Arpita,
In my (admittedly limited) experience, the ttl is your issue here. The
ttl (or Time-To-Live) is the number of hops across switches your
packet will survive. In your case, if it is more than one switch in
the chain the packet will get dropped. I tend to raise it to 15-20 if
it is going across my network alone, and to 35 if it is going across
the internet.
--
Adam Tygart
Beocat Sysadmin
On Mon, May 9, 2011 at 22:46, arpita k <arpita...@gmail.com> wrote:
> Hi,
>
>
>
> I am using ganglia framework in our application. The gmetad and gmond both
> are installed in one machine named A and the gmond is installed another
> machine named B. I need to publish all the metric data from machine B to
> machine A.But both the machines are present in different subnet. The Ip of
> machine B is like: 34. 456.123.21 and machine A’s ip is like: 295.45.23.21.
> So, I am trying to make unicast communication.
>
>
>
> Here is the gmond configuration present in the machine B:
>
>
>
> cluster {
>
> name = "test_ganglia"
>
> owner = "unspecified"
>
> latlong = "unspecified"
>
> url = "unspecified"
>
> }
>
>
>
> udp_send_channel {
>
> host = 295.45.23.21
>
> port = 8649
>
> ttl = 1
>
> }
>
>
>
> Gmond of machine A:
>
>
>
> udp_recv_channel {
>
> mcast_join = 239.2.11.71
>
> port = 8649
>
> bind = 239.2.11.71
>
> }
>
> udp_recv_channel {
>
> port = 8649
>
> }
>
>
>
> /* You can specify as many tcp_accept_channels as you like to share
>
> an xml description of the state of the cluster */
>
> tcp_accept_channel {
>
> port = 8649
>
> }
>
>
>
> But the data is not getting published in the machine A.
>
>
>
> Please help me.
>
>
>
> Thanks in Advance!!
>
>
>
> Regards,
>
> Arpita
>
> ------------------------------------------------------------------------------
> Achieve unprecedented app performance and reliability
> What every C/C++ and Fortran developer should know.
> Learn how Intel has extended the reach of its next-generation tools
> to help boost performance applications - inlcuding clusters.
> http://p.sf.net/sfu/intel-dev2devmay
> _______________________________________________
> Ganglia-general mailing list
> Ganglia-general@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/ganglia-general
>
>
------------------------------------------------------------------------------
Achieve unprecedented app performance and reliability
What every C/C++ and Fortran developer should know.
Learn how Intel has extended the reach of its next-generation tools
to help boost performance applications - inlcuding clusters.
http://p.sf.net/sfu/intel-dev2devmay
_______________________________________________
Ganglia-general mailing list
Ganglia-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ganglia-general
