[prev in list] [next in list] [prev in thread] [next in thread] 

List:       mico-announce
Subject:    Strange communication problem (was: Re: [mico-devel] Assertion from
From:       Kretschel Klaus <Klaus.Kretschel () dlr ! de>
Date:       2004-10-07 8:13:20
Message-ID: 200410071013.20462.Klaus.Kretschel () dlr ! de
[Download RAW message or body]

Hi,

[problem was an "InvalidName" exception at start-up]

On Friday 13 August 2004 21:25, Karel Gardas wrote:
> IMHO (3) should run well, you just need to find the reason why ORB throws
> InvalidName exception, have you changed your setup in some way after
> upgrade?

Meanwhile it seems that the problem had nothing to do with Linux upgrade but 
probably with routing, since it occurs now also on another cluster where 
nothing but the routing entries have been changed. Unfortunately nobody saved 
the old entries, thus we cannot try the old status.

The topology in both cases is the same: a front-end node connected to a 
compute cluster (a local net) and to the general net on the other hand. The 
problem is that all clients throw the "InvalidName" exception while trying to 
connect to the name service if the MICO deamons run on the front-end, but 
everything runs fine when the MICO deamons run one of the nodes.

I heard some rumors about a missing "multicast" option, so I tried to add net 
244.0.0.0 to routing but that did not help (224.0.0.1is available from the 
kernels). On the other hand, I had a look at the port with tcpdump and saw 
that packets where transmitted from the node on eth0 and received at the 
front-end on eth0 on the one cluster and on eth3 on the other one, so I guess 
that communication is okay.

Any help?

Thank you, Klaus

PS:
Karel, maybe that we will need commercial support with this problem (though I 
cannot decide about our money). Would you please mail me terms and 
conditions?
_______________________________________________
Mico-devel mailing list
Mico-devel@mico.org
http://www.mico.org/mailman/listinfo/mico-devel
[prev in list] [next in list] [prev in thread] [next in thread]