Has anybody had success building a cluster with 3 or more servers using
the latest snapshot from the nightly RPM repository?  I've had this
working in the past, but the last couple of cluster builds I've done
have resulted in strange behavior.  All have been installed on a CentOS
6.2 64-bit VM, using the nightly snapshot repo, with the sipxaastra and
sipxopenacd packages added to the default sipXecs groupinstall.

 

The first server added to the cluster (ID 2) will go into a status of
"Configured" after running sipxecs-setup, no problem.  But when adding
the next server to the cluster (ID 3), all three servers will be in a
"Configured" state for a short while (10 - 15 seconds tops), then server
ID 2 goes into an "Uninitialized" state while server ID 3 remains
"Configured".

 

If I run sipxecs-setup on server ID 2, in an attempt to get it running
again, it results in the statuses flip-flopping.  ID 2 becomes
Configured, but at the expense of ID3 which goes then goes
Uninitialized.  This is reproducible at will after this point, swapping
back and forth between ID 2 and ID 3.

 

I had to delete server ID 3 so that our testers could get back to work,
but plan to reproduce the scenario again this evening to gather some
logs.  But, I was hoping that in the meantime perhaps somebody had a
similar experience, or had an idea to help diagnose the underlying
issue.

 

Please let me know if there are logs or other specific information which
could help troubleshoot.

 

Thanks!

 

Dave.

_______________________________________________
sipx-dev mailing list
[email protected]
List Archive: http://list.sipfoundry.org/archive/sipx-dev/

Reply via email to