Broadsoft Cluster Nodes "Out of Sync"10 May 2011
I am in the process of fixing two cluster installations which were originally performed by a third party consulting group. The
peerctl lists were not setup correctly and thus replication is not working from the primary node to the secondary node on either platform.
There are a few tasks you can perform to validate whether
replication to your secondary nodes are healthy. First, ensure
dbSyncCheck is running in the maintenance scheduler for all AS and NS nodes. Second, verify whether you are receiving alarms from your cluster nodes stating the datbase is out of sync. Third, see if you can perform an
importdb.pl on the secondary node without it failing. In the event your platform is not in sync or importdb.pl fails, you must reconfigure the
peerctl list on the primary server. This example assumes your are setting the peerctl list for an application server cluster.
** On the primary AS
peerctl set <old peerctl name> <hostname> <ipaddress> | example: peerctl set as1.markholloway.com as1 192.168.1.10 peerctl set <old peerctl name> <hostname> <ipaddress> | example: peerctl set as2.markholloway.com as2 192.168.1.11 peerctl setprimary as1 repctl restart bwBackup.pl AppServer db-JAN-11-2009.db scp db-JAN-11-2009.db as bwadmin to as2
** On the secondary AS
repctl stop stopbw bwRestore.pl AppServer db-JAN-11-2009.db importdb.pl AppServer 192.168.1.10 | Verifies importdb.pl works as expected
You will now have proper
replication between your primary and secondary nodes.