[Spread-users] need help with spread configuration on VMs cluster

John Schultz jschultz at spreadconcepts.com
Wed Apr 24 09:55:14 EDT 2013

As a first guess, it is possible that your spread segment address is wrong.

Is the broadcast address for your virtual network really  Or should it be

You could also try a multicast address rather than a broadcast address (e.g. -

I would also check that all your virtual machines can talk to each other directly using ping or the spsend and sprecv programs using the addresses in your configuration.

You can definitely have IPC without using a loopback/ entry.  So long as a singleton daemon starts up and the local clients can connect to it, then IPC should work.  If you have a multiple daemon configuration then IPC will still work, however, it might seem to hang if the daemons are struggling to establish a membership and can't finish, which can happen when your network + spread configuration aren't proper.


John Lane Schultz
Spread Concepts LLC
Phn: 301 830 8100
Cell: 443 838 2200

On Apr 23, 2013, at 10:19 AM, Tarashchanskiy, Yevgeniy wrote:

I am a newbie. I am trying to do simplest things to evaluate if I can use spread in my project.
I have a segment of several virtual machines that should talk to each other plus I would like to use spread for IPC between servers and clients on every VM.
I created very simple configuration file:
At best two VMs can finish configuration and "see" each other:
Num Segments 1
        2   4803
In most of the attempts only one VM finish configuration and other VMs print the line:
"Conf_load_conf_file: My name: . . ., port 4803"
and stop.
I can start spuser(s) on VMs but they can't send messages to each other.
( My initial testing on one VM with configuration file that include line:
"loopback     "
allowed me to create a common group to several spusers and send messages to a group)
I need help to understand what can be wrong with configuration and how I can figure out my problems?
Can I have IPC without loopback?
Spread-users mailing list
Spread-users at lists.spread.org

More information about the Spread-users mailing list