[AG-TECH] Debugging network problems in new AG setup

[Andrew A Rowley]

Hi,

Could it be that multicast is not enabled on the network that you are using?  
Try using a bridge on an external server (it won't work if one of your client machines is the bridge, as the bridge and vic and rat will try to bind to the same unicast ports.  Note that because of the external server, you might then have firewall problems.

Andrew :)

============================================
Access Grid Support Centre,
RSS Group,
Manchester Computing,
Kilburn Building,
University of Manchester,
Oxford Road,
Manchester, 
M13 9PL, 
UK
Tel: +44(0)161-275 0685
Email: Andrew.Rowley at manchester.ac.uk 

> -----Original Message-----
> From: owner-ag-tech at mcs.anl.gov [mailto:owner-ag-tech at mcs.anl.gov] On
> Behalf Of R. P. Channing ["Rick"] Rodgers
> Sent: 20 December 2005 00:49
> To: ag-tech at mcs.anl.gov
> Cc: rodgers at nlm.nih.gov; aronson at nlm.nih.gov
> Subject: [AG-TECH] Debugging network problems in new AG setup
> 
> Dear AG Colleagues,
> 
> I am trying to establish a small AG setup at UCSF, on two Dell XPS Gen 4
> PCs
> running Windows XP.  We have installed AGTk 2.4, started the server on one
> of
> the machines, configured nodes on both, and started the AG client on both.
> We seem to be able to enter the venues we created, and each machine sees
> the
> other in the list of participants.  However, no chat, no video, and no
> audio
> seems to travel between the two machines (although vic and rat do come up
> on
> each machine, and appear to be operating properly).  Although both
> machines are on the same subnet, this smells like some kind of traffic
> filtering
> problem to me.  I append all of the log files (realizing that not all of
> them
> may be useful, and having pared them down a bit to the more recent
> entries).
> Someone versed in interpreting these may find that we have made some error
> in
> setting up the server/node/services though I think not.
> 
> We have disabled some Norton software that might have geen
> responsible, and ensured that the built-in firewall in XP is inactive,
> but that did not solve the problem.  Could there be a bridge or other
> device
> between the hosts that is causing the problem?  Would a reasonable
> starting
> point in debugging this be to connect the two machines together using a
> cross-over cable (eliminating the network as a variable) -- or would that
> be
> an inadequate test for the problem being in the network rather then in our
> AG
> setup on the PCs?  Another group on the
> campus is in charge of the network, so if we can establish that the
> problem is
> network-related, we can punt the problem to them -- though the better we
> can
> characterize what is wrong, the faster it is likely to get fixed.  Any
> suggestions about debugging strategies?  I'm a UNIX guy and feel
> completely
> out of water working with PCs ;)  I suppose we could use netcat and/or
> etherial
> as aids -- is there any kind of troubleshooting guide for this sort of
> problem?
> 
> Sorry to couch this question at a rather high level.  Many thanks in
> advance
> for any helpful pointers.
> 
> Best Regards, Rick Rodgers
> --------------------------------------------------------------------------
> ------
> R. P. C. Rodgers, M.D. * rodgers at nlm.nih.gov * (301)435-3267 (voice, fax)
> OHPCC, LHNCBC, U.S. National Library of Medicine, NIH
> Bldg 38, Rm. B1N-30F2, 8600 Rockville Pike, Bethesda MD 20894 USA
> http://lhc.nlm.nih.gov/staff/rodgers/rodgers.html
>