay: resetDelay = 210:Fixnum
>  INFO prop.resetTries: resetTries = 1:Fixnum
>  INFO Experiment: load tut_topo_2
>  INFO Topology: Loading topology 'system:topo:active:grid'.
>  WARN system:topo:active:grid: Ignoring missing node '12 at 14'
> Number of Active nodes on this tesbed: 9
>  INFO stdlib: Waiting for nodes (Up/Down/Total): 0/4/4 - (still down:
> n_19_15,n_19_12,n_20_5)
>  INFO stdlib: Waiting for nodes (Up/Down/Total): 0/6/6 - (still down:
> n_19_15,n_19_12,n_20_5)
> ERROR mainTopology: Topology - Inventory has no information on [ath0] for
> node [19,12] (domain: grid)
> FATAL run: Exception: Topology - Inventory has no information on [ath0]
> for node [19,12] (domain: grid) (RuntimeError)
>         /opt/nodehandler4-4.2.0/lib/handler/topology.rb:231:in
> `buildMACBlackList'
>         /opt/nodehandler4-4.2.0/lib/rexml/element.rb:891:in `each'
>         /opt/nodehandler4-4.2.0/lib/rexml/xpath.rb:53:in `each'
>         /opt/nodehandler4-4.2.0/lib/rexml/element.rb:891:in `each'
>         /opt/nodehandler4-4.2.0/lib/handler/topology.rb:229:in
> `buildMACBlackList'
>         /opt/nodehandler4-4.2.0/lib/handler/topology.rb:215:in `each'
>         /opt/nodehandler4-4.2.0/lib/handler/topology.rb:215:in
> `buildMACBlackList'
>         /opt/nodehandler4-4.2.0/lib/handler/nodeSet.rb:338:in
> `setMACFilteringTable'
>         /opt/nodehandler4-4.2.0/lib/handler/nodeSet.rb:894:in `initialize'
>         /opt/nodehandler4-4.2.0/lib/handler/nodeSet.rb:950:in `new'
>         /opt/nodehandler4-4.2.0/lib/handler/nodeSet.rb:950:in
> `method_missing'
>         ./tut_topo_2.rb:187
>         (eval):20:in `require'
>         (eval):20:in `getProc'
>         /opt/nodehandler4-4.2.0/lib/handler/oconfig.rb:137:in `call'
>         /opt/nodehandler4-4.2.0/lib/handler/oconfig.rb:137:in `load'
>         /opt/nodehandler4-4.2.0/lib/handler/experiment.rb:33

:in `load'
>         /opt/nodehandler4-4.2.0/lib/handler/nodeHandler.rb:186:in `run'
>         /opt/nodehandler4-4.2.0/lib/handler/nodeHandler.rb:527
>         /opt/nodehandler4-4.2.0/app/handler.rb:3:in `require'
>         /opt/nodehandler4-4.2.0/app/handler.rb:3
>  INFO ExecApp: Application 'commServer' finished
>  INFO run: Experiment grid_2008_02_13_01_57_10 finished after 0:36
>
>
> Thanks a lot,
> Shrini
>
> --
> Shrinivas Mohan
> Graduate Student - Computer Science
> University of California at Los Angeles
> _______________________________________________
> orbit-user mailing list
> orbit-user at orbit-lab.org
> http://orbit-lab.org/cgi-bin/mailman/listinfo/orbit-user
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://orbit-lab.org/pipermail/orbit-user/attachments/20080213/76b9d928/atta
chment.html

-- 
Shrinivas Mohan
Graduate Student - Computer Science
University of California at Los Angeles 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://orbit-lab.org/pipermail/orbit-user/attachments/20080213/db27748e/attachment-0001.htm 